Reforming Education and Challenging Inequalities in Southern Contexts: Research and policy in international development

Table of contents :
Cover
Half Title
Series
Title
Copyright
Dedication
Contents
List of figures and tables
Notes on contributors
Acknowledgements
List of abbreviations
Foreword: a tribute to Christopher Colclough
1 Education and the reform of social inequalities in the Global South: an introduction
Part I The economics and politics of educational reform
2 The changing pattern of returns to education: what impact will this have on earnings inequality?
3 Unequal access to education: accounting for change and counting costs
4 Education for All in India and Sri Lanka: the drivers and interests shaping egalitarian reforms
5 Public–private partnerships in education: do they offer an equitable solution to education in India and Pakistan?
6 The influence of politics on girls’ education in Ethiopia
Part II Challenges and opportunities in addressing inequalities through education
7 Overriding social inequality? Educational aspirations versus the material realities of rural families in Pakistan
8 Confronting social inequality through fertility change in Punjab, Pakistan: the role of girls’ schooling
9 Teenage pregnancy and social inequality: an impediment to achieving schooling for all in Uganda
10 Complementary basic education: parental and learner experiences and choices in Ghana’s northern regions
11 Addressing dilemmas of difference: teachers’ strategies to include children with disabilities in rural primary schools in India
12 Social distance, teachers’ beliefs and teaching practices in a context of social disadvantage: evidence from India and Pakistan
Index

Citation preview

“This volume is a both a provocative tribute to Professor Chris Colclough, and a welcome contribution to the study of education and social inequality. With chapters by some of the world’s leading social scientists, the book draws on Christopher Colclough’s intellectual legacy and a wide range of theories and cutting-edge empirical work, to highlight the cross-sectoral roots of disadvantage and marginalization in education. The book not only advances our understanding of social inequality in education – but considers policies most likely to lead to a more equitable future.” – Karen Mundy, Professor of International and Comparative Education, Ontario Institute for Studies in Education and Munk School of Global Affairs and Public Policy, University of Toronto, Canada “This beautifully written and well edited book is a fitting acknowledgement of Chris Colclough’s immense contribution to the global agenda and action on education for all. It brings together masterful analytical and practical insights into how social and economic inequalities impact on outcomes of schooling, and how education reforms should respond to achieve more equitable outcomes. An important read for anyone who seeks to understand and address educational inequalities in global south contexts.” – Kwame Akyeampong, Professor of International Education and Development, The Open University, UK “This is an exemplary collection of scholarly papers exploring the issue of inequality in schooling in the global south. Focusing especially on experiences from South Asia and Sub-Saharan Africa, they bring new insights into our understanding of the various dimensions of the problem and warn how persistence of unequal access to quality education could negate the benefits accruing from quantitative expansion of schooling in the post-Jomtien decades. The book is a befitting tribute to Chris Colclough who devoted his entire professional life to the study of this subject.” – Rangachar Govinda, Emeritus Professor and former Vice Chancellor, National University of Educational Planning, New Delhi, India “The right to education is the foundation stone for more equitable and inclusive societies. Nobody championed that right more powerfully or more effectively than Christopher Colclough. This book is a fitting tribute to his life and work, and a reminder that ‘education for all’ is an unfinished business.” – Kevin Watkins, CEO, Save the Children UK

Reforming Education and Challenging Inequalities in Southern Contexts

This book offers in-depth analyses of how education interacts with social inequality in Southern contexts. Drawing on a range of disciplinary frameworks, it presents new analyses of existing knowledge and new empirical data which define the challenges and possibilities of successful educational reform. It is a tribute to the work of the late Christopher Colclough, who, as a leading figure in education and international development, played a key role in the global fight for education for all children. The book critically engages with international evidence of educational access, retention and outcomes, offering new understandings of how social inequalities currently facilitate, mediate or restrict educational opportunities. It exposes the continuing influence of wealth and regional inequalities and caste and gendered social structures. Researchers in Ethiopia, Ghana, India, Pakistan and Uganda highlight how the aspirations of families living in poverty remain unfulfilled by poor-quality education and low economic opportunities and how schools and teachers currently address issues of gender, disability and diversity. The book highlights a range of new priorities for research and identifies some necessary strategies for education reform, policy approaches and school practice, if educational equality for all children is to be achieved. The book will be of great interest to researchers, scholars, educational practitioners and policy-makers in the fields of economics, politics and sociology of education, international education, poverty research and international development. Pauline Rose is Professor of International Education at the University of Cambridge, UK. Madeleine Arnot is Emeritus Professor of Sociology of Education at the University of Cambridge, UK. Roger Jeffery is Professorial Fellow in Sociology at the University of Edinburgh, UK. Nidhi Singal is Professor of Disability and Inclusive Education at the University of Cambridge, UK.

Education, Poverty and International Development Series Series Editor Madeleine Arnot

Faculty of Education, University of Cambridge, UK

This series of research-based monographs and edited collections, which was set up in collaboration with the late Professor Christopher Colclough, contributes to global debates about how to achieve education for all. A major set of questions faced by national governments and education providers concerns how the contributions made by education to reducing global poverty, encouraging greater social stability and equity, and ensuring the development of individual capability and wellbeing can be strengthened. Focusing on the contributions that research can make to these global agendas, this series aims to provide new knowledge and new perspectives on the relationships between education, poverty and international development. It offers alternative theoretical and methodological frameworks for the study of developing-country education systems, schools, colleges and universities in the context of national cultures and ambitious global agendas. It aims to identify the key policy challenges associated with addressing social inequalities, uneven social and economic development, and the opportunities to promote democratic and effective educational change in the name of social justice. The series brings together researchers from the fields of anthropology, economics, development studies, education research, politics, international relations and sociology. It includes work by some of the most distinguished writers in the fields of education and development, along with new authors working on important empirical projects. The series contributes significant insights on the linkages between education, the economy, communities, and society, based on interdisciplinary, international and national studies. Selected volumes include critical syntheses of existing research and policy, work using innovative research methodologies, and in-depth evaluations of major policy developments. Some studies address topics relevant to poverty alleviation, national and international policy-making and aid, while others are anthropological or sociological investigations of how education functions within local rural and urban communities, for households living in poverty or for particular socially marginalised groups. The authors explore a diverse range of themes from the challenges associated with providing quality teacher, professional and entrepreneurial education, to those associated with promoting gender equality, reducing gender violence, understanding the impact of poverty on constructions of childhood, or assessing the effectiveness of learner- centred school pedagogies. They offer sharp, critical studies that are intended to have a strategic influence on the thinking of academics, researchers and policy-makers.

Published Titles   1 Education Outcomes and Poverty A Reassessment Edited by Christopher Colclough   2 Teacher Education and the Challenge of Development A global analysis Edited by Bob Moon   3 Education Quality and Social Justice in the Global South Challenges for policy, practice and research Edited by Leon Tikly and Angeline Barrett   4 Learner-centred Education in International Perspective Whose pedagogy for whose development? Michele Schweisfurth   5 Professional Education, Capabilities and the Public Good The role of universities in promoting human development Melanie Walker and Monica McLean   6 Livelihoods and Learning Education for All and the marginalisation of mobile pastoralists Caroline Dyer   7 Gender Violence in Poverty Contexts The educational challenge Edited by Jenny Parkes   8 The ‘Poor Child’ The cultural politics of education, development and childhood Edited by Lucy Hopkins and Arathi Sriprakash   9 Educating Entrepreneurial Citizens Neoliberalism and youth livelihoods in Tanzania Joan DeJaeghere 10 Education, Poverty and Global Goals for Gender Equality How people make policy happen Elaine Unterhalter and Amy North For more information on the series, please visit: www.routledge.com/EducationPoverty-and-International-Development/book-series/EPID

Reforming Education and Challenging Inequalities in Southern Contexts Research and policy in international development Edited by Pauline Rose, Madeleine Arnot, Roger Jeffery and Nidhi Singal

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 selection and editorial matter, Pauline Rose, Madeleine Arnot, Roger Jeffery and Nidhi Singal; individual chapters, the contributors The right of Pauline Rose, Madeleine Arnot, Roger Jeffery and Nidhi Singal to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-367-26489-5 (hbk) ISBN: 978-0-367-74093-1 (pbk) ISBN: 978-0-429-29346-7 (ebk) Typeset in Galliard by Apex CoVantage, LLC

This book is dedicated to the late Christopher Colclough, our friend and mentor, and to his wife Sarah and his son Giles.

Contents

List of figures and tablesxiii Notes on contributorsxv Acknowledgementsxviii List of abbreviationsxix

Foreword: a tribute to Christopher Colclough

xxi

SIR RICHARD JOLLY

  1 Education and the reform of social inequalities in the Global South: an introduction

1

PAULINE ROSE, MADELEINE ARNOT, ROGER JEFFERY AND NIDHI SINGAL

PART I

The economics and politics of educational reform17   2 The changing pattern of returns to education: what impact will this have on earnings inequality?

19

HARRY ANTHONY PATRINOS

  3 Unequal access to education: accounting for change and counting costs

37

KEITH M. LEWIN

  4 Education for All in India and Sri Lanka: the drivers and interests shaping egalitarian reforms

59

ANGELA W. LITTLE

  5 Public–private partnerships in education: do they offer an equitable solution to education in India and Pakistan? MONAZZA ASLAM AND GEETA GANDHI KINGDON

79

xii  Contents   6 The influence of politics on girls’ education in Ethiopia

98

LOUISE YORKE, PAULINE ROSE AND ALULA PANKHURST

PART II

Challenges and opportunities in addressing inequalities through education121   7 Overriding social inequality? Educational aspirations versus the material realities of rural families in Pakistan

123

ARIF NAVEED

  8 Confronting social inequality through fertility change in Punjab, Pakistan: the role of girls’ schooling

144

FEYZA BHATTI AND ROGER JEFFERY

  9 Teenage pregnancy and social inequality: an impediment to achieving schooling for all in Uganda

165

FLORENCE KYOHEIRWE MUHANGUZI AND GRACE BANTEBYA KYOMUHENDO

10 Complementary basic education: parental and learner experiences and choices in Ghana’s northern regions

185

LESLIE CASELY-HAYFORD WITH ADOM BAISIE GHARTEY AND JUSTICE AGYEI-QUARTEY

11 Addressing dilemmas of difference: teachers’ strategies to include children with disabilities in rural primary schools in India

206

NIDHI SINGAL

12 Social distance, teachers’ beliefs and teaching practices in a context of social disadvantage: evidence from India and Pakistan

224

ANURADHA DE AND RABEA MALIK

Index244

Figures and tables

Figures 2.1 2.2 2.3 2.4 2.5 2.6 3.1 3.2 6.1 6.2 6.3 10.1 12.1 2.2 1 12.3

Years of schooling and returns over time 25 Returns to schooling and average years of schooling by period 26 Standard deviation of rate of return to year of schooling 27 Changes in returns to education since 1980 28 Changes in returns to education since 1980, men 29 Changes in returns to education since 1980, women 29 Types of enrolment by grade in LICs and LMICs 40 LICs and LMICs classified by percentage of girls enrolled by grade 43 Gross enrolment rate at primary level (grades 1–8) by gender, 1995/96–2016/17100 Gross enrolment rate at upper primary level (grades 5–8) by gender and region, 2016/17 101 Categorisation of stakeholders’ attitudes towards girls’ education 114 Incidences of OOSC and access to primary schools 188 Stylised framework linking teacher practice and effort with identity, attitudes and beliefs, alongside other factors 228 Two channels linking teachers’ beliefs with student learning 230 Percentage of teachers who agree or strongly agree that specific background factors affect children’s ability to learn 234

Tables 2.1 2.2 3.1 3.2 5.1 5.2 6.1 6.2 6.3 7.1 8.1 8.2

Returns to schooling over time (%) Education–earnings relationship (% of countries) LICs and LMICs classified by enrolment types LICs and LMICs classified by patterns of participation by gender Typologies of state and non-state providers  Typologies of PPP arrangements in Pakistan Government stakeholders included in the key informant interviews Strategies for girls’ education outlined in the Education Sector Development Plans (I–V) Categorisation of stakeholders’ attitudes towards girls’ education The 2016 interview sample by the highest educational level achieved Sample characteristics for Punjabi women aged 25–34, 1990/91, 2006/07 and 2012/13 Sample characteristics for qualitative studies, by education levels and location

26 30 41 44 82 89 103 106 110 128 148 149

xiv  Figures and tables 8.3 Percentage distributions of desired family size of Punjabi women aged 25–34 by schooling grade completed, 1990/91, 2006/07 and 2012/13 8.4 Percentage of reported current use of contraception by married, non-pregnant Punjabi women aged 25–24 by schooling, 1990/91, 2006/07 and 2012/13 9.1 The National Legal and Policy Framework 9.2 Trends in teenage pregnancy rates (%) in Uganda between 1995 and 2016

150

153 168 170

Contributors

Justice Agyei-Quartey is a monitoring and evaluation specialist with over 18 years of experience working with development partners including: USAID, DFID, Social Impact, Associates for Change and a number of civil society agencies in Ghana. Justice has provided technical assistance to the Complementary Basic Education Programme under the Ministry of Education, Ghana Education Service and Non-Formal Education Division. Madeleine Arnot, Emeritus Professor of Sociology of Education, University of Cambridge, has published extensively on educational issues relating to social inequality, gender relations, youth and citizenship as part of the RECOUP project in Africa and South Asia. She edits the Routledge series Education, Poverty and International Development and has contributed to the recent UNICEF/CUP project, The Learning Passport, supporting the education of displaced youth globally. Monazza Aslam is the Managing Partner of an education in international development research think tank (Oxford Partnership for Education Research and Analysis). She is also an Honorary Senior Research Fellow at the Institute of Education, University College London. Monazza is an education economist with more than 15  years of experience in girls’ education, teacher quality, political economy of education systems and student learning. She also advises governments and donor agencies. Feyza Bhatti is a lecturer in Economics and Vice Dean of the Business Faculty at the Girne American University, north Cyprus. Between 2002 and 2011, she was Senior Research Fellow at Mahbub ul Haq Human Development Centre, contributing to annual Human Development in South Asia reports and conducting research for the Research Consortium on Outcomes of Education and Poverty (RECOUP). Leslie Casely-Hayford is the Director of Associates for Change, which conducts research and evaluation work with a variety of government, civil society and development partners. She has conducted several longitudinal and impact studies in the education and social sectors. Her research focus is in the areas of education equity, complementary basic education, social development and gender equality. Anuradha De is a Director of Collaborative Research and Dissemination (CORD), a not-for-profit research organisation based in New Delhi, India. She is an economist by training and has been involved in surveys and research on issues related to education, labour and governance. Her particular focus has been on education statistics, finance and policy.

xvi  Contributors Adom Baisie Ghartey holds a PhD degree in Development Studies. He is a consultant with extensive experience in development planning and management, socio-economic research, organisational development, governance and results-based management. He has worked for various governments, development partners and NGOs in various countries, especially in sub-Saharan Africa. Roger Jeffery is Professorial Fellow in Sociology, University of Edinburgh. He has carried out sociological research in Pakistan and India, where his work has focused on health policy as well as secondary education. Currently he is a coinvestigator for research on nitrogen-based pollution in South Asia; he is also researching how India has influenced the city of Edinburgh. Sir Richard Jolly is an honorary Professor in the Institute of Development Studies at the University of Sussex, focusing on world development and the role of the United Nations in global governance. He was Assistant Secretary General of the United Nations, Deputy Executive Director of UNICEF and Coordinator of the UNDP’s Human Development Report. Among his many publications are five volumes on UN history and the influential Adjustment with a Human Face: Protecting the Vulnerable and Promoting Growth. Geeta Gandhi Kingdon is Professor at the Institute of Education, University College London. Her research on Economics of Education is based mostly on statistical analysis of education datasets from developing countries. On the basis of this research, she advises governments and donor agencies. She also serves on the board of a K-12 registered society school in India. Grace Bantebya Kyomuhendo is Professor of Women and Gender Studies, Makerere University, Kampala, and a distinguished social anthropologist, researcher and advocate for women’s and girls’ empowerment, gender equality and social transformation and justice. Grace has carried out research in the fields of gender norms, women empowerment, poverty, reproductive health and has over 37 publications in the respective fields. Keith M. Lewin is Emeritus Professor of International Development and Education at the University of Sussex (see www.keithlewin.net). His research covers educational planning, finance, assessment, science and teacher education. He directed the DFID-supported Centre for Research on Educational Access, Transitions and Equity (CREATE) and was research advisor to the Indian government’s programme to universalise secondary education. He chairs the Trustees of the UK Forum for International Education and Training (UKFIET). Angela W. Little is Professor Emerita at the University College London Institute of Education. She is a member of the Academy of Social Sciences and was previously a Fellow of the Institute of Development Studies at the University of Sussex. She continues to be actively engaged in the field of Education and International Development through research, teaching and practice (www. angelawlittle.net). Rabea Malik is a sociologist of education. She is Assistant Professor at the School of Education in Lahore (currently on leave) and a Fellow at the Institute of

Contributors xvii Development and Economic Alternatives (IDEAS). Her research focuses on inequalities in education, early childhood education and reforming delivery systems for social services. Florence Kyoheirwe Muhanguzi is Senior Lecturer at the School of Women and Gender Studies, Makerere University, Kampala. Her area of interest is gender-focused research in the fields of education, sexual and reproductive health, population and development, sexual and gender-based violence and adolescent girl’s well-being. She is a gender trainer and activist for women’s and girls’ empowerment. Arif Naveed is Lecturer at the Department of Education, University of Bath. Trained in economics and sociology of education, he investigates the intergenerational social mobility resulting from the expansion of mass schooling in the Global South. He implemented community-based poverty reduction projects in rural Pakistan, was the country lead for RECOUP (2009–10) and contributed to the design of the educational reforms in Punjab (2013–18). Alula Pankhurst is the Ethiopia Country Director of Young Lives and a member of the RISE Ethiopia team. He taught Social Anthropology at Addis Ababa University. His research has focused on poverty and well-being, child work and marriage, adolescence and youth development, migration and displacement, food security and social exclusion, customary justice and the longerterm impacts of interventions. Harry Anthony Patrinos is Education Global Practice Manager for Europe and Central Asia at the World Bank. He specialises in the economics of education, particularly the returns to schooling, school-based management, demand-side financing and public–private partnerships. Pauline Rose is Professor of International Education at the University of Cambridge, where she is Director of the Research for Equitable Access and Learning (REAL) Centre in the Faculty of Education. Pauline directed UNESCO’s Education for All Global Monitoring Reports (2012–14). Her publications examine inequalities in educational policy and practice, in relation to poverty and gender and with respect to financing and governance. Nidhi Singal is Professor of Disability and Inclusive Education at the Faculty of Education, University of Cambridge. Nidhi’s research problematizes simplistic binaries between mainstream and special settings and proposes a more sophisticated understanding of inclusive education. It provides new conceptual and methodological insights which move beyond access to examining issues of quality and outcomes of educational provision for children/young people with disabilities. Louise Yorke is Research Associate with the REAL Centre, Faculty of Education, University of Cambridge as part of the Research for Improving Systems of Education (RISE) Ethiopia project. Her current research interests include girls’ education and gender equality, equity in education, education systems analysis and the politics of education.

Acknowledgements

The seeds for this book were sown in the tributes that flooded in from friends and colleagues around the world following the untimely death of Professor Christopher Colclough in June 2017. Chris had clearly inspired the academic careers of numerous colleagues not only at the Institute of Development Studies at Sussex University and the University of Cambridge, where he worked but also collaborators, for example, in Ghana, Kenya, India, Pakistan and South Africa. The editors of this book were fortunate to meet some of these colleagues at a memorial service to commemorate Chris’s life held at Corpus Christi College, Cambridge on 3rd November 2017, where we reflected on his life, his personal achievements and his extraordinary contribution to the field of international education and development economics. Many of us had the opportunity to meet again at the conference in Cambridge organised by the Research for Equitable Access and Learning (REAL) Centre that had built on Chris’s research and teaching legacy. With over 100 participants, the conference provided the opportunity to learn more about the work that Chris had inspired during his long career. We are grateful to all of those who participated in these events and who encouraged us to put together this collection of research that was inspired by Chris’s example and the opportunities he created for so many of us to collaborate across national boundaries and disciplines. It was apparent that it would be fitting and to mark his legacy by revisiting the field of education and international development and the fight against educational inequality globally to which Chris had dedicated his career. This book is included in the Routledge series on Education, Poverty and International Development, which was co-founded and co-edited over a period of 10 years by Chris and Madeleine Arnot. We are particularly thankful to Madeleine for initiating this editorial collaboration and for her guidance, encouragement and sustained support throughout the process. We are also grateful to all those who have contributed chapters to the book. Amongst the many others who have contributed to the production of this book in different ways we would like to pay particular thanks to Holly Ruffhead in the REAL Centre, Faculty of Education at Cambridge and Emilie Coin and Swapnil Joshi at Routledge. All of us who have contributed to this book have different memories of our engagement with Chris, but we share a common concern with each other and with Chris that what is needed is rigorous research that generates the evidence on how to tackle the social injustices that hold back progress in education. We hope that the chapters in this book help to move further forward these efforts that aspire to the goal of achieving equality in education.

Abbreviations

ABL ASER CBE CPR CSA DFID DISE DPEP ECD EFA EPRDF ESDFP

Activity Based Learning, India Annual Status of Education Report, India Complementary Basic Education, Ghana Contraceptive Prevalence Rate Central Statistical Agency, Ethiopia Department for International Development, UK District Information System on Education, India District Primary Education Programme, India Early childhood development Education for All Ethiopian People’s Revolutionary Democratic Front Education Sector Development Framework and Programme, Sri Lanka ESDP Education Sector Development Plan, Ethiopia ETP Education and Training Policy, Ethiopia EVS Education Voucher Scheme FAWE Forum for African Women Educationalists FDRE Federal Democratic Republic of Ethiopia FGD Focus Group Discussion GDP Gross Domestic Product GEQIP-E General Education Quality Improvement for Equity, Ethiopia GES Ghana Education Service GOP Government of Pakistan GPE Global Partnership for Education GPHC Ghana Population and Housing Census GTP Growth and Transformation Plan, Ethiopia HBV(s) Home-Based Volunteers, India IDS Institute for Development Studies, UK IERT(s) Inclusive Educational Resource Teachers, India IFFEd International Finance Facility for Education IMF International Monetary Fund JVP Janatha Vimukthi Peramuna, India KP Khyber Paktunkhwa, Pakistan

xx  Abbreviations LC LHW MoE MoWA NGO NR OB OOSC PDHS PPP PROBE PTA R4D REAL RECOUP RISE RMSA RTE SDG SMC(s) SNNPR SRH SSA TEACh TFR TLM UER UNESCO UNFPA UNICEF UPE USAID WCEFA WEF

Local Committee, Ghana Lady Health Worker, Pakistan Ministry of Education, Ethiopia Ministry of Women’s Affairs, Ethiopia Non-governmental Organisation Northern Region, Ghana Operation Blackboard, India Out-Of-School Children Pakistan Demographic and Health Survey Public Private Partnership Public Report on Basic Education, India Parent Teacher Association Results for Development Research for Equitable Access and Learning Research Consortium on Educational Outcomes and Poverty Research on Improving Systems of Education, Ethiopia Rashtriya Madhyamik Shiksha Abhiyan, India Right to Education, India Sustainable Development Goal School Management Committee(s) Southern Nations, Nationalities, and Peoples’ Region, Ethiopia Sexual and Reproductive Health Sub-Saharan Africa Teaching Effectively All Children, India and Pakistan Total Fertility Rate Teaching and Learning Materials Upper East Region, Ghana United Nations Educational, Scientific and Cultural Organisation United Nations Fund for Population Activities United Nations International Children's Fund Universal Primary Education United States Agency for International Development World Conference on Education for All World Education Forum

Foreword

A tribute to Christopher Colclough

Christopher Colclough was my first graduate student in 1967. He had come from Bristol with his first degree and the Powesland Memorial Prize in economics. We would have weekly supervisions in what was then the Department of Applied Economics – myself, Chris and Udi Gachaga from Kenya. I am not sure who learned the most and from whom, but it was always stimulating and enjoyable. Soon it became obvious that Chris had real talent for thoughtful economics and careful analysis, and he wanted to study further. It didn’t take long before we had agreed that he would stay on for doctoral study. At the time, manpower planning was a mainstream issue, designed to set an economic frame for expanding higher education in Africa. Chris decided to work on this, with Professor H.A. Turner as his formal supervisor and myself in the background as an informal one. By then, I had gone to the Institute of Development Studies (IDS) and Chris would often come down, contributing to a joint study we were by then undertaking, reviewing the manpower plans of some 40 African countries. It was not difficult to see Chris’s growing talents and working together was always fun and rewarding. By 1971, with his newly minted doctorate, Chris was off to Botswana to use these manpower planning skills for the recently independent government and form many of the close friendships and professional relationships which he retained over much of his life. These qualities gained for him much trust and respect from those in government. Chris went back many times to Botswana and wrote an early book about its economy. His professionalism and friendship were increasingly valued. Much later, someone remembers Chris saying that he thought he ought to leave Botswana at that time because he was becoming too powerful. What a beautiful and sensitive comment from someone who always was so modest. How many other expatriates have shown such awareness of the proper limits to their role in an independent country? In 1975, Chris was back in the UK and to my delight, came to IDS as a Fellow, undertaking research, some consultancies and teaching. In 1982 he wrote what he later thought was probably his most important single paper, certainly one of the most quoted. Prepared for the World Bank, he reviewed a very wide range of sources – sociological, anthropological and medical as well as economic – about the wider impacts of primary education. He showed how primary education in

xxii  Foreword the developing world brought productivity benefits for work in the informal sector and for smallholder agricultural production. Through literacy and numeracy, primary education also contributed to social and economic life, especially through better health, nutrition, birth spacing and speeding the demographic transition (Colclough 1982). One of the impressive features of this article to me is how it shows the openness and questing mind of Chris, shifting from his earlier economic focus on higher education and manpower planning to a broader and thoroughly documented rethinking of the place and benefits of primary education. The piece played a major role in helping the World Bank and many donors to start funding primary education, which until then they had treated as consumption and ineligible in their concentration on what they thought was investment. Chris’s influence and reputation grew. This paper played a key part in shifting aid-supported education programmes away from secondary and higher education, towards the primary level – a shift undertaken by the World Bank, by what became the UK Department for International Development (DFID) and by a number of other bilateral donors over the following years. The article also helped set the stage for the first global UN Conference on Education for All held in Jomtien, Thailand, in 1990, at which 155 governments and supporting non-governmental organisations made commitments for expanding primary education, so that all girls and boys would have a place. For this, Chris co-authored a key strategy paper, later published in 1993 as Educating All the Children: Strategies for Primary Schooling in the South (Colclough & Lewin 1993). This became the first standard reference documenting the practical possibilities of achieving universal primary provision, even in poorer countries with severe budgetary constraints. As an economist, Chris never allowed himself to be trapped within the limited confines of education, let alone of neoclassical orthodoxy. His writings and consultancy reports set education within the broader context of wages and employment, incomes policy, human resources planning, public sector pay and alternatives to structural economic adjustment (now called austerity). He critically examined economic and policy issues, often challenging orthodoxy and putting forward alternatives. States or Markets (Colclough & Manor 1991) attracted attention in the world of academia towards the end of the Thatcher era, arguing that (for practical policy) the choice should never be state or markets but a judicious balancing of each, requiring careful analysis in relation to country, time and context. It was typical of Chris that he never played to the gallery, whether of students, colleagues, political or international. Although a strong academic, Chris was never an ivory tower economist. Between 1993 and 2000, Chris enjoyed a long-term policy advisory role in South Africa. Working initially with the African National Congress until transition, and then with the new ministry post-apartheid, he helped design a new legislative framework for education, securing structural change away from excessive expenditures allocated to the schooling of Whites, to a system of equal subsidies for the different racial groups. The essential features of this new school financing policy are retained to this day. During this time also, he led a multi-country research

Foreword xxiii programme on Gender and Primary Schooling in Africa in collaboration with the Forum for Women Educationalists (Colclough et al. 2003). The legacy of this programme is evident through some of the chapters included in this book. His international work had helped set the Jomtien goals for Education for All and for UNESCO’s annual monitoring of country progress towards them. He had been an adviser to the UK Parliamentary Committee on Overseas Aid and Development, the Norwegian Agency for Development Cooperation, Irish Aid, the World Bank, UNICEF and the Rockefeller Foundation. In the mid-2000s, he provided the technical analysis and the evidence base behind the UK making a ten-year pledge to provide £8.5  billion from its aid budget to support basic education. The international highpoint of Chris’s career came in 2002 when he was appointed by UNESCO as the founding director of the Education for All (EFA) Global Monitoring Report. This report has become the main instrument for holding governments and agencies to account for the commitments they have made on education. Education for All (EFA) was by then high on the agenda of international development and this authoritative, frank and well-presented report documented progress towards the EFA goals agreed at Jomtien. The 2002 Education for All: Is the World on Track, the first report, received front-page treatment in the press of some 160 countries. This outreach was in large part the result of Chris insisting that the report must be free of UN jargon and bureaucratic censorship, values easy to praise in a university, not so easy to achieve in the United Nations. Chris set the stage for these arrangements, unprecedented, I think, for UNESCO, by holding out for them right up to negotiations with UNESCO’s Director-General. The result was that Chris alone, as director of the report, was responsible for the report’s content and conclusions. The next two global reports were equally bold: Gender and Education for All: The Leap to Equality (2004) and Education for All: The Quality Imperative (2005). In short, by the time he came back to Cambridge as a professor, Chris had established himself as one of the world’s leading economists specialising in education. His career was already well established but in fact it became the stepping stone for another burst of creativity and contribution. In 2004, Chris was elected as Commonwealth Professor of Education and International Development and as Director of the newly established Centre for Commonwealth Education and subsequently Director of the Centre for Education and International Development. He remained in this position until his retirement in 2013. In 2004–05, he was President of the British Association for International and Comparative Education (BAICE). He won an Honorary Degree from the University of Katholieke University of Leuven. Chris was successively Fellow and Life Fellow of Corpus Christi College, where he had gained his doctorate many years earlier and a Fellow of the Academy for Social Sciences. In Cambridge, Chris used his base and international contacts for establishing an international Research Consortium on Educational Outcomes and Poverty (RECOUP), bringing together a multidisciplinary team of researchers from seven institutions in India, Pakistan, Ghana, Kenya and the UK. This led to an edited collection, Education Outcomes and Poverty: A  Reassessment (2012) and

xxiv  Foreword then to a Routledge book series on Education, Poverty and International Development, which Chris co-founded and co-edited with Madeleine Arnot. In this pinnacle of his academic career, the Faculty of Education in Cambridge benefitted enormously from his leadership and teaching, in bringing together comparative education and international development. This work offered new insights on education access, quality and outcomes in the Global South. It is a fitting tribute that this edited collection is published in this series, written by authors many of whom worked with Chris during different stages of his career, including ones associated with RECOUP. Chris died too young, from cancer at the age of 70. But he had already made important and lasting contributions to future generations through education and economic policy in many countries, especially but not only in Africa. His research and writings cover a wide field and were often pioneering. Chris was thoughtful, honest and careful as an economic analyst, a true professional. He was also a talented pianist and cello player, which helped shape his deep sense of humanity. His qualities, not always found in the academic world but treasured when they are, were those of scholarly rigour, calm rationality and high regard for intellectual debate combined with generosity of spirit, graceful leadership and his ability to make everyone feel their contribution was valued and should be heard – as well as good humour, a sense of fun. Sir Richard Jolly Honorary Professor and Research Associate Institute of Development Studies

References Colclough, C. (1982) The impact of primary schooling on economic development: A review of the evidence. World Development, 10(3), pp. 167–185. Colclough, C. (2012) Education Outcomes and Poverty: A Reassessment. Abingdon: Routledge. Colclough, C., Al-Samarrai, S., Rose, P., & Tembon, M. (2003) Achieving Schooling for All in Africa: Costs, Commitment and Gender. Aldershot: Ashgate. Colclough, C. & Lewin, K. (1993) Educating All the Children: Strategies for Primary Schooling in the South. Oxford: Clarendon Press. Colclough, C., & Manor, J. (Eds.) (1991) States or Markets? Neo-liberalism and the Development Policy Debate. Oxford: Clarendon Press. UNESCO (2002) Education for All: Is the World on Track? Education for All Global Monitoring Report. [Online] Paris: UNESCO. Available from: https:// en.unesco.org/gem-report/report/2002/education-all-world-track [Accessed 21 July 2020]. UNESCO (2004) Gender and Education for All: The Leap to Equality. Education for All Global Monitoring Report. [Online] Paris: UNESCO. Available from: https:// en.unesco.org/gem-report/report/2003/gender-and-education-all-leap-equality [Accessed 21 July 2020]. UNESCO (2005) Education for All: The Quality Imperative. Education for All Global Monitoring Report. [Online] Paris: UNESCO. Available from: https://en.unesco. org/gem-report/report/2005/education-all-quality-imperative [Accessed 21 July 2020].

1 Education and the reform of social inequalities in the Global South An introduction Pauline Rose, Madeleine Arnot, Roger Jeffery and Nidhi Singal This book represents our tribute to Professor Christopher Colclough (1946– 2017), one of the most influential world-leading development economists who was devoted to rigorous evidence for tackling educational inequalities. Chris’s influence went far beyond academia. Throughout his career, whether at the Institute for Development Studies (IDS) at Sussex University or at the University of Cambridge, Chris worked to ensure that research was relevant to the field of education and international development, to the needs of educational policy-makers and to national governments and communities. In addition to his impressive research and publications record, he spent four years at UNESCO establishing and directing the flagship Education for All Global Monitoring Reports (UNESCO 2002, 2004, 2005), providing independent evidence-based assessments of progress towards providing every child in the world with an education. Throughout his career, he produced evidence for international agencies and national governments with the aim of shifting their priorities towards the world’s poorest and most disadvantaged children. For the editors of this book, Chris was our much-esteemed colleague, a mentor and a great friend. Support for the training and careers of others was important to him, including for many of the authors of the chapters in this book. The tributes to him that poured in at memorial events referred movingly to his exceptional collegiality. We are all deeply indebted to his scholarship and wisdom and greatly admired his persistent, firm pursuit of social justice, in particular in some of the poorest countries in the world. His humanity spoke through his research at every level. Colleagues from many different institutions and countries were fulsome in their praise, with some describing him as having ‘intellectual distinction with a human face’;1 he was said to mix ‘the science of argumentation with the art of kindness’.2 On 3rd November  2017, when we sadly came together at the memorial event held in his honour, we became aware that his colleagues were keen to celebrate his work and its continuing relevance. This led to a conference on 9th March  2018 held at the University of Cambridge, including presentations by many of the contributors to this book.3 We decided that a fitting tribute to Chris and his great contribution to the field of education and international development was to build on it, demonstrating how his agenda of tackling and reducing

2  Pauline Rose, et al. education and social inequality in Southern contexts is being taken forward into the contemporary climate. In 2016, Chris pointed out in a major speech celebrating 25 years of IDS work that the macro strategy of neoliberals letting the market work was associated with an enormous rise in inequality. His analytical acuity recognised that, globally, social inequality was a growing not a lessening problem that needed addressing continuously and with detailed analysis. His insights into the world’s predicaments were highlighted in his conclusion: [I]nequality has risen enormously over the past two generations. Its consequences are patent everywhere from the ballot boxes of the rich industrialised nations to the migrant boats from Africa and Asia. Its impact is felt in the rise of nationalism in countries of both the South and the North, and even, partly, in the growth of terrorism. These matters should have a closer focus. (Colclough 2016) As editors, we have therefore chosen as the theme of this collection an investigation into how patterns of social inequality have been addressed by educational reforms over the last few decades and how these reforms are reshaping social inequalities today. We invited Chris’s colleagues to contribute to this collection by revisiting their research data in light of contemporary changes or by offering critical insights into how the expansion of education in their own countries was now impacting on social inequality.4 The relationships between schooling and social inequalities must be looked at holistically and dynamically. By their nature, they can be aggravated by even the most well-intentioned reforms. The starting point for authors in this volume is a recognition that schooling systems have a history of sifting and sorting generations of children (such that educational inequalities are produced internally), but that the unequal patterns found in educational access and in the outcomes of schooling are also shaped by wealth, power and status inequalities in the wider society. Therefore, attempts to address the problem of education inequalities only through education reforms, whilst important, are likely to fail if they do not take account of the economic and social contexts surrounding schools. Such contexts are nationally shaped, leading to a vital need to get to grips with the particular contextual factors that influence national and local educational scenarios. Policy processes, for example, are fraught with difficulties and outcomes are, more often than not, unintended. Social inequality is not, as sometimes assumed, greater within urban spaces rather than rural cultures. The latter tend to be represented as homogeneous and collective, yet as authors of several studies in this volume indicate, within the collective lie many complex sets of unequal power relations. Within each space and culture lie inequalities not just of wealth but also of capability (Sen 1999; Walker & Unterhalter 2007), of different forms of economic, social and cultural capital (Bourdieu 1986) and of status, value and respect (Jeffrey et al. 2007; Jeffrey 2010).

Education and the reform of social inequalities 3 A major challenge for educational policy-makers and practitioners, therefore, is how to address those inequalities that lie outside of schooling institutions, but which enter into its classrooms and shape the identities and aspirations and strategies of young people, their teachers and parents. Research on these patterns and restrictions is vital to achieving the ambitions set in 2000 by the international development community through the Millennium Development Goals and the Education for All agenda and, more recently, with respect to the Sustainable Development Goals (SDG) (Sachs 2012). Sustainable development as an ambitious set of goals affects all global, national and local development contexts. Its overarching concern for ‘leaving no-one behind’ sets an agenda beyond income alone, to incorporate both economic and social dimensions of disadvantage, associated with factors such as poverty, gender, caste, disability, migration and locality. Over the last ten years, many resource-poor countries have expanded their education systems dramatically, moving from elite to mass systems of schooling, designed to be universal (UNESCO 2014; World Bank 2018). While such moves have provided new opportunities for some children and their families, the goals of universal coverage are still to be achieved. Wide diversities in classrooms pose new challenges for teacher education, teachers, schools and approaches to teaching and learning (Akyeampong 2017; Moon 2013; Sayed 2018; Schweisfurth 2015). With its emphasis on addressing cross-sectoral challenges for sustainable development, the SDG agenda for 2030 presents new research questions that require sustained attention. In particular, researchers need to focus on the nature and extent of social inequalities within specific national development contexts and on how such inequalities affect, transform and militate against current educational reform initiatives (e.g. Unterhalter  & North 2019). It demands new research approaches that take account of the political economy of reform processes within countries (Pritchett 2018). For example, the long-standing debate on whether marketisation and privatisation in education perpetuate inequalities (Colclough & Manor 1993; Colclough 1997) has, if anything, intensified in recent years and deserves closer scrutiny (Day Ashley et al. 2014; Macpherson et al. 2014; Srivastava  & Walford 2018; Verger et  al. 2012). In the context of this ambitious agenda, it is also now particularly important to reassess how social-economic and related unequal relations such as gender, caste, disability and sexuality shape the choices made of these different types of schooling or the ways in which different school experiences shape future lives – how, in effect, the inequalities associated with such relations hold back progress. These issues require careful consideration of the relations of power and elite privilege that underlie the failure so far to develop effective policies and practice. These challenges are taken up by the community of authors in this book. We begin by identifying the intellectual agenda which we associate with Chris, before describing the contributions which individual authors bring to this agenda. We highlight, in particular, the continued relevance of drawing upon different disciplinary perspectives and a range of methodologies, the importance of linking

4  Pauline Rose, et al. the international agendas to national, institutional and local structures, processes and experiences and the need to work in close partnerships with colleagues in the South.

In tribute to Christopher Colclough Sir Richard Jolly, a long-time close friend of Chris describes in his foreword how and why Chris was recognised as a world-leading development economist working in education. Chris promoted an agenda, based on his own and colleagues’ research, on the ways in which education was a vital lever for poverty alleviation and economic development. Through his immense personal influence, and extraordinarily successful engagements with the aid community, Chris encouraged world leaders and global institutions and agencies to develop wide-ranging accountability systems that would encourage all governments to engage in the task of implementing the principle that all children could achieve their right to education, regardless of their background. His co-authored book Educating All the Children (Colclough with Lewin 1993) drew the world’s attention to the economic and social value of primary schooling, specifically in Southern contexts, and to the financing that would be needed to address the costs of it. Another early seminal intervention, as Richard Jolly points out, was the edited book States or Markets? Neo-liberalism and the Developing Policy Debate (Colclough  & Manor 1993), which engaged with the issue of who should provide education in the name of human development and under what conditions. Chris’s publications repeatedly addressed the strong links of education to labour market outcomes, health and gender (Colclough 1997; Colclough et al. 2003). The debates provoked by his findings are still central to our understanding of policy options and possibilities. These empirically based convictions notably led Chris to support the need for interdisciplinary dialogue and contextual relevance through bringing together contributions from a wide range of disciplines and national settings. He recognised that the economic approaches to education (in particular through the adoption of human capital theory), while important, were not sufficient in themselves. In a reflexive piece in 2010, Chris also argued strongly that the most challenging of international debates in comparative education and development studies should come together in a ‘common cause’ (Colclough 2010). If such a cause could be found, research in these two fields of study could find a way in which educational reforms (locally, nationally and internationally) contribute to the transformations in economic, political and social orders in lower-income countries so that they reduced the widespread poverty found worldwide. Arguably, the bringing together of these two fields of study would allow national comparisons of the effects of decentralising or privatising policies on different educational systems; the external factors (such as poverty, gender, race/ethnicity, disability and sexuality) affecting schooling outcomes (such as literacy and numeracy and the significance of cultural diversification); and the promotion of empowerment as an educational goal. In this context, we welcome the number of country case studies in this volume.

Education and the reform of social inequalities 5 Drawing on his experience of working and living in Botswana and researching in sub-Saharan Africa and South Asia, Chris was aware that the economics of education needed to take account of the diversity of cultural, social, economic and political factors that are integral to the uneven economic development of Southern contexts. He recognised the complexities of structural inequalities and how these are further manifested through a complex interplay of colonial histories and current local norms and values. Consequently, Chris was also keen to connect his quantitative evidence with qualitative researchers for more nuanced perspectives on global challenges such as poverty alleviation. This approach was initially apparent in the Rockefeller-funded research on ‘Gender Inequalities and Primary Schooling in Africa’ (Colcough et  al. 2003), which Chris led at the Institute of Development Studies and later in the DFID-funded research programme that he led at the University of Cambridge through the Research Consortium on Educational Outcomes and Poverty (RECOUP) (Colclough 2012).5 Both projects, using quantitative and qualitative methods, offered original analyses of education’s role in Southern contexts in reproducing social disadvantage. In each of these programmes, quantitative data highlighted the extent of disadvantage some children faced in terms of their access to a good-quality education, while qualitative data usefully exposed the complexity of the reality of people’s lives that influenced their children’s education. Such data had a major contribution to make to those policy-makers who might be tempted by unicausal explanations of complex social patterns or by ‘quick fixes’. Unusually for someone with a strong quantitative background, Chris also supported those of us who produced evidence of social inequalities that seemed initially to be far from policy-makers’ agendas, but which he recognised were essential contributions to the field of education and international development. Research openness requires learning from researchers who are most knowledgeable about challenges and possibilities in their local contexts and the impact of international agendas on their national governments and societies (Mitchell et al. 2020; Tuhiwai Smith et al. 2018; Denzin et al. 2008; Robinson-Pant 2005; Robinson-Pant  & Singal 2013; Breidlid 2012). When leading major empirical research projects, Chris worked in partnership with scholars based in Southern institutions and those who strongly identify themselves as Southern researchers, given their personal and professional positionings. A number of such partners have contributed to this collection (Anuradha De, Geeta Gandhi Kingdon and Nidhi Singal, India; Monazza Aslam, Feyza Bhatti, Rabea Malik and Arif Naveed, Pakistan; Florence Kyoheirwe Muhanguzi, Uganda and Leslie CaselyHayford, Ghana). Such knowledge allows for richer perspectives and research that engages with the complexities of their contexts (Aikman  & Dyer 2012; Aikman et  al. 2016; Connell 2007). As Connell (2007, p.  21) points out, the knowledge and power of social science from Southern contexts is a ‘tremendous resource that has been disregarded by mainstream social science’ – a knowledge source that has ‘wide-ranging implications for social science in the 21st century’.

6  Pauline Rose, et al. Individual chapters reflect Chris’s priorities when they consider evidence in relation to education and social inequalities in various South Asian and sub-Saharan African contexts, some of which are furthest from achieving global goals, including some countries that have made rapid progress in terms of increased enrolment numbers but less so in global learning metrics associated with literacy and numeracy (including Ethiopia, Ghana, India, Pakistan, Sri Lanka, Tanzania and Uganda). Using a range of disciplinary perspectives, contributors to this collection employ different types of data – from large-scale internationally generated quantitative data sets, through to national and local surveys, classroom observations and qualitative interviews with teachers and other key education stakeholders, to in-depth analysis of family voice data. These data are used to understand how educational trajectories, choice of types of schooling, professional identities and school practices, patterns of social deprivation and social and individual aspirations come together. The contribution of the book therefore, as Chris would have wished, is its diversity, as well as the range of issues and contexts it addresses, drawing on the perspectives of scholars based in institutions in both the North and the South. In the following pages we describe the key themes addressed in individual chapters.

Key themes Continued challenges in closing inequalities within and around educational systems internationally can largely be traced to their historical, economic and political roots. The interconnections between these roots are highlighted in many of the chapters in Part I which show variations in this influence in different contexts and across types of schools. In some instances, the root causes of inequalities have been or are being more successfully challenged than in others. In Part II, authors delve deeper into the lives and realities faced particularly by families living in poverty and the choices and aspirations of parents and young people – sons, daughters and children with disabilities. Authors further consider how the social distance between teachers and their students can affect children’s educational experiences.

Part I: the economics and politics of educational reform Part I  begins by establishing the links between the economics and politics of reform processes and their implications for social inequality, particularly the impact of economic differences between families and the impact of poverty, social class, caste and region on young people’s prospects. It identifies the economic benefits of expanding access (Chapter 2) as well as the costs of doing so (Chapter 3). Recognising the distance that many education systems are from achieving SDG4 which aims to ‘ensure inclusive and equitable quality education and promote lifelong learning opportunities for all’, the chapters analyse the political challenges of successful reform associated with tackling the quality of education and diversified forms of provision that has resulted in differential progress for populations within countries such as Ethiopia, India, Pakistan and Sri Lanka.

Education and the reform of social inequalities 7 Building on his work with Chris for RECOUP, and related to his own PhD which Chris supervised, Harry Anthony Patrinos employs human capital theory to highlight the benefits that education has for achieving economic returns. In Chapter 2, he develops his influential work on the patterns of returns to education to show how these have changed over time, shifting from more emphasis on primary schooling to returns becoming more apparent at higher levels of education. Patrinos attributes this to three possible facts that: primary schooling is now near universal globally, resulting in a fall in returns at that level; the expansion of primary and secondary schooling could be at the expense of quality; and technological change is resulting in demand for higher-order skills. From an equity perspective, Patrinos highlights important policy considerations that derive from his analysis, notably that public funds should continue to support the development of basic cognitive skills from early years of education, while also highlighting a need to expand the education system at secondary and higher levels to avoid a widening of income inequality. This approach is influential in making the case to national governments and international agencies to invest even more in education, as well as to guide the distribution of public resources across different levels of education in ways that are beneficial for economies and for addressing inequalities. In Chapter 3, Keith M. Lewin provides an insightful analysis of how the educational landscape for low- and lower middle-income countries has changed in the last three decades and the factors that continue to shape unequal access to education and the pattern of educational financing. He employs a typography of four different transitions that shape how education in low- and lower-middle income countries will develop: demography; the flow of students through patterns of enrolment; gendered participation; and, the number of children out of school. Building on his work together with Chris for the Jomtien World Education Forum in 1990 (Colclough with Lewin 1993), he identifies the consequences that unequal patterns in access have for aid and national spending for education, noting that aid is diminishing in importance. Highlighting shifting trends in financing over the past 30 years, he identifies the need for a twopronged approach in educational financing. The first prong is to support research and investment in education system planning and management to promote durable gains in efficiency and effectiveness so that education systems can be financed from domestic revenue. The second prong is to invest in supporting fiscal reforms that can increase domestic revenue. Angela W. Little, who also collaborated with Chris at the 1990 Jomtien World Education Forum, went on to study in-depth the historical and political ‘drivers and interests’ shaping egalitarian education reforms in India and Sri Lanka in pre-colonial and post-colonial times. In Chapter 4, she compares and contrasts the political paths taken by the two countries leading up to, but especially after, national independence in relation specifically to social inequality in access to basic schooling. Her detailed account identifies how Sri Lanka has been more successful in promoting equitable education progress which could, in part, be attributed to India’s greater diversity, resulting in a greater degree of competition and

8  Pauline Rose, et al. conflict between a larger number of diverse interests groups. She suggests that this potentially slowed the ability of successive Indian governments at all levels to introduce and implement policies designed to reduce inequalities in education. Even so, enrolment overall has expanded in both countries, which leads Little to ask if the changing social composition of state and private enrolments could promote or thwart the ability of the education system in each country to increase social equality in the longer term. Linking this question to a contemporary context, in Chapter 5, Monazza Aslam and Geeta Gandhi Kingdon build on their work with Chris for the RECOUP project to explore the possibilities and challenges faced not by private schools per se but by different models of public–private partnerships (PPP). They consider if such models offer an ‘enabling environment’ that helps reduce inequalities or whether the aims of improving access to education and achieving better quality education through PPPs could reinforce or even exacerbate inequalities that exist in the education system. Through a careful assessment based on rigorous reviews of evidence in the Indian and Pakistani contexts (two countries where experimenting with private–public partnerships has been particularly common), they argue that these partnerships have the potential to extend access and may also improve the quality of education, including for the poorest. They highlight different forms of partnerships, ranging from charter school-type models where the private sector takes over the management of public schools to voucher schemes that aim to provide parents with a choice for their children’s schooling. Recognising that this is a highly contested area of reform, they conclude by identifying the importance of adopting an explicit plan for when, how and why the potential of public–private partnerships should be recognised and adopted with the aim of achieving equity in education, rather than the ad hoc approach that is currently the norm. Chris had a particularly close interest in sub-Saharan Africa, having started his academic career in Botswana and subsequently led a major nine-country study in collaboration with the Forum for African Women Educationalists (FAWE) in the late 1990s. His co-authored book Achieving Schooling for All in Africa: Costs, Commitment and Gender (Colclough et al. 2003) offered an analysis of new, rich, quantitative and qualitative data that complemented the, then, existing econometric evidence on the determinants of gender differences in schools across different contexts in sub-Saharan Africa. The value of the study was to bring gender into the heart of discussions on education reform, importantly with respect to debates on the financing of schooling. Given this record, it was not surprising that gender was embedded in the Education for All Global Monitoring Reports which Chris led, with one report specifically dedicated to this theme (UNESCO 2004). Returning to one of the countries in the original study, Ethiopia (some 30 years later) has allowed Pauline Rose along with Louise Yorke and Alula Pankhurst to demonstrate that there has been considerable progress in moving from an elite to a mass system of primary schooling. However, Chapter 6 reports that, although throughout the period there has been explicit commitment to gender equality,

Education and the reform of social inequalities 9 this commitment has not sufficiently been realised. Gender gaps remain with respect to enrolment and wider societal factors that result in girls’ early marriage as well as sustaining their burden of domestic work. Drawing on an innovative approach to understanding how politics influences education systems, they adopt the lens of Hickey and Hossain’s (2019) domains of power framework to identify how formal and informal institutions as well as Ethiopian officials and educators shape and reinforce gender inequalities in and around education. The authors advocate a more transformative approach to girls’ education, with political leaders moving away from the dichotomy of mainstreaming and targeted models and from simplistic notions that an increase in girls’ access and retention will lead to greater gender equality. They also emphasise that the politics of educational reform demand that all stakeholders are included in reforms to tackle gender inequalities.

Part II: challenges and opportunities in addressing inequalities through education The six chapters in Part II contribute to these ongoing debates by offering deep insight into how inequalities in the wider society impact on, and are embedded in, schooling itself. The first three chapters, using socio-cultural qualitative analyses, highlight the importance of looking beyond the school to understand how education shapes and is shaped by aspirations. Chapter 7 by Arif Naveed and Chapter 8 by Feyza Bhatti and Roger Jeffery are both based on family-focused research carried out in Pakistan in the original RECOUP fieldwork sites. They consider the meanings and values attached to schooling of villagers who are poor or in precarity, just above poverty lines in Punjab villages. The appeal of schooling for families living in poverty is that it is an aspirational resource which can be within their compass to achieve, at least to a limited extent. Arif Naveed’s chapter takes us deep into the relationality of family life to explore this aspect. Using data from a follow-up study of the same families used in the RECOUP project, he employs Appadurai’s (2004) concept of ‘capacity to aspire’ to understand their negotiation of social inequality in the village. His novel methodology draws on intergenerational interviews (with mothers, fathers, sons and daughters in the same family) to show the complex engagements with what he calls the pentagonal structure of the village – the inequalities associated with landownership, kinship and caste structures, religion, patriarchy and relationships of patronage. Whilst families have high hopes of using schooling to achieve a form of social mobility or taraqqi (a better life), these structural power relations and their material realities reduce or even destroy such aspirations, thus sustaining the cycles through which social inequalities are reproduced. Educational and economic reforms need to break such painful cycles, not just at the individual level but for the collective. Failing to do this has led to increasing evidence that families living in poverty are giving up on formal schooling when they see the route

10  Pauline Rose, et al. to better forms of employment is blocked. The educational aspirations of families and communities could be used by policy-makers more effectively, as Appadurai noted: [I]t is in culture that ideas of the future, as much as of those about the past, are embedded and nurtured. Thus, in strengthening the capacity to aspire, conceived as a cultural capacity, especially among the poor, the futureoriented logic of development could find a natural ally, and the poor could find the resources required to contest and alter the conditions of their own poverty. (Appadurai 2004, p. 59) Feyza Bhatti and Roger Jeffery’s (Chapter 8) research indicates that schooling itself can be a way of raising the ‘capacity to aspire’. Their study of attitudes to fertility and childbearing of young mothers in Pakistani Punjab shows how their own schooling is closely linked to their aspirations for their children. As family systems are reshaped under modernising pressures, young women – whatever their own schooling – are increasingly aware of the need to limit their families (both boys and girls). Although schooling outcomes are unpredictable, household ambitions involve heightened aspirations, internalised by these young women. In this revealing study, young women have more freedom, in conjunction with their husbands, to take decisions relating to their own fertility and they use this shared decision-making power in similar ways across these lower-class households. Household ambitions for their children, the expenses of everyday life and within them, schooling costs (especially if paying for increasingly highly priced and often private schooling) are all changing the ideals of young mothers with respect to family size, the use of contraception and the value of girls when compared to boys. At the same time, despite the resulting fertility decline, the new staging of childbearing and more investment into childrearing, the ‘liberating force’ that education could represent for young mothers and their families may be insufficient to make a real difference to their futures. Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo (Chapter 9) pick up on themes from the research programme on Gender and Primary Schooling in sub-Saharan Africa and links with Chapters 7 and 8 by going deeper into the relationships between gender, sexuality and education. Their work demonstrates the importance of recognising that, whilst increasing girls’ schooling is beneficial, the global agenda on education does not yet engage with sufficient understanding about the ways in which teenage pregnancy represents a major impediment to schooling achievements and acts as a cause of continuing gender inequalities. Drawing on an illuminating qualitative study that uncovers the perspectives of young women and men and community members in Uganda, the authors show how beneath the surface of such pregnancies are the realities of poverty, sexual pressure on girls and male-dominated cultures. Despite the government’s high-level commitment to both improving girls’ schooling and reducing

Education and the reform of social inequalities 11 the prevalence of teenage pregnancy, its policies have been poorly implemented. Prevention strategies have included sexual and reproductive health education in schools, but this has been resisted by teachers and village elders. It is anyway thought by girls to be inadequate, leaving them being blamed when they become pregnant. Strong patriarchal norms and patterns of behaviour protect young men but not the young mothers whose education careers are badly affected. To prevent this tight coupling of teenage pregnancy with poverty, schooling – not just schools – needs to provide more support to young women to avoid pregnancy, and policy-makers need to hear and take heed of young women’s views about the pressures they face. The final three chapters explore how education systems interact with social inequality, and bring classroom practices and interactions into focus. All three chapters include researchers who were centrally involved in the RECOUP programme. The authors report on their recent research, which includes mixed methods approaches, linking survey data to in-depth understandings of the culture of schooling systems. Their work provides windows into these pedagogic systems by focusing on some of the classroom experiences offered to the most marginalised: children who are out of school, those with disabilities and those who live in relatively unstable, precarious households. They emphasise the need to look beyond efforts to increase school access for children from traditionally excluded groups by offering a critical analysis of what happens in the classroom. Leslie Casely-Hayford, Adom Baisey Ghartey and Justice Adjei-Quartey revisit findings from a major study of complementary basic education for outof-school children in two communities in the northern region of Ghana (Associates for Change 2013). Chapter 10 describes how parents and children, when asked about why they choose complementary rather than formal education, describe the mismatch between the pedagogy, discipline and culture of mainstream primary schools and their own perspectives on how schools should be organised. The authors’ extensive investigation of the complementary basic education programme provides important lessons for formal education provision, such as the importance of using enthusiastic local teachers, introducing mother tongue as the medium of these early years of schooling and drawing on indigenous cultures. Parents of those children who would be out of school normally value fee-free education, flexible timescales and a less disciplinary and more supportive child-centred pedagogy that develops children’s confidence in their learning ability. All these features of the complementary system could offer lessons for the formal system that would enhance the educational experience of their pupils. The authors of Chapters 11 and 12 address the danger that children who have entered schools in recent years as enrolment has expanded have little opportunity to learn. The poor quality of schooling they receive disproportionately affects those from disadvantaged backgrounds. In Chapter 11, Nidhi Singal draws on her influential work on poverty and disability to explore teachers’ engagement in promoting inclusion for children with disabilities and the challenges they face in doing so in rural north India. She first demonstrates that, although inclusive

12  Pauline Rose, et al. education has been a major policy shift with respect to the schooling of children with disabilities over the past two decades in India, little has been written that illuminates and engages with the complexity of how to deliver on this promise. She argues that, in international development literature, global proclamations driven by a largely Northern lens of the what and the how of inclusive education have stifled deeper engagement with the challenges and opportunities in Southern educational systems. Drawing on classroom-based data, and using Minow’s (1990) dilemmas of difference, she moves the discussion of inequalities experienced by children with disabilities to a deeper understanding of the difficult and dilemmatic decisions that teachers are faced within ‘inclusive’ school settings. Evidence shows that teachers in these schools were not simply rejecting children with disabilities, rather a closer look at their practices highlights the many quandaries that they are faced with on a daily basis as they navigate teaching and learning in complex and resource-constrained classrooms. Her chapter argues that, in order to address the continued exclusion of children with disabilities, there is a need for more contextually driven engagement with some fundamental questions around the purpose and value of schooling, rather than continuing to uphold the rhetoric of ‘inclusive education’ which is, in itself, shrouded in ambiguity. The real challenge here is to listen to teachers working in these contexts, as makers of change, rather than positioning them through a deficit-driven lens and blaming them for inadequacies. In Chapter  12, Anuradha De and Rabea Malik add to Nidhi Singal’s concern about teacher practices by showing how teachers are affected by their social distance from the students they teach. Building on their work with RECOUP, their chapter draws on rich data collected as part of the ‘Teaching Effectively All Children’ (TEACh)6 research project, based in India and Pakistan, to provide a deeper focus on how differences and potential inequalities were addressed by teachers. Their work recognises what a tragedy it would be if children who are in school do not learn. In that context, their study reports on the expectations of a sample of Indian and Pakistan teachers and their characterisations of pupils which, they show, can strongly influence the capacity of pupils to learn and to do well in class. Through their interactions with students, teachers communicate their expectations about how students from different backgrounds will perform. The teachers’ conceptualizations of ‘good’ students are much more likely to be present in children from socially and economically stable and well-off homes, somewhat similar to the teachers’ own backgrounds. While teachers engage more with ‘good’ students, they might not openly exclude the others; nevertheless, there is a ‘silent process of non-inclusion’. Through these expectations and their practices, teachers may be reproducing within class the disadvantages that the student arrives with, instead of addressing and alleviating the disadvantages that such children face. If this is the case more generally, then educational reformers need to engage far more deeply with the personal and professional identities and practices of teachers in such settings. Taking the educational research agenda forward, these two chapters forefront the need for more research focusing on what teachers do, alongside understanding

Education and the reform of social inequalities 13 how they explain why they act in these ways. Critical exploration of classroom processes provides a powerful lens for re-examining processes of inclusion and exclusion and pointing to how classroom spaces become important sites for the reproduction of social inequalities. Therefore, for educational reforms to be successful, these must be located firmly in the everyday happenings of classrooms in mainstream schools.

Reforming education This collection of research studies reveals new directions in the field of international development, identifying how education might more successfully challenge a range of social inequalities. The research it showcases contributes critical insights into how educational reforms in the name of social inequality can be investigated globally, nationally and locally, through different disciplinary fields, using a diverse, broad and deep set of methodologies and data. There are benefits of comparing and contrasting the experience of national government-led reforms across countries to assess differential progress towards global goals. At the same time, there is also a need to delve more deeply into particular national contexts, communities and institutions from a variety of perspectives. Attention should be paid to the range of different types of educational provision (from PPPs to complementary basic education) and to different governmental and nongovernmental organisation initiatives, locating them in specific demographic contexts. What are revealed here are some of the restrictions on closing large learning gaps and the loss of talent and opportunity at national and regional levels. What are also revealed are the rising reform agendas and shifts in the patterns of thinking about poverty contexts but also the unevenness of such reform agendas in terms of affecting communities, families and individuals (whether teachers, students or parents). The authors’ ambitions are diverse, but they share a goal of reviewing and moving forward policy debates such that they engage with the economic and cultural complexity and intersectionality of inequalities and also about the play of power of governments, elites and interest groups. Such power relations limit the resourcing and possibilities of those who wish to make education genuinely accessible and effective and the achievements of communities, families, teachers and students who wish to take up the opportunities that education can offer. Spurred on by the influence of Christopher Colclough, our community of researchers believe that, without insights into the various dimensions and factors explored and reviewed in this volume, genuinely inclusive education systems will not easily be achieved. On the 10th anniversary of the Education for All, Chris wrote a blog for UNESCO7 in which he noted that by 2010 some 40 million more children than in 2000 were managing to get to primary schooling. But he admitted that one should not be too optimistic. There are indeed now worries that numbers out of school will increase again and, that with the global COVID-19 pandemic, there will be a manifold increase in poverty and a simultaneous cutting back of aid

14  Pauline Rose, et al. budgets. As Chris noted, it needs a determined effort by wealthy countries to help lower-income countries, if the formers’ own economies are shrinking. The goal, he said, is clearly ‘morally right’ but, objectively, the truth is that it is in the interests of donor nations as well as recipient states to continue to support the sustainable education goals of quality schooling for all.

Notes 1 See Adrian Wood at www.ukfiet.org/wp-content/uploads/2017/07/Chris-Aselection-of-appreciations-from-colleagues-and-friends-in-IDS-and-Sussex-Univer sity.pdf. 2 Madeleine Arnot speaking at the 29th March  2019 Conference, Celebrating the Work of Professor Colclough, Conference, REAL Centre, Faculty of Education, University of Cambridge. 3 The conference was organised by the Research for Equitable Access and Learning (REAL) Centre in the Faculty of Education and was attended by over 100 participants, many of whom had worked with Chris over the years. 4 This anthology adds to the Routledge series on Education, Poverty and International Development, established by Christopher Colclough and Madeleine Arnot in 2010, which has as its mission to publish ‘Sharp, critical and innovative studies’ that are likely to have ‘a strategic influence upon the thinking of academics and policy-makers’. www.routledge.com/Education-Poverty-and-International-Development/bookseries/EPID 5 See RECOUP project website: http://ceid.educ.cam.ac.uk/researchprogrammes/recoup/ 6 See the TEACh research project website: www.educ.cam.ac.uk/centres/real/ researchthemes/teachingandlearning/effectiveteaching/ 7 See https://gemreportunesco.wordpress.com/2012/06/04/protecting-educationaid-is-more-vital-than-ever/

References Aikman, S.,  & Dyer, C. (2012) Education and inclusion: Re-examining the narratives, Compare: A  Journal of Comparative and International Education, 42(2), pp. 177–185. Aikman, S., Robinson-Pant, A., McGrath, S., Jere, C., Cheffy, I., Themelis, S.,  & Rogers, A. (2016) Challenging deficit discourses in international education and development. Compare: A  Journal of Comparative and International Education, 46(2), pp. 314–334. Akyeampong, K. (2017) Teacher educators’ practice and vision of good teaching in teacher education reform context in Ghana. Educational Researcher, 46(4), pp. 194–203. Appadurai, A. (2004) The capacity to aspire: Culture and the terms of recognition. In Rao, V., & Walton, M. (Eds.) Culture and Public Action. Stanford, CA: Stanford University Press, pp. 59–84. Associates for Change (2013) The Quality of Education in Northern Ghana: Assessing Learning Efficiency and Effectiveness across Ghana’s Three Northern Regions. VSO/ Comic Relief Commissioned Study. Available from: www.associatesforchange.org Bourdieu, P. (1986) The forms of capital. In Richardson, J.G. (Ed.) Handbook of Theory and Research for the Sociology of Education. Westport, CT: Greenwood, pp. 241–258.

Education and the reform of social inequalities 15 Breidlid, A. (2012) Education, Indigenous Knowledges and Development in the Global South. London: Routledge. Colclough, C. (Ed.) (1997) Marketizing Education and Health in Developing Countries: Miracle or Mirage? Oxford: Oxford University Press. Colclough, C. (2010) Development studies and comparative education: Where do they find common cause? Compare: A Journal of Comparative and International Education, 40(6), pp. 821–826. Colclough, C. (Ed.) (2012) Education Outcomes and Poverty: A Reassessment. London: Routledge. Colclough, C. (2016) States or Markets – 25 Years On. IDS Bulletin, 47(2A). Colclough, C., Al-Samarrai, S., Rose, P., & Tembon, M. (2003) Achieving Schooling for All in Africa: Costs, Commitments and Gender. Aldershot: Ashgate Press. Colclough, C., & Lewin, K. (1993) Educating All the Children: Strategies for Primary Schooling in the South. London: Clarendon Press. Colclough, C., & Manor, J. (1993) States or Markets? Neo-liberalism and the Development Policy Debate. London: Clarendon Press. Connell, R. (2007) Southern Theory: Social Science and the Global Dynamics of Knowledge. London: Allen and Unwin. Day Ashley, L., McLoughlin, C., Aslam, M., Engel, J., Wales, J., Rawal, S., Batley, R., Kingdon, G., Nicolai, S., & Rose, P. (2014) The Role and Impact of Private Schools in Developing Countries. [Online] London: Department for International Development (DFID). Available from: www.odi.org/sites/odi.org.uk/files/odi-assets/ publications-opinion-files/10145.pdf [Accessed 8 June 2020]. Denzin, N.K., Lincoln, Y.S., & Tuhiwai Smith, L. (2008) Handbook of Critical and Indigenous Methodologies. Los Angeles, CA: Sage. Hickey, S, & Hossain, N. (Eds.) (2019) The Politics of Education in Developing Countries: From Schooling to Learning. Oxford: Oxford University Press. Jeffrey, C. (2010) Timepass: Youth, Class and the Politics of Waiting in India. Stanford, CA: Stanford University Press. Jeffrey, C., Jeffery, P., & Jeffery, R. (2007) Degrees Without Freedom? Education, Masculinities, and Unemployment in North India. Stanford, CA: Stanford University Press. Macpherson, I., Robertson, S., & Walford, G. (Eds.) (2014) Education, Privatisation and Social Justice. Oxford: Symposium. Minow, M. (1990) Making All the Difference: Inclusion, Exclusion and American Law. New York: Cornell University Press. Mitchell, R., Asare, S., & Rose, P. (2020) Education research in sub-Saharan Africa: Quality, visibility, and agendas. Comparative Education Review, 64(3), pp. 363–383. Moon, B. (2013) Teacher Education and the Challenge of Development: A  Global Analysis. London: Routledge. Pritchett, L. (2018) The Politics of Learning: Directions for Future Research. RISE Programme Working Paper No. 18/020. Robinson-Pant, A. (2005) Cross Cultural Perspectives on Educational Research. Maidenhead: McGraw-Hill Education. Robinson-Pant, A., & Singal, N. (2013) Researching ethically across cultures: Issues of knowledge, power and voice. Special Issue. Compare, 43(4), pp. 417–421. Sachs, J.D. (2012) From millennium development goals to sustainable development goals. The Lancet, 379(9832), pp. 2206–2211. Sayed, Y. (Ed.) (2018) Continuing Professional Teacher Development in Sub-Saharan Africa. London: Bloomsbury.

16  Pauline Rose, et al. Schweisfurth, M. (2015) Learner-centred pedagogy: Towards a post-2015 agenda for teaching and learning. International Journal of Educational Development, 40, pp. 259–266. Sen, A. (1999) Development as Freedom. Oxford: Oxford University Press. Srivastava, P.,  & Walford, G. (2018) Non-State Actors in Education in the Global South. London: Routledge. Tuhiwai Smith, L., Tuck, E., & Yang, K.W. (2018) Indigenous and Decolonizing Studies in Education. Mapping the Long View. New York: Taylor and Francis Group. UNESCO (2002) Is the World on Track? Education for All Global Monitoring Report, 2002. [Online] Paris: UNESCO. Available from: https://unesdoc.unesco.org/ ark:/48223/pf0000225660 [Accessed 18 May 2020]. UNESCO (2004) Gender and Education for All – The Leap to Equality? Education for All Global Monitoring Report, 2003–2004. [Online] Paris: UNESCO. Available from: https://en.unesco.org/gem-report/report/2003/gender-and-educationall-leap-equality [Accessed 18 May 2020]. UNESCO (2005) Education for All  – The Quality Imperative. Education for All Global Monitoring Report, 2005. [Online] Paris: UNESCO. Available from: https://en.unesco.org/gem-report/report/2005/education-all-quality-imperative [Accessed 18 May 2020]. UNESCO (2014) Teaching and Learning: Achieving Quality for All. EFA Global Monitoring Report, 2013–2014. [Online] Paris: UNESCO. Available from: https:// unesdoc.unesco.org/ark:/48223/pf0000225660 [Accessed 18 May 2020]. Unterhalter, E., & North, A. (2019) Education, Poverty and Global Goals for Gender Equality: How People Make Policy Happen. London: Routledge. Verger, A., Novelli, M.,  & Kosar-Altinyelken, H. (Eds.) (2012) Global Education Policy and International Development: New Agendas, Issues and Programmes. London: Bloomsbury. Walker, M., & Unterhalter, E. (Eds.) (2007) Amartya Sen’s Capability Approach and Social Justice in Education. London: Palgrave Macmillan. World Bank (2018) World Development Report 2018: Learning to Realize Education’s Promise. Washington, DC: World Bank.

Part I

The economics and politics of educational reform

2 The changing pattern of returns to education What impact will this have on earnings inequality? Harry Anthony Patrinos1 The pattern of economic rates of return to investments in education can help us to understand the benefits of schooling. Previous estimates showed that the private labour market returns to education were highest for the primary level of education and lower for secondary and higher education levels (see, e.g. Psacharopoulos 1981). However, Colclough et al. (2010) showed that this pattern was already changing such that the returns for higher education were then higher than for primary. The causes of such changes, and their implications for both education and labour market policy, are significant. The possible reasons put forward for this latter pattern include: (1) technological change favouring higher-order skills; (2) increasing access at lower levels of schooling and (3) the quality of both primary and secondary schooling worsening as access increased. Of course, it could be a combination of all three factors. The issue of changing patterns of returns and the implications are revisited here. The patterns of the returns to education are important for students and their families as they use this information to make important and costly decisions. Better and more accurate information about future earnings potential significantly affects schooling decisions (Jensen 2010; Hastings et  al. 2015). They are also important signals for policy-makers (Patrinos 2016). For example, evidence on the returns to early childhood investments is used by policy-makers to extend funding for pre-schools and returns to higher education have been cited in efforts to reform student finance systems (Barr & Crawford 2003). Since the 1980s, although the overall returns to schooling increased (Psacharopoulos & Patrinos 2018), recently, the returns to higher education have increased the most (Montenegro & Patrinos 2014). In this chapter I first review current patterns of the returns to schooling by level and region, along with a discussion of the possible causes of changes and the subsequent policy implications. Among other aspects, I assess the education investment patterns over time. The analysis is based on estimates of the returns to schooling from 149 countries and more than 1,000 estimates from 1950 to 2014. These data, disaggregated by gender, come from countries all over the world representing more than 90% of the world’s population. The more educated have significantly increased their earnings relative to the less educated despite an increase in the numbers of people who have completed,

20  Harry Anthony Patrinos for example, secondary schooling. This must mean that the demand for more educated workers has increased more than supply over time, causing an increase in the overall returns to schooling. This has implications for equity. If supply of more educated workers does not catch up with demand soon, then earnings inequality will continue to rise. This puts those with lower levels of schooling at a distinct disadvantage. This chapter is structured as follows. First, I examine the link between education and earnings in depth. This is followed by a focus on the changing patterns of the returns to education before focusing on the causes of these changes. Throughout the chapter, the arguments assume a reasonably competitive labour market in which wages and salaries are set by the balance between supply and demand. The conclusions I reach are valid, despite a variety of barriers to freely competitive labour markets.

The link between education and earnings The origins of scientific interest in the contribution of schooling to development are attributed to Adam Smith, although classical philosophers in the East and the West promulgated education as a characteristic of moral and personal development. In the modern era, education entered economics in the late 1950s (see, e.g. Mincer 1958). The concept of the rate of return on investment in education is like that for any other investment. While there are other motivations for education, the one being considered here is the monetary benefits associated with educational attainment. It is a summary of the costs and benefits of the investment incurred at different points in time, and it is expressed as an annual (percentage) yield, like that quoted for savings accounts or government bonds. Returns on investment in education have been estimated since the late 1950s. Investments in education increase future productivity. Time and money spent on education builds human capital, hence one should be able to estimate the rate of returns on such investment, like the investment in physical capital. Families invest in education with an expectation that the investment will provide a benefit in the form of higher earnings. While human capital is more than education  – it has been defined as ‘the stock of knowledge, habits, social and personality attributes, including creativity, embodied in the ability to perform labour so as to produce economic value’ (Goldin 2018) – nevertheless, most empirical work by economists assessing human capital estimates the impact of schooling on earnings. This is because of data constraints – for example, we do not have comparable productivity measures for most workers in the informal sector. Nevertheless, attempts to include informal sector workers produce similar rates of return patterns (see, e.g. Lassibille & Tan 2005). Human capital theory has been used by economists of education since the 1950s. Human capital refers to the aggregate stock of competencies and knowledge embodied in people which allows them to create economic value. Human capital theory assumes that individuals take actions that will likely increase their future earnings and overall well-being. Such investments are costly: they might

The changing pattern of returns to education 21 involve direct costs such as tuition and fees for school and indirect costs such as foregone earnings during the period spent in school. These investments result in some expected future benefits. The benefits might include a higher wage, but can also be anything that the individual values, for example, better working conditions or a longer life. Human capital theory typically models investment decisions, such as those resulting from an optimisation process: an individual will invest in such activities in order to maximise well-being over the course of a lifetime. Observed outcomes in the marketplace will be the result of an equilibrium process where the demand for specific skills and abilities is balanced with its supply. Similarly, it is also assumed that schooling promotes equity as earnings rise with the level of education. But it is sometimes necessary to incentivise the demand for schooling, especially for students from lower-income backgrounds who might tend to underestimate future returns (Jensen 2010). On average, women obtain less schooling than men. The gap is much smaller in industrial countries, but a sizeable gap remains in most developing countries (Patrinos 2008). Again, incentives can help. There is evidence that financial support can increase girls’ enrolment and help close the gender gap in school enrolment (Patrinos 2008). The rise in earnings inequality experienced during the 1980s and 1990s in many countries led to renewed interest in estimates of returns on educational investment (see, e.g. Levy & Murnane 1992). The literature suggests that systematic changes in the production process brought about by changes in technology and the growth of the knowledge-based economy whereby product cycles become shorter and flexibility is needed, led to changes in the demand for skilled labour, especially in North America (Juhn et al. 1993) and other high-income regions. Estimates of the returns to schooling were used to justify investments in basic education and away from higher levels, away from vocational and higher education towards academic and basic education and to make the case for cost recovery in higher education (Heyneman 2005; Jones 1997). Estimates of the returns to education indicate the profitability of investing in a certain level of education, both from the individual’s (private returns to education) and society’s (social returns to education) point of view. The high private rates of return to primary education, relative to secondary and higher education, have led to calls for increased investment in this level of education. One of the main reasons for this is the assumed positive contribution that expansion of primary education would make towards economic equality. The justification for increasing schooling attainment was the high social benefits of education (Colclough 1982) and the relatively higher social returns to primary education (Psacharopoulos 1985). High returns to higher education, by extension, justify further investment in higher education. But to enrol in university, one needs secondary schooling; and to get secondary schooling, one needs primary. Thus, education is sequential, and higher returns at one level necessitate demand rising for preceding levels as well. Thus, social demand for each level of schooling is driven by returns at higher levels as well. Thus, education has a return, and a further

22  Harry Anthony Patrinos benefit as it provides opportunity for further study – in other words, an option value (Shafiq 2007; Weisbrod 1962). Evidence on the returns to education was therefore used to justify increased spending on education to achieve Education for All, the Millennium Development Goals and the Sustainable Development Goals. At the World Bank, returns to schooling evidence were used to justify shifting investment from vocational secondary schooling towards basic education (Psacharopoulos & Loxley 1985). Public intervention may be needed to promote equity and make sure that those without the financial means enrol in schooling and are able to realise the returns of education. In its report, the Education Commission (International Commission on Financing Global Education Opportunity 2016) says that successful education systems must reach everyone, including the most disadvantaged and marginalised children. The learning gap will not be closed unless additional steps are taken to include and support those at greatest risk of not learning – the poor, the disadvantaged, girls and those facing multiple disadvantages. Because of this, the Commission has called on stakeholders to take up a strategy called ‘progressive universalism’ – pursuing quality education for all while prioritising education in the early years of a child’s life and reaching out to the most disadvantaged populations, where social returns are often the highest (International Commission on Financing Global Education Opportunity 2016). High private returns do not necessarily mean that there should be more social spending. On the contrary, if the returns to schooling are high for post-compulsory schooling then this justifies calls for cost-sharing. For social policy, the priority is to finance basic education and promote equality of opportunity so that the poorest can enrol and experience the returns to schooling. As it now stands, many countries are far from achieving the Sustainable Development Goal of equal access to higher education by 2030. It is necessary to reduce pro-rich education spending  – that is, public spending that benefits the wealthiest segments of the population (Ilie & Rose 2018).

Education and inequality There is an expectation that education contributes to greater equality in society (Becker 1964). Education has a strong effect on individual earnings, so it also affects the distribution of earnings. Historically, the net effect of an expansion of schooling has been a reduction in the dispersion of earnings and, therefore, a more equal distribution. This effect, however, depends on which level of schooling is expanded. The equity impact is highest for primary education, since the low earnings of those with otherwise no schooling are raised. But if university education is expanded, then the equity effect may be negative since the earnings of a group of workers already earning above the mean are raised further. Since most university students come from the higher-income groups in most countries (Hansen & Weisbrod 1969; Ilie & Rose 2018; Jin et al. 2019;

The changing pattern of returns to education 23 Vawda 2003), then state subsidies for their education will boost their future earnings at the expense of others who are less likely to enrol their children in higher education. A large literature examines the benefits of education investments across the income distribution. Overall, public education expenditures are regressive, with a higher share of public spending going to groups from the highest family income categories. However, this has a lot to do with the fact that mostly individuals from high-income families enter university, which is associated with a much higher expenditure per student (Psacharopoulos & Patrinos 2008). The idea that educational policy can be used to eradicate poverty and contribute to equality has been with us for some time (Goldin & Katz 2007). This led to rapid educational expansion during the 1950s and 1960s. Educational investment is a profitable investment for the individual and a good social investment by governments and can contribute to economic equality (see, e.g. Colclough 1982; World Bank 2018). Moreover, educational policy is an exogenous policy instrument that the government can use to achieve such goals as a more equitable distribution of education and a more equitable distribution of income. Increasing an individual’s education makes that person more productive and able to gain a high-paying position, particularly if the quality of education is also high and the labour market is competitive. As the supply of highly paid, skilled labour increases and the supply of unskilled labour decreases, the differential in pay between the two groups of workers will diminish due to the laws of supply and demand. This process eventually leads to a more equitable distribution of income, other things being equal and in the absence of discrimination and other factors that negatively affect the effective functioning of the labour market. However, there are limitations to this. There are still instances where discrimination plays a role, thus limiting the beneficial power of education (Welch 1975). Also, returns to schooling accrue when there is a need for learning – such as in formal labour markets or when there is technological shock that puts a premium on learning, such as the Green Revolution in India (Rosenzweig 1995). According to Human capital theory and as its empirical application, estimates of the returns to education show that the returns to education should decline following an expansion of the supply of schooling (Psacharopoulos 1989). An educational expansion that increases the number of the more educated relative to the less educated in the labour market should cause the wage differential between the two groups to fall. This is because an increase in supply leads to a reduction in price, assuming of course that demand does not rise. This, in turn, will lead to a decline in the returns to education, since the returns, which is the ‘price’ of education from the worker’s perspective, will fall. In other words, earnings differentials are compressed. All this, of course, should lead to a decrease in income inequality, other things being equal. Most research has generally validated this hypothesis (see, e.g. Mincer 1974; Marin & Psacharopoulos 1976; Arshed et al. 2019; Abdullah et al. 2015; Coady & Dizioli 2018).

24  Harry Anthony Patrinos

Returns to investments in education There is a sizeable literature on the returns to education (Psacharopoulos  & Patrinos 2018; Montenegro  & Patrinos 2014; Patrinos 2016). What is more interesting is the variation in returns by level of education – primary, secondary and tertiary education. Economists have been estimating the returns to schooling systematically since the late 1950s. If the private returns to primary schooling were high, then it could be assumed  – if information flows widely and can be acted upon  – that parents would value education and ensure their children were enrolled. Growing enrolments would result in higher productivity of the poor and lead to lower inequality in the future. The empirical evidence from the 1960s to the 1990s suggested that in most countries the relationship between education and earnings was concave (Psacharopoulos 1994; Psacharopoulos & Patrinos 2004). The slope is steep at low levels of education (that is, the return to an additional year of schooling is high) but becomes progressively flatter (that is, the marginal returns fall) at higher levels (Colclough et  al. 2010). If the shape of the education– earnings relationship is concave, then an extra year of education at low levels of education brings substantially greater increases in earnings than it does at higher levels of education. Similarly, marginal increases in education at low levels of education (where the poor are typically concentrated) raise earnings substantially.

Time trends of the returns to education Returns to investments in schooling are estimated by computing the difference in average earnings between those with a given level of schooling and those at the level below it (Mincer 1974). Typically, one estimates an earnings function where the coefficient on education can be interpreted as the private (that is, what the individual expects to earn as a result of their investment) return to an additional year of education (Patrinos 2016). A review of current returns patterns by level and region is presented using two large, recent databases. The analysis is based on estimates of the returns to schooling from 149 countries and more than 1,000 estimates from 1950 to 2014 (the full set of countries can be seen in the source studies, namely: Montenegro  & Patrinos 2014; Psacharopoulos  & Patrinos 2018). The data come from countries all over the world representing more than 90% of the world’s population and are disaggregated by gender. Montenegro and Patrinos (2014) presented comparable estimates of the rate of returns to investment in education in a database that used the same specification, estimation procedure and similar data. They provided comparable estimates because they held constant the definition of the dependent variable, the set of control variables, the sample definition and the estimation method for all surveys in the sample. Psacharopoulos and Patrinos (2018) updated their decennial review of the literature and reviewed the

Years of schooling/returns (%)

The changing pattern of returns to education 25 10 9 9 8 8 7

Pre 2000 Mean years of schooling

Post 2000 Rate of return (%)

Figure 2.1  Years of schooling and returns over time Source: Computed by the author using data from Psacharopoulos and Patrinos (2018)

latest trends and patterns. This allows us to update the findings from Colclough et al. (2010). Montenegro and Patrinos (2014) found not only that there was a decreasing pattern in the private returns to schooling over time across the sample of countries, but also that the returns to tertiary education were the highest. Psacharopoulos and Patrinos (2018) found that the private average global returns to a year of schooling was 9% a year and that the private returns to higher education increased. Both found higher returns for women. The rapid expansion in schooling in terms of access, enrolment and total years of schooling is expected to reduce the earnings advantage, everything else being equal, of course. Average years of schooling worldwide were only one year in 1900, rising to 3.2 by 1950, to more than 8.5 by 2010 (Barro & Lee 2015). However, although the average schooling level of the working population increased in the early 2000s, the overall private rate of return to schooling also increased (Figure 2.1). Before 2000, average years of schooling were 7.8 and the returns to schooling were 8.7%. After 2000, average schooling rose to 8.6 years and the returns edged up to 9.1%. In fact, the returns to schooling were decreasing over time as the level of education rose (Figure 2.2). From 1980 to 2010, the returns to schooling declined significantly while schooling levels increased. While schooling levels increased on average almost 11% every five years, the returns to schooling fell by 6%. But between 2011 and 2013, schooling levels went up by 6%, and the returns to schooling increased by more than 4%. This is based on a significant number of economies (149) undertaken with a consistent methodology (Montenegro  & Patrinos 2014). Between 1970 and 1990, the returns to schooling for both men and women declined. However, in the 2000s the returns started to increase (see Table  2.1). This is significant, because at the same time, the gap in years

26  Harry Anthony Patrinos 14

Returns to schooling (%)

12 10 8 6 4 2 0

1980–1985

1986–1990

1991–1995

1996–2000

Returns to schooling

2001–2005

2006–2010

2011–2013

Average years of schooling

Figure 2.2  Returns to schooling and average years of schooling by period Source: Constructed by the author using data from Montenegro and Patrinos (2014: Annex Table 1, pp. 20–36)

Table 2.1  Returns to schooling over time (%) Decade

Men

Women

1970s 1980s 1990s 2000s

8.5 8.5 7.3 9.3

10.2 9.8 8.8 11.1

Source: Computed by the author using data from Psacharopoulos and Patrinos (2018)

of schooling has narrowed, even though women have less years of schooling. Women have almost quadrupled their average years of schooling (Barro & Lee 2013). Rising or stagnant returns, or even slowly declining returns, while schooling levels are rising, goes against economic theory. The evidence from estimated returns is giving us a hint of possible growing earnings inequality (Figure 2.3). The variance in the returns to schooling is increasing. Just as the returns to schooling are increasing, so is the variance. We see this in the standard deviation of the returns to schooling which is increasing over time. Since the standard deviation is a measure of the dispersion of the returns from their mean, then the higher deviation signals that the returns are becoming more dispersed. In this sense, the returns to schooling are becoming more volatile. The increased inequality coincides with the fact that now the returns to tertiary education are highest. It was not always like that. The empirical evidence from

The changing pattern of returns to education 27

Standard deviation

8

6 4 2 1970

1980

1990 Year

2000

2010

Figure 2.3  Standard deviation of rate of return to year of schooling Source: Computed by the author using data from Montenegro and Patrinos (2014)

the 1960s to the 1990s suggested that in most countries the relationship between education and earnings was concave (Psacharopoulos 1972, 1981, 1985, 1994; Psacharopoulos & Patrinos 2004). That is, the private rate of returns was higher at the primary school level than at secondary or higher level (Psacharopoulos 1981). The same pattern was evident by the early 1980s (Psacharopoulos 1985). Similar evidence was produced for the 1980s and 1990s, but already the pattern was shifting (Colclough et al. 2010). In fact, by the early 1990s, the returns to primary schooling were decreasing, while the returns to higher education were rising (Psacharopoulos 1994). More recent evidence suggests that the rate of returns to primary education is lower than that to higher levels of education. Many studies using cross-sectional data from the 1990s and early 2000s find that the returns to primary education in wage employment is significantly lower than that to higher education (Psacharopoulos & Patrinos 2018). But analyses of current rates and levels may be problematic because of assumptions about foregone earnings, costs, assumptions, comparators – like university versus secondary education. What might be more telling are the trends. This includes the trends on the returns on years of schooling and returns by level. Globally, primary school enrolment has gone up from 72% in 1970 to 90% in 2017 and from 41% to 77% at the secondary level (UNESCO Institute for Statistics database). The returns to schooling follow a predictable pattern at those levels: they decreased, based on evidence from 746 observations between 1970 and 2014 (Montenegro  & Patrinos 2014; Psacharopoulos  & Patrinos 2018) from 116 economies, representing more than 98% of the world’s population. The returns to primary education have decreased by 21% and the returns to secondary have decreased by one-third (Figure 2.4). The returns to schooling have decreased slightly over time (since the 1960s to the early 2000s), but they have remained remarkably consistent and stable.

28  Harry Anthony Patrinos

Ratio of returns

1.20

1.00

0.80

0.60

1980

1990

2000

2014

Year Primary

Secondary

Higher

Figure 2.4  Changes in returns to education since 1980 Source: Computed by the author using data from Montenegro and Patrinos (2014)

In the 1960s and 1970s, the returns to a year of schooling were 13% in Africa, 11% in Asia and 9% in advanced economies (Psacharopoulos 1981). By 2014, the returns to schooling were 10.5% in sub-Saharan Africa, 8.1% in South Asia and 8% in advanced economies (Psacharopoulos & Patrinos 2018). At the same time schooling levels doubled in advanced economies from six to 11 years, more than quadrupled in sub-Saharan Africa from one to five years and quintupled from one to five years in South Asia. The situation is very different at the tertiary education level. Tertiary level enrolments have more than tripled since the 1970s (UIS), from 10% in 1970 to 38% in 2017. At the same time, the returns to tertiary education have increased by almost 3%. In less developed regions, the returns to higher education are high, but so too are the returns to primary schooling, and access to (secondary and) higher education is limited. In these regions, the priority is to expand access to basic and secondary education, so that more people can eventually access tertiary education opportunities. The situation for men and women is similar at the primary and secondary levels. The returns to schooling at the primary and secondary levels for men decreased by one-third (Figure  2.5). The returns to higher education stayed roughly the same between 1980 and 2014. For women, only the returns to primary education decreased by 15%. The returns to higher education increased by 8% (Figure 2.6). Standardised data across countries and over a long period of time reveal a compelling picture of a steep decline in the wage returns to primary schooling across the developing world over the past half-century. The evidence also shows an increasing private rate of returns to higher education, especially for women. While the returns to primary schooling have decreased less for women than for

The changing pattern of returns to education 29

Ratio of returns

1.20

1.00

0.80

0.60

1980

1990 Primary

2000 Secondary

2014 Higher

Figure 2.5  Changes in returns to education since 1980, men Source: Computed by the author using data from Montenegro and Patrinos (2014)

Ratio of returns

1.20

1.00

0.80

0.60

1980

1990 Primary

2000 Secondary

2014 Higher

Figure 2.6  Changes in returns to education since 1980, women Source: Computed by the author using data from Montenegro and Patrinos (2014)

men, the returns to schooling for secondary have decreased much more for women. Since secondary is required for access to higher education, this could be a deterrent to enrolling in secondary schooling or might become one (Kingdon 1998). This is important since investment in schooling of women tends to decrease the proportion of earnings differential with men that is attributable to factors such as discrimination (Dougherty 2005). It is also possible that academically able women are accessing higher education. Since other factors are at play that limit women from fully experiencing access and returns to education,

30  Harry Anthony Patrinos more research is needed into those factors. At the same time, education offers a path to equality. This justifies investment in girls’ schooling and compels us to ensure high levels of schooling through to higher education to reap the significant wage returns. To make it more attractive for girls and their families, then interventions designed to increase access are needed. These could be targeted financial programmes, which have been shown to be effective (Muralidharan & Prakash 2017). Over time comparisons using the same method and the same definitions of education levels across the countries show that, in general, the returns to education increase as the level of education rises in 41 of 49, or 84%, cases in developing economies with over time data. In all cases latest year data shows higher returns to higher education than to secondary schooling. Overall, based on 596 observations or data points, the returns to higher education are greater than the returns to primary education in 83% of cases. The returns to higher education are greater than the returns to secondary education in 97% of cases. And, in terms of latest year data for 104 economies, the returns to higher education are greater than the returns to primary education in 73% of cases, and the returns to higher education are greater than the returns to secondary education in 95% of cases. However, the returns to secondary schooling are greater than the returns to primary schooling only 33% of the time. While in some countries the pattern of increase is not monotonic with the level of education a generally mixed pattern of returns is observed across most countries (Table 2.2). For females, the differences in returns are greatest at the secondary level, suggesting that this level of schooling is paramount for generating gender equity in the future. The very sharp declines, and low overall rates, for secondary education are a bit misleading. There is an option value to completing secondary education – and primary as well. The option value is the benefit of obtaining further education and the rewards accompanying it (Weisbrod 1962). It is distinct from the conventionally accepted wage gain. In the case of secondary education, there is an option value arising from the fact that completion of secondary education provides a benefit other than wage returns: the option to invest in post-secondary levels of education, which is not available to those who do not complete secondary education. Estimates from Bangladesh suggest significant returns to the option value (Shafiq 2007). Table 2.2  Education–earnings relationship (% of countries) Comparison of returns by level

Higher > Primary Higher > Secondary Secondary > Primary

Incidence of returns being greater (%) Total

Male

Female

73.3 95.2 32.7

82.6 95.7 29.3

72.2 93.8 38.1

Source: Computed by the author with data from Montenegro and Patrinos (2014)

The changing pattern of returns to education 31

Causes of the changed pattern The returns to schooling have not declined very much over time, while the supply of schooling has increased. This suggests that the supply of schooling is not keeping up with demand. There could be several reasons for this. We explore three possible explanations: technological change favouring higher-order skills; increasing coverage at lower levels of schooling and the quality of schooling. The higher returns to tertiary education seem to suggest that the education supply curve has been shifting more to the right, relative to the demand curve (Tinbergen 1975; Psacharopoulos 1989; Psacharopoulos & Patrinos 2018; Goldin & Katz 2009). That is, employers are demanding workers with higher levels of education. This is most evident in the returns to higher education. While enrolment in higher education has gone up threefold since 1970, the returns have not changed overall and have increased for women. The increased share of the labour force with higher education should have reduced the rate of returns on the investment. Yet, the rates of returns over time do not fluctuate much because as the supply of educated labour increases, so does the demand for higher skills, hence, not depressing the returns to education. As discussed by Goldin and Katz (2009), skill-biased technological progress is producing consequences for income inequality. Increased global trade and open economies is leading to changes in skilled/unskilled wage ratios (Colclough et al. 2010). But increased openness has not reduced skill differentials but widened them, with increased global demand for higher skills (Colclough et al. 2010; Patrinos et al. 2018). Education systems do not seem to be able to mediate this relation. While higher education has increased substantially, the premium on high skills has continued to increase. This suggests that educational advancements were insufficient to countervail demand due to technological progress. The evidence on the skill-biased technological change came from more developed countries (see, e.g. Acemoglu 2002). Yet we have seen that recent evidence on returns to education suggests that similar tendencies exist in a good number of low-income as well as middle-income countries (Himaz  & Aturupane 2011; Ogundari  & Aromolaran 2014; Ackah et  al. 2014). That we are seeing the same thing happen in developing countries such as Ghana, Nigeria and Sri Lanka is interesting. That is, the returns to higher education are increasing in countries which have not even achieved universal primary education. But what is different in developing countries is the low quality of education (World Bank 2018). In principle, the relative decline in the wage returns to primary education over time may be due to both supply-side and demand-side factors (Colclough et al. 2010). The main causes are likely to be that primary schooling is now more likely to be a means to getting to secondary rather than an end in its own right. The last few decades have seen significantly increasing coverage at lower levels of schooling. The supply of primary and secondary school completers has greatly increased over the past 15 years. The out-of-school secondary school rate also fell rapidly during this time. As education systems expand, the qualifications required for jobs rise (Colclough et  al. 2010). The reduced access to jobs provided by

32  Harry Anthony Patrinos primary and secondary schooling will be associated with downward pressure on wage returns at these levels. The quality of schooling may be suffering because of the rapid expansion of school systems in developing countries. The expansion of primary school enrolments seems to be accompanied by lower test scores (Angrist et al. 2019), either by selectivity as poor and less well-prepared students or unprepared teaching staff enter the system. Recent students may be less well-prepared than earlier cohorts (Colclough et al. 2010). This would help explain the decrease in the returns to primary schooling. Educational expansion can reduce income inequality but only if the quality of schooling is improving (Coady & Dizioli 2018).

Conclusion The fact that returns to education have improved, in particular for the more educated despite an increase in their numbers, must mean that the demand for more educated workers has increased more than supply over time, causing an increase in the overall returns to schooling. While schooling and earnings inequality are related, the returns to schooling may increase despite an increase in the average level of schooling (or a decrease in the variance of schooling) if the demand for schooling has also increased. When there are no changes in the demand for educated labour, an increase in the supply of educated workers leads to a decrease in the earnings premium they have over less educated workers and the returns to schooling decrease. The technological revolution and prospects of automation and artificial intelligence put added strain on skills (Frey & Osborne 2017). For most developing countries, the message is to invest early in basic cognitive skills. In situations of rapid technological change, it becomes more difficult to predict job changes and therefore schooling policy. If schools are unable to meet the demand for skills in the future, then the premium on high-order skills and competencies will rise, further raising the returns to schooling and increasing income inequality (Tinbergen 1975; Goldin & Katz 2009). When technological or other developments occur that lead to an increase in the demand for educated labour substantially large enough to overcome increases in the supply of educated labour, however, then the returns to schooling will increase (Psacharopoulos 1989). The continued decline in the relative returns to primary schooling could be expected to cause the demand for primary education to fall. But since we are trending towards universal primary schooling in most parts of the world, the demand for primary schooling may be unlikely to fall. But given the problems of quality and low completion, is it the case that the issue of public investment in primary remains a priority? Given the sequential nature of the schooling system, one needs to invest in primary before secondary and secondary before higher education. Therefore, high enrolment and completion rates will drive the demand for secondary school places. However, the relatively lower returns to secondary puts further gains in doubt. The option value of secondary schooling is likely to be positive in terms of demand. The fact that the returns to higher education have been increasing gives one hope.

The changing pattern of returns to education 33 But the demand for post-primary schooling also depends on affordability. Secondary education is more expensive than primary education and in shorter supply. Further expansion will strain government coffers or lead to expansion without quality. The extent to which primary school returns have declined due to falls in quality would require further attention. Decreasing quality affects not only the labour market opportunities of those with only primary schooling but limits their ability to access high-quality secondary school opportunities (see also Zubairi & Rose 2019). There is a need to expand the education system at secondary and higher levels. If not, then income inequality will rise. But to effectively expand coverage and reduce the inequality tendencies, the quality of primary education must rise, too. Thus, the priority for governments is to improve quality and expand secondary education. The high and rising returns justify cost recovery at the higher education level. However, to avoid growing inequity, then innovative forms of financing are needed to allow worthy but credit-constrained students to attend. The latter should be taken up by donor institutions and the private sector.

Note 1 Practice Manager, Education, World Bank. [email protected]. Views expressed here are those of the author and should not be attributed to the World Bank Group. This chapter is dedicated to the memory of Christopher Colclough, supervisor, mentor, friend and great development economist. I am forever grateful for the opportunity he gave me. I thank the editors of this volume for comments on earlier drafts of the chapter.

References Abdullah, A., Doucouliagos, H.,  & Manning, E. (2015) Does education reduce income inequality? A meta-regression analysis. Journal of Economic Surveys, 29(2), pp. 301–316. Acemoglu, D. (2002) Technical change, inequality, and the labor market. Journal of Economic Literature, 40(1), pp. 7–72. Ackah, C., Acquah, A., Turkson, F., & Adjasi, C. (2014) Education, Skill, and Earnings: Further Evidence from Ghana. WIDER Working Paper No. 2014/073. Angrist, N., Djankov, S., Goldberg, P.K., & Patrinos, H.A. (2019) Measuring Human Capital. World Bank Policy Research Working Paper No. 8742. Arshed, N., Anwar, A., Hassan, M.S., & Bukhari, S. (2019) Education stock and its implication for income inequality: The case of Asian economies. Review of Development Economics, 23(2), pp. 1050–1066. Barr, N., & Crawford I. (2003) Myth or magic? The Guardian, December 1, p. 20. Barro, R.J., & Lee, J.W. (2013) A new data set of educational attainment in the world, 1950–2010. Journal of Development Economics, 104(C), pp. 184–198. Barro, R.J., & Lee, J.W. (2015) Education Matters. Global Schooling Gains from the 19th to the 21st Century. Oxford: Oxford University Press. Becker, G.S. (1964) Human Capital Theory. New York: Columbia University Press. Coady, D., & Dizioli, A. (2018) Income inequality and education revisited: Persistence, endogeneity and heterogeneity. Applied Economics, 50(25), pp. 2747–2761.

34  Harry Anthony Patrinos Colclough, C. (1982) The impact of primary schooling on economic development: A review of the evidence. World Development, 10(3), pp. 167–185. Colclough, C., Kingdon, G., & Patrinos, H. (2010) The changing pattern of wage returns to education and its implications. Development Policy Review, 28(6), pp. 733–747. Dougherty, C. (2005) Why are the returns to schooling higher for women than for men? The Journal of Human Resources, 40(4), pp. 969–988. Frey, C.B., & Osborne, M.A. (2017) The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, pp. 254–280. Goldin, C. (2018) Human capital. In Diebolt, C., & Haupert, M. (Eds.) Handbook of Cliometrics. New York: Springer, pp. 55–86. Goldin, C.,  & Katz, L.F. (2007) The Race between Education and Technology: The Evolution of US Educational Wage Differentials, 1890 to 2005. National Bureau of Economic Research Paper No. 12984. Goldin, C., & Katz, L.F. (2009) The Race between Education and Technology. Cambridge: Harvard University Press. Hansen, W.L, & Weisbrod, B.A. (1969) The distribution of costs and direct benefits of public higher education: The case of California. Journal of Human Resources, 4, pp. 176–191. Hastings, J., Neilson, C.A., & Zimmerman, S.D. (2015) The Effects of Earnings Disclosure on College Enrolment Decisions. National Bureau of Economic Research Paper No. 21300. Heyneman, S.P. (2005) The history and problems in the making of education policy at the World Bank, 1960–2000. In Baker, D.P., & Wiseman, A.W. (Eds.) Global Trends in Educational Policy (International Perspectives on Education and Society, Vol. 6. Bingley: Emerald Group Publishing Limited, pp. 23–58. Himaz, R., & Aturupane, H. (2011) Education and Household Welfare in Sri Lanka from 1985 to 2006. Economics Series Working Papers, 527. Ilie, S., & Rose, P. (2018) Who benefits from public spending on higher education in South Asia and sub-Saharan Africa? Compare: A  Journal of Comparative and International Education, 48(4), pp. 630–647. International Commission on Financing Global Education Opportunity (2016) The Learning Generation: Investing in Education for a Changing World. [Online] The Education Commission. Available from: https://report.educationcommission. org/downloads/ [Accessed 20 May 2020]. Jensen, R. (2010) The (perceived) returns to education and the demand for schooling. Quarterly Journal of Economics, 125(2), pp. 515–548. Jin, H., Jirasavetakul, L.B.F., & Shang, B. (2019) Improving the Efficiency and Equity of Public Education Spending: The Case of Moldova. International Monetary Fund. Working Paper No. 19/42. Jones, P.W. (1997) On World Bank Education Financing. Comparative Education, 33(1), pp. 117–130. Juhn, C., Murphy, K.M., & Pierce, B. (1993) Wage inequality and the rise in returns to skill. Journal of Political Economy, 101(3), pp. 410–442. Kingdon, G.G. (1998) Does the labour market explain lower female schooling in India? The Journal of Development Studies, 35(1), pp. 39–65. Lassibille, G.,  & Tan, J.P. (2005) The returns to education in Rwanda. Journal of African Economies, 14(1), pp. 92–116.

The changing pattern of returns to education 35 Levy, F., & Murnane, R.J. (1992) US earnings levels and earnings inequality: A review of recent trends and proposed explanations. Journal of Economic Literature, 30(3), pp. 1333–1381. Marin, A.,  & Psacharopoulos, G. (1976) Schooling and income distribution. The Review of Economics and Statistics, 58(3), pp. 332–338. Mincer, J. (1958) Investment in human capital and personal income distribution. Journal of Political Economy, 66(4), pp. 281–302. Mincer, J. (1974) Schooling, Experience and Earnings. Cambridge, MA: National Bureau of Economic Research. Montenegro, C.E.,  & Patrinos, H.A. (2014) Comparable Estimates of Returns to Schooling around the World. World Bank Policy Research Working Paper No. 7020. Muralidharan, K.,  & Prakash, N. (2017) Cycling to school: Increasing secondary school enrollment for girls in India. American Economic Journal: Applied Economics, 9(3), pp. 321–350. Ogundari, K., & Aromolaran, A.B. (2014) Impact of education on household welfare in Nigeria. International Economic Journal, 28(2), pp. 345–364. Patrinos, H. (2008) Returns to education: The gender perspective. In Tembon, M., & Fort, L. (Eds.) Girls’ Education in the 21st Century: Gender Equality, Empowerment and Economic Growth. Washington, DC: World Bank, pp. 53–661. Patrinos, H. (2016) Estimating the return to schooling using the Mincer equation. IZA World of Labor 2016, p. 278. doi: 10.15185/izawol.278. Patrinos, H., Thang, P.V., & Thanh, N.D. (2018) The Economic Case for Education in Vietnam. World Bank Policy Research Working Paper No. 8679. Psacharopoulos, G. (1972) Rates of return to investment in education around the world. Comparative Education Review, 16(1), pp. 281–302. Psacharopoulos, G. (1981) Returns to education: An updated international comparison. Comparative Education, 17(3), pp. 321–341. Psacharopoulos, G. (1985) Returns to education: A further international update and implications. Journal of Human Resources, 20(4), pp. 583–604. Psacharopoulos, G. (1989) Time trends of the returns to education: Cross-national evidence. Economics of Education Review, 8(3), pp. 225–231. Psacharopoulos, G. (1994) Returns to investment in education: A  global update. World Development, 22(9), pp. 1325–1343. Psacharopoulos, G., & Loxley, W. (1985) Diversified Secondary Education and Development: Evidence from Colombia and Tanzania. Baltimore, MD: The Johns Hopkins University Press. Psacharopoulos, G.,  & Patrinos, H.A. (2004) Returns to investment in education: A further update. Education Economics, 12(2), pp. 111–134. Psacharopoulos, G., & Patrinos, H.A. (2008) Education and human capital. In Dutt, A.K., & Ros, J. (Eds.) International Handbook of Development Economics, Vol. 1. Cheltenham: Edward Elgar Publishing. Psacharopoulos, G.,  & Patrinos, H.A. (2018) Returns to investment in education: A  decennial review of the global literature. Education Economics, 26(5), pp. 445–458. Rosenzweig, M.R. (1995) Why are there returns to schooling? American Economic Review, 85(2), pp. 153–158. Shafiq, M.N. (2007) Household rates of return to education in rural Bangladesh: Accounting for direct costs, child labour, and option value. Education Economics, 15(3), pp. 343–358.

36  Harry Anthony Patrinos Tinbergen, J. (1975) Income Distribution: Analysis and Policies. Amsterdam: Elsevier. Vawda, A.Y. (2003) Who benefits from public education expenditures? Economic Affairs, 23(1), pp. 40–43. Weisbrod, B. (1962) Education and investment in human capital. Journal of Political Economy, 70(5), pp. 106–123. Welch, F. (1975) Human capital theory: Education, discrimination, and life cycles. American Economic Review, 65(2), pp. 63–73. World Bank (2018) World Development Report: Education. Washington, DC: World Bank. Zubairi, A., & Rose, P. (2019) Equitable Financing of Secondary Education in subSaharan Africa. [Online] Background paper for the Mastercard Foundation. Available from: https://mastercardfdn.org/wp-content/uploads/2019/03/SEAFinance-Equity_REAL_Final-Version_Feb-2019.pdf [Accessed 21 May 2020].

3 Unequal access to education Accounting for change and counting costs Keith M. Lewin

Access to education is both part of the definition of development and a means to achieve it. It lies at the heart of inequalities and is a central vector for social mobility out of poverty. Article 26 of the UN Charter of Human Rights (United Nations 1948) asserts that everyone has the right to basic education that is free and that education at all levels should be provided equitably to promote the full development of the human personality and respect for human rights and fundamental freedoms. Despite all of the efforts to realise rights to education in poor countries since the 1970s the problems of how to finance equitable and inclusive education systems have persisted. Aid to education has made a difference but can no longer be seen as a panacea. New approaches are needed that lead to sustainable financing of universal access to education. Aid needs to be catalytic not gap filling, focussed on sustainable increases in efficiency and effectiveness and directed towards fiscal reforms that are the only pathway to equitable and sustainable financing that is more equitable. The challenges of mobilising resources to universalise access to education globally were addressed systematically at the World Conference on Education for All (WCEFA) (UNESCO 1990a). The analysis provided the first global estimates of the costs and the reform packages that would be necessary to make Education for All (EFA) affordable and more equitable (Colclough & Lewin 1990; ­UNESCO 1990b).1 WCEFA, and the World Education Forum (WEF) (UNESCO 2000) ten years later in Dakar in 2000, reaffirmed beliefs that knowledge and skill translate into the capabilities that underpin development. It is the distribution of access to opportunities to learn that shapes who gets what? in the competition for valued lifestyles, jobs, livelihoods and well-being (Lewin & Sabates 2012). It is the utility of educational outcomes that determines whether those who acquire knowledge and skills translate these into development and share the social and economic benefits. Common threads in the last four decades of education and development recognise that: (1) greater equity and enhanced learning outcomes have always been part of any useful definition of educational development (Colclough 1977; Lewin 1985); (2) education is an investment in human capital without which development would not take place (Colclough  & Manor 1993; Lewin 2000);

38  Keith M. Lewin (3)  mass education systems are essential publicly available merit goods (Colclough 1997; Lewin 2007) and (4) the neoliberal prescriptions of privatisation fail tests of non-rivalrous and non-exclusive access and equity essential to public goods (Lewin 1994; Colclough 1997). This chapter first details transitions in demography, patterns of enrolment, gendered participation and the numbers of out-of-school children that will shape the future development of education in low-income countries. The second section explores how aid to education is evolving and discusses how gaps in educational financing have changed. It identifies the opportunities that exist to focus on aid that catalyses sustainable reform, which increases efficiency and effectiveness and encourages the development of fiscal states that can finance education systems from domestic revenue. Concluding remarks suggest ways forward.2

The changing landscape of education in low- and lower middle-income countries Since 1990 the topography of educational development has changed with many implications for access, participation, equity and financing. Some of the most significant transitions are discussed in the following pages.

Demographic transitions The demographic transition rewrites the map for planning and financing education. It changes the ratio of adults to children and taxpayers to dependents. It can transform the basic arithmetic of youth unemployment. Transition has already occurred in most of East Asia and in China. It is well underway in most of South and South East Asia and in South America. In India about half the states are already seeing a fall in the number of school-age children by 20% or more. Most of the remaining states will experience demographic transition after 2020 (Lewin et al. 2015a, p. 23). In sub-Saharan Africa, transition is occurring more slowly, but the direction of travel is the same (Canning et al. 2015). The implications for educational planning and financing are extensive since the most fundamental driver of costs is the size of the school-age population. Falling enrolments mean there will be more workers per child and, ceteris paribus, more tax revenue per child to translate into educational investment. Increases in the resources available per child should make it easier to reduce inequalities in access and educational quality if the political will exists to redirect resources to those with least financial and cultural capital and make school attendance less burdensome to poor households. In the last three decades of the twentieth century the challenge of expanding participation in low- and lowermiddle-income countries (LICs and LMICs)3 was made difficult by the need to keep pace with high population growth that meant having to double the number of schools, teachers and learning materials every 20 years or so. This is no longer true especially in the LMICs.

Unequal access to education 39 In sub-Saharan Africa, the benefits of demographic transition are appearing more slowly than in other parts of the world. From 1990 to 2015, the number of primary school-age children in sub-Saharan Africa grew by about 2.5% per year. There were twice as many children at the end of the period as at the beginning. Over the same period, the average Gross Enrolment Rate in primary schools increased from 74% to 99%. Enrolments in primary school increased at an annual rate of over 4% from about 60 million to over 150 million. More recently, child population growth rates in sub-Saharan Africa have slowed, averaging about 2.1% for the LICs and 1.4% for LMICs (UNESCO Institute for Statistics 2018). The prospect is one of much slower growth in enrolments at primary level between 1% and 2% per year. The biggest long-term challenge is one that was anticipated in 1990: that of managing progress towards full enrolment at secondary level (Colclough & Lewin 1990). Many LICs are far from full secondary enrolment and most have historic structures of costs that make universal access unavailable without reforms (Lewin & Caillods 2000; Verspoor 2008).

Children in school A second important set of transitions since the 1990s is that education systems in LICs and LMICs have evolved into five characteristic types determined by patterns of enrolment, dropout and completion (Lewin 2008). Data on enrolments from more than 60 LICs and LMICs have been charted to show patterns of enrolment from grades 1–12 and how these have changed over time. An Enrolment Index that compares those enrolled with the number in the child population in the relevant age group can be used to explore how countries differ in patterns of enrolment. There are five patterns that are defined by the shape of the enrolment curve. These are: (1) convex; (2) highly convex; (3) linear attrition; (4) concave and, (5) linear full. Figure 3.1 shows the patterns aggregated from the country data. Countries falling into each pattern are listed in Table 3.1. More detailed analysis is available in Lewin (2017). •

•

Type 1 countries have convex enrolment curves through to grade 12. Intake levels into grade 1 are similar to the number of children in the entry age group. The participation index (number enrolled/number in age group for grade) is close to 1 for grade 1. The tipping point, where there are as many children in the age group as are enrolled in school, is in grade 1 or grade 2. Dropout starts in grade 1 and results in fewer than 50% completing grade 6. Completion rates may be below 40% at primary school level and are less than 20% for lower secondary. Development at secondary level is strongly constrained by the output from primary level. Type 2 countries have very convex enrolment curves with high rates of overenrolment in the early grades of primary. Tipping points are typically around grade 3. Enrolment in grade 1 may exceed 200% of the number of children in the age group. High dropout means that less than 70% of the age group complete grade 6 and less than 50% reach grade 9. Over-enrolment arises from

40  Keith M. Lewin

•

•

•

many children entering who are overage and from high rates of repetition. In some countries this pattern has persisted for more than two decades. The implication is that one equilibrium with low enrolment, low drop out and low completion (Type 1) has been replaced by another with a very high intake, high enrolments and a higher rate of drop out leading to low completion rates. Type 3 countries have enrolments that decline linearly with increasing grade, and the tipping point is around grade 4. It includes countries where the intake rate to grade 1 is high but is less than 50% greater than the number of six-yearolds. No more than 75% of children in an age group reach the end of primary school. There may be serious issues with overage children and repetition and with persistent dropout such that fewer than 50% completes lower secondary school. Primary completion rates constrain expansion of secondary school. Type 4 countries have concave enrolments and include countries that are close to achieving universal completion of grade 6 but have less than 50% completing grade 9. Tipping points are around grade 6 or higher. These countries are more likely to have regularised intake into grade 1 so that all children are within a year of the appropriate age. Most of those who start primary finish at the right age. The biggest attrition occurs in lower secondary and less than half of all children succeed in entering upper secondary. Type 5 countries have full enrolment with similar numbers of children enrolled in each grade as there are in the relevant age cohort. Enrolment curves are linear and track the population growth of single age cohorts of children. There is no tipping point. These systems have achieved universal enrolment up to the end of lower secondary. 1.Convex Low Enrolment and High Drop Out

2.5 Lower Secondary

Primary

Upper Secondary

2

2.Highly Convex Over Enrolment in Grade 1 and High Drop Out.

Enrolment Index

2 1.5

3

3.Linear Decline Middle Level Enrolment and Drop Out

4 5

1

4.Concave Middle Level Enrolment and Low Drop Out

1 0.5

0

1

2

3

4

5

6 7 Grade

8

9

10

11

Figure 3.1  Types of enrolment by grade in LICs and LMICs Source: Lewin (2017)

12

5.Full Enrolment High Enrolment at all Levels and Low Drop Out

Unequal access to education 41 LICs are concentrated in Types 1, 2 and 3. LMICs are predominantly Type 4 and Type 5 systems. Time series analysis of enrolments since 2000 indicates how the direction of travel of systems varies, as reported elsewhere (Lewin 2017). In general, many Type 1 LMICs will graduate to become Type 2 or Type 3 within the next decade. It is probable that Type 1 LICs will become Type 2 systems, and Type 2 systems become Type 3. Inequalities in completion rates at secondary level associated with household poverty are likely to persist for decades in all except Type 5 countries if the future mirrors the past. Countries within each type have issues with educational quality, achievement and equity which go well beyond what is illustrated by enrolment data alone. Inequalities are greatest and most persistent in the poorest LICs.

Table 3.1  LICs and LMICs classified by enrolment types Type

LICs

LMICs

Comment

Cote D’Ivoire, Mauritania, Pakistan, Senegal

Intake rate and enrolment to grade 1 low and likely to include overage children; low primary completion rates and very low lower secondary completion; progression strongly associated with household wealth.

1 Convex: Low Enrolment High Drop Out

Burkina Faso, Eritrea, Gambia, Guinea, Haiti, Liberia, Mali, Niger, South Sudan, Sierra Leone

2 Highly Convex: Over- Enrolment in Grade 1 and High Drop Out

Benin, Burundi, Cameroon Chad, Central African Republic, Comoros, Congo, Democratic Republic of Congo, Ethiopia, Madagascar, Malawi, Mozambique, Rwanda, Timor Leste, Togo, Uganda,

Intake and enrolment to grade 1 very high with double the number of children in lower grades than in the age group; high dropout with less than 75% completing primary; less than 50% completing lower secondary; progression strongly associated with household wealth.

3 Linear Attrition: Middle-Level Enrolment and Drop Out

Afghanistan, Bangladesh, Cambodia, Lao People’s Democratic Republic, Myanmar, Nepal,

Intake and enrolment to grade 1 up to 40% more than in the age group; most but not all complete primary but less than 50% reach the end of lower secondary; children from richer households survive longer.

Nigeria, Yemen

(Continued)

42

Keith M. Lewin

Table 3.1 (Continued) Type

LICs

4 Tanzania Concave: Middle-Level Enrolment and Low Drop Out

5 Linear Full: High Enrolment and Low Drop Out

Tajikistan

LMICs

Comment

Bhutan, Ghana, Kenya, Honduras, Lesotho, Nicaragua, Sao Tome and Principe, Tanzania, Vietnam, Zambia, Zimbabwe

Intake and enrolment rates in grade 1 up to 10% more than in the age group; low dropout through primary with high completion rates; dropout accelerates through lower to upper secondary; children from richer households survive longer.

Albania, Georgia, Kyrgyzstan, Moldova, Mongolia, Uzbekistan,

Full intake and enrolment in primary grades through to grade 9 with little dropout.

Source: Lewin (2017)

Wherever there is significant dropout, there will be inequalities of attainment since not all students will complete a cycle of schooling. Dropout is strongly related to household income with the poorest having much less chance of completing primary or secondary school in LICs and LMICs. The most significant correlate of exclusion across LICs and LMICs is household wealth. Much less of the variance is contributed by location and gender. This is evident from the World Inequality Database on Education and from many analyses in India and elsewhere (e.g. UNICEF 2015; Lewin et al. 2015b; Lewin 2017).

Patterns of participation and gender The third transition is to be found in convergence towards gender parity in primary schooling. In sub-Saharan Africa this gathered pace between 1980 and 1995 when 28 out of 39 countries saw improvements in the ratio of female-tomale enrolments. In 15 cases, gross enrolment rates increased, but in 13 cases the gain took place when gross enrolment rates were falling at the same time as economic recession was widespread. This suggests enrolment of boys fell faster than girls in these countries. In only four cases did the ratio of girls to boys deteriorate (Colclough et al. 2003, p. 30). Economic recovery in the 1990s was accompanied by increasing gender equality. On average, in 1990 the Gender Parity Index for all developing countries4 for primary enrolment was 0.86, and for sub-Saharan Africa it was 0.79. By 2015, gender parity was close to being achieved on average, reaching 0.99 for all

Unequal access to education 43 developing countries and 0.94 for sub-Saharan Africa. At the secondary level, the average Gender Parity Index had reached 0.96 globally and 0.88 in sub-Saharan Africa. In all regions, girls out-enrolled boys at tertiary level except in South Asia and sub-Saharan Africa. The Gender Parity Index by country (as found in UNESCO 2017a) identifies the differences between countries that are concealed by averages. A key issue is that the exclusion of boys has become much more apparent in some countries, especially amongst older age groups at higher educational levels (UNESCO 2018b). Few would have predicted that by 2015 there would be an average of 130 girls to 100 boys enrolled in higher education in Europe, North and South America and the Caribbean. This suggests boys suffer from growing kinds of social exclusion that may differ from those affecting girls, for example, shifting labour market preferences for girls in the service sector, increasing underachievement of boys, male bullying and violent, non-aspirational male youth culture (e.g. Miller 1991). In most LICs and LMICs differences in attainment between girls and boys are greatest amongst the poor and gender differences in enrolment rates are also greatest amongst the poor (UNICEF 2015). Patterns of enrolment of girls and boys have been synthesised from 60 LICs and LMICs into Figure 3.2 to profile characteristic participation by grade (Lewin 2017). Figure 3.2 describes four different patterns of gendered exclusion in LICs and LMICs. These can be described as (1) strong exclusion of girls in all grades; (2)

60% 55%

Percentage Girls

50%

Pattern 4

45% 40%

Pattern 3

35% Pattern 2

30% 25% 20%

Pattern 1

G1

G2

G3

G4

G5

G6

G7

G8

G9

G10

G11

Figure 3.2  LICs and LMICs classified by percentage of girls enrolled by grade Source: Lewin (2017)

G12

44  Keith M. Lewin weak exclusion of girls in primary, strong exclusion at secondary; (3) near equity in primary and weak exclusion of girls at secondary and, (4) gender equity or enrolment of more girls than boys in most grades. Table  3.2 indicates which countries fit the patterns described in Figure 3.2. • • • •

Pattern 1 shows high inequality with large differences in enrolment in favour of boys at all levels. These countries have low overall level of participation. Pattern 2 countries have middle levels of inequality in enrolments and have 45% or fewer girls through primary grades. Pattern 3 countries have low levels of inequality up to the end of primary. At secondary level girls’ participation begins to fall off. Pattern 4 reflects equal enrolments with participation of girls and boys within the 48–52% range. There is a tendency for more girls than boys to be enrolled in higher grades.

Table 3.2  LICs and LMICs classified by patterns of participation by gender Pattern

LICs

LMICs

Comment

Pattern 1 High Inequality

Afghanistan, Central African Republic, Chad

Pattern 2 Middle Inequality

Benin, Democratic Republic of Congo, Eritrea, Guinea, Mali, Niger, Togo,

Cote d’Ivoire, Pakistan, Yemen

45–47% girls in grade 1 falling to below 45% by grade 6 and below 40% by grade 9.

Pattern 3 Low Inequality

Burkina Faso, Burundi, Congo, Guinea Bissau, Ethiopia, Liberia, Mozambique, Sierra Leone, Tanzania

Cameroon, Lao People’s Democratic Republic, Nigeria

47–50% girls in grade 1 with at least 45% up to grade 6. Grade 9 averages about 45%.

Pattern 4 Equal Enrolment

Bangladesh, Bhutan, Cambodia, Gambia, Myanmar, Madagascar, Malawi, Nepal, Rwanda, Sao Tome and Principe, Timor Leste, Uganda, Vietnam

Ghana, Guyana, Honduras, Kenya, Lesotho, Nicaragua, Senegal, Vietnam, Zambia

Average of 49% girls in grade 1 and 50% in grade 6 and grade 9; more girls than boys in high enrolment countries; girls increase with grade level.

Source: Lewin (2017)

40–45% girls in grade 1 falling to less than 35% by grade 9.

Unequal access to education 45 In Pattern 1, boys outnumber girls in all grades. However, about 80% of girls and boys have a similar enrolment status to each other at primary level. Only 5% of LICs and LMICs are in Type 1. They are all conflict-afflicted and very low-income countries, for example, Afghanistan, Central African Republic and Chad (Figure  3.4). In Pattern 2, about 90% girls and boys have the same enrolment status at primary level. Unequal enrolment between boys and girls is concentrated amongst the 10% of children who have different enrolment status, suggesting sharply targeted interventions are most likely to have an impact on the differences. In Patterns 3 and 4 well over 90% of girls and boys have the same enrolment status. This does not mean that gender equity is achieved. Critical indicators other than enrolment and completion rates are needed to identify, monitor and reduce forms of gendered preference and differential exclusions of girls or boys and gain insight into interactive effects. Analysis of the datasets indicates that in LICs and LMICs gendered enrolment patterns tend to diminish as overall enrolment rates increase and Patterns 3 and 4 become the most common. About 45% of the countries are Type 4 and 29% Type 3. Higher enrolment countries have less gendered enrolment patterns. This is empirically true for LICs and LMICs. Since the possible range of gendered difference reduces as enrolments approach universal levels this is partly a statistical inevitability. Gender differences in enrolments are typically larger at secondary than primary school level in low enrolment countries (Lewin 2009). LMICs are more likely to have parity or more girls enrolled at secondary level than LICs (Lewin 2017). In most countries in sub-Saharan Africa, girls tend to enrol at a younger age and leave school earlier than boys, who repeat more often and remain until greater ages. Where gross enrolment rates at secondary are above 50% girls tend to out-enrol boys almost invariably (Lewin 2009). Time sequence data shows that most LICs and LMICs have made substantial progress towards gender equity and 75% of LICs and LMICs are now either Type 3 or Type 4. Strikingly, data on wealth inequalities shows much less progress towards equity in the chances of enrolment at different grade levels and less consistency in the direction of travel, than do gender differences (Lewin 2017; World Inequality Database on Education 2017). The five patterns of enrolment and the four patterns of gendered participation interact in complex ways. The distribution of education systems between LICs and LMICs is also uneven. So is the direction of travel for each system since some are evolving rapidly and others are experiencing stasis. Analysing the patterns can identify bottlenecks and zones of exclusion and locate viable pathways forward using medium- to long-term planning methods. Evidence-based and demand-led planning must complement supply-led educational reforms. They also need to place target setters and target getters in the same room (Lewin 2015). The taxonomies presented earlier provide a starting point.

46  Keith M. Lewin

Children out-of-school The fourth transition relates to the number of children thought to be out of school. This has fallen dramatically. In the late 1980s we estimated that about 130  million children of primary school age were not enrolled (Colclough  & Lewin 1990). By the time of WEF in 2000 the number had fallen to about 94 million (UNESCO 2000). When the Incheon World Education Forum convened the number had fallen to about 60  million (UNESCO 2015b). Half of the 60  million out-of-school primary-age children are now in sub-Saharan Africa compared to only 40% in 1990 indicating that sub-Saharan Africa is lagging behind in the transition to full enrolment. Over 20  million of these are located in just six countries: Ethiopia, Mali, Niger, Nigeria, South Sudan and Tanzania (UNESCO Institute for Statistics 2017a). Enrolments in secondary schooling have increased fivefold from 11 million in 1990 to 55 million in 2015, but still about 60% remained out-of-school. Other parts of the world, especially East and South East Asia and South America have succeeded in achieving more equitable access to primary and secondary school more rapidly than in sub-Saharan Africa. Strikingly, the problem of out-of-school children is no longer constructed in terms of primary school-age children of 6–11 years without access to education but is about teenagers. In the last five years the global definition of outof-school children has expanded to include children above primary school age (UNESCO Institute of Statistics 2017b). Over 53% of the 262  million now thought to be out of school are of upper secondary age (16–18) and 23% are of lower secondary age (13–15) (UNESCO Institute for Statistics 2018) and overall 50.5% are boys. This rewrites the map of the problem of out-of-school children and the cost of addressing it. It raises questions about whether the right to education extends to the end of the teenage years and if so, how will the delivery of this right be financed? It also places in sharp focus the equity trade-off between the advocacy of the Global Partnership for Education (GPE), which stresses that more investment to achieve these global goals for education requires ‘a stronger focus on the most marginalized’ (arguably those who have never attended school (UNESCO 2016, p. 1) or alternatively investing at secondary and higher levels where the largest numbers are excluded (UNESCO 2017c).

Evolution of aid to education The four transitions detailed earlier have rewritten central parts of the educational planning landscape. They have had profound consequences for aid to education and the analysis and responses to the financing gaps that are central to the crises in access and achievement that have persisted over the last three decades. Much has been made in the last decade of the fact that aid to education has plateaued since 2010 (UNESCO 2017b). Aid to education from Member States

Unequal access to education 47 of the Development Assistance Committee (DAC) rose from the year 2000 to reach about USD 12 billion per year by 2010. Flows of aid have stagnated despite much advocacy to commit greater amounts. Aid to education as a proportion of all aid averaged about 10% for the period 2000–10 and then fell to around 7% (UNESCO 2017b). Aid to basic education is now concentrated in a relatively small number of countries. The Global Partnership for Education (GPE) is the largest source of grant aid for education in LICs, disbursing over USD 500 million per year. About 24% of countries receiving this aid account for 68% of all its aid by value. Some large countries like Ethiopia, Pakistan and the Democratic Republic of the Congo are the major beneficiaries. On the other hand, 42% of aid recipients receive less than 5% of all aid. This means that many countries receive small amounts of aid with the likelihood of higher transaction costs and small-scale impact (Lewin 2017, p. 45). Aid is becoming less important as a source of educational financing in LICs and LMICs. The amount the GPE can disburse is little more than 2% of the additional amounts needed for recurrent financing for the Education 2030 agenda. Significantly, at the 2018 GPE Replenishment conference in Dakar (Global Partnership for Education 2018; UNESCO 2018a) countries likely to be in receipt of GPE grants pledged to increase spending on education to at least 20% of their public budget and to between 4% and 6% of Gross Domestic Product (GDP). These domestic pledges amounted to USD 110  billion annually, dwarfing the USD 2.3 billion pledged by the donors to the GPE over these years. This was a reminder that most of the financial challenges for education are now for domestic financing not aid (Lewin 2017; Zubairi & Rose 2019). If educational inequalities persist, the heart of the problem does not signal a need for more aid. It indicates the need for more domestic commitment backed by political will to change historic patterns of resource allocation to favour greater equity. Development is happening in many LICs and LMICs. Economic growth results in increased domestic revenue and widens the opportunities to generate tax income and facilitate more equitable financing of education. Our best estimates of growth in GDP amongst LICs and LMICs anticipate an average of nearly 5% per annum based on the most recent five-year projections of the International Monetary Fund (IMF) (Lewin 2017). The range is wide from less than 2% per annum to over 8%. At 5% real growth, GDP will double in 15  years with at least the same gain in domestic revenue. At 7% it will double in ten years. Economic growth will move about half of the current LICs into the LMIC category by 2030. These transitions should reduce gaps in educational financing as more revenue is collected. The impact of COVID-19 on medium-term growth is unknown, but most projections continue to assume a return to trend over a ten-year period. There are risks that, if the appetite for aid weakens, countries may borrow to meet short-term needs for recurrent expenditure. Currently, 20 sub-Saharan African countries are formally in or at risk of debt distress according to the International Monetary Fund (IMF 2018). Economic development should lead to less demand for concessionary loans to sub-prime borrowers and more financing

48  Keith M. Lewin from domestic revenue. Some potential lenders continue to argue more credit is needed and make a case for more gap filling aid. Thus, the International Finance Facility for Education (IFFEd) has recently argued that as countries get richer, they need to receive more aid, not less: As countries transition from LIC to LMIC status, aid falls faster than tax receipts rise. Just when many countries start to emerge from very low per capita income, their growth is constrained as domestic taxes and marketrelated public borrowing fail to expand fast enough to compensate for loss of concessional finance. (International Finance Facility for Education, Strategic Case 2019, p. 13) This kind of ‘compensation’ is a strange idea that both Easterly (2013) and Alice in Wonderland would enjoy. If aid was guaranteed, despite aid-related development targets being met, this would provide a perverse incentive to suppress domestic revenue collection and underinvest domestic resources in education. It would increase inequalities between countries. There is an assumption that more lending will drive more growth and that lending can resolve shortfalls in recurrent spending without risk. This is a not so much a theory of change but a pathway towards dependence and default (Lensick & White 1999). Countries which have sub-prime credit ratings as a result of unwillingness to collect revenue are unlikely to evolve sustainable public finances by borrowing more and creating more debt.

Educational financing gaps, past and present The World Conference on Education for All (WCEFA) resulted in The Framework for Action to Meet Basic Learning Needs (UNESCO 1990a). This made clear commitments to mobilise international resources and invest simultaneously in universal access to education and enhanced learning outcomes (UNESCO 1990b). It thus anticipated recent advocacy that what is needed now is access plus learning (Learning Metrics Task Force 2015). The final report from WCEFA stressed the importance of educational quality and included recognition of the ‘third way’ of delivering education through mass media and the informal sector and what would now be called social networks, expert forums and demand-led peer-to-peer learning networks. Learning was always a priority for educational development. However, providing sufficient resources proved an intractable problem. WCEFA committed the international community to mobilise up to USD 2 billion a year over and above existing levels of expenditure to meet the financial challenges of Education for All (Usher 1990, p. 8). The amount of aid needed represented a 40% increase in the level of aid to education in 1990. It was a very affordable amount equivalent to two naval aircraft carriers at a time when defence spending was averaging about 5% of GDP in sub-Saharan Africa. A small peace dividend would have paid the bills. The amount would have been higher without

Unequal access to education 49 a set of reforms to control costs, improve quality, enhance equity and generate enough finance to support Schooling for All (Colclough & Lewin 1993, p. 239). The reforms proposed are worth recalling. Without similar restructuring updated to reflect development since 1990, progressive universal access by 2030 will turn out to be unaffordable. The first set of reforms were cost saving and included double shifting, class size and teaching load management and classroom assistants to extend the reach of qualified teachers. The second group of reforms were cost shifting and included being permissive of self-financed private schools for those who could afford to pay, community contributions to the costs of school building and a freezing of higher education subsidies unless it became more cost-efficient and equitably accessed. The third set of reforms were quality enhancing and included investment in learning materials, increases in teachers’ salaries to ensure recruitment and motivation and measures to increase internal efficiency through reducing repetition and dropout, improved management of learning, limits to the costs to household of attendance and circles of support around children (Colclough & Lewin 1990). The portfolio of possible reforms was designed to balance competing ambitions. The terms of reference were to: • • • •

Remain within plausible increases in financing; Focus on universal basic education and restrain growth at higher levels; Assume assistance would largely take the form of grants not loans; Anticipate that economic growth and increased government allocations to education would close financing gaps after 2005; • Limit aid to levels that did not create long-term dependence; • Allocate most aid to countries where the needs were greatest and use aid to enhance equity; • Gain political and professional commitment for the proposed reform agenda. Fast forward to 2019 and the dimensions of the financing dilemma facing LICs and LMICs echo the earlier analysis in 1990. The level of ambition has changed beyond recognition and includes the kind of ‘successful globalisation’ that leads to education for sustainable development that respects environment, society and economy (Little & Green 2009). Now the 2015 UN Sustainable Development Goals (SDGs), and in particular the education-specific SDG 4, anticipate universal enrolment from pre-primary through to grade 12 with relevant learning outcomes for all (UNESCO 2015b). They also include expanded access to further and higher education. This generates very large gaps between the resources currently allocated to education by governments in LICs and LMICs and the funding necessary, as noted in following discussion. These new estimates are the result of ambition untempered by credible planning and realistic revenue streams. They are untroubled by what can be learnt from the experience of Highly Indebted Poor Countries in the 1980s and 1990s and the problems of financing recurrent costs during economic downturns. Aspiration to universalise early childhood development, pre-school and upper-secondary

50  Keith M. Lewin enrolment is running ahead of the fiscal prioritisation that would be necessary to support these ambitions. There are six issues that will facilitate and challenge progress towards the kind of sustainable financing that can create the conditions for reducing inequalities in participation and attainment. (1) Recent modelling for the GPE (Lewin 2017, p. 54) indicates that, if both primary and lower secondary school were to be universalised with plausible efficiency gains, the amounts needed for education would average between 6% and 6.5% of GDP in LICs and LMICs. This scenario would still leave almost half of all children in LICs without access to upper secondary and less than 15% of the age group enrolling in higher education. Providing universal access to pre-school would add 15% to the total cost and further and higher education more. The current estimated total public expenditure on education across the LICs is about USD 19 billion and for LMICs USD 68 billion, representing 3.7% and 4.6% of GDP, respectively. This includes current aid contributions. To reach or exceed 6% of GDP would cost at least another USD 13 billion per year for the LICs and USD 22 billion for the LMICs totalling over USD 35 billion a year for schools alone. (2) The IFFEd has generated much higher estimates of costs for the SDGs to be achieved in LICs and LMICs (International Finance Facility for Education 2016, p. 105). In their estimates, DAC donors would have to increase aid to education in LICs alone from USD 12 billion to USD 49 billion a year, nearly four times that of current levels. This can be compared to the 1990 estimate of USD 2 billion a year albeit that this was simply to universalise primary education (Colclough & Lewin 1990). The IFFEd modelling makes the heroic assumption that it is realistic for the LICs to spend nearly 12% of their GDP on education with half of that being financed by aid on a recurrent basis. LMICs would need 7.5% of GDP. No country has ever sustained such high levels of GDP allocation. The projections fall outside the envelop of the credible plan advanced at the WEF 2000 as the criterion for external financing of education in LICs. The amounts are four times as big as the Marshall Plan to reconstruct Europe after World War II which was worth only 3% of GDP and it took some countries more than 60 years to pay off associated loans. There is no clear pathway to end dependence on large-scale loans even though this would be essential for any credible plan. If such large volumes were mobilised as loans lasting aid dependence would be the result. The burden will be even less sustainable if countries contract much more debt as a result of COVID-19. (3) The Education 2030 Framework for Action agreed at WEF 2015 ‘urges adherence to the international and regional benchmarks of allocating at least 4–6% of GDP and/or at least 15–20% of total public expenditure to education’ (UNESCO 2015a, p. 9). Currently 40% of LICs and LMICs spend less than 4% of GDP on education and less than 15% allocate more than 6% of GDP. Fewer than 20% of LICs and LMICs spend more than 20% of their

Unequal access to education 51 government budgets on education. There is therefore a long way to travel to reach and sustain spending at target levels. The proportion of GDP allocated to education is currently 3.7% in LICs and 4.6% in LMICs. These average levels have persisted despite sustained advocacy for increases. The proportion of government budgets allocated to education in LICs averaged 16% and LMICs averaged 17% (UNESCO 2017a, p. 404). If the share of the government budget for education was 20% (which is 33% greater than the current average for LICs and LMICs), and the amount collected from domestic revenue was the LIC/LMIC average of 16% of GDP, then this would result in education expenditure being only 3.2% of GDP without aid (i.e. 20% of 16%). It would need at least 35% of the government budget to provide 5.8% of GDP which is the amount likely to be needed. It is important to note that when GDP falls, as it did in the UK and many other countries after the 2008 financial crisis, the proportion of government spending allocated to education can increase although the actual cash amounts are falling (Lewin 2018, p. 4). COVID-19 will reduce GDP in most LICs and LMICs and education spending as a proportion of GDP will appear to increase, but this will be a pyrrhic victory if the real value falls as a result of the contraction in GDP. The key to sustainable financing lies in increasing domestic revenue as a share of GDP at the same time as there is economic development and growth in GDP. (4) Revenue generation can be problematic. The best estimates suggest that in LICs and LMICs in sub-Saharan Africa income tax charged on personal income amounts to between 5% and 10% of all tax revenue. This compares with a share of over 40% in Organisation for Economic Co-operation and Development (OECD) countries. Income tax is paid by only about 5% of all people who live in Africa, compared to 50% of adults in the OECD. In one East African country only 5% of all company directors pay any income tax, and few of the wealthiest officials pay any income tax at all. A  recent leak revealed that about 5,000 Africans held assets of over USD 6 billion in just one Swiss Bank. The wealthiest client with a personal account balance of over USD 700  million came from one of Africa’s poorest countries. Corporate tax avoidance and evasion appears widespread. There is no doubt that large amounts of income and assets – some suggest USD 500 billion a year – are lost through transfer pricing, money laundering and simple theft. The result is gaps in educational financing. (5) The good news is that some developments are encouraging progress towards sustainable educational financing which could accelerate reductions in inequality. National systems for raising revenue are becoming more efficient and effective. This is transforming the landscape of educational financing and the size of the gaps that exist between what is currently financed and what is needed. Aid to sub-Saharan Africa was greater than tax receipts from 1986 to 1995. Since then it has fallen relative to GDP and tax revenues are now twice the value of aid (Moore et al. 2018). Some LMICs already use some of their revenue to finance their own aid programmes. India, with one of the

52  Keith M. Lewin largest financing gaps projected by the IFFEd, actually gives three times as much aid as it receives. The trends are likely to continue with aid shrinking and tax revenues growing. Indeed, this is what is supposed to happen when countries develop and when aid programmes are effective. As low-income economies grow, direct taxes become a larger share of revenue, and total revenue should grow faster than the economy as the modern sector increases its share of economic activity. Taxes will also become more difficult to avoid with better biometric identification, electronic tracking of transactions and compliance with international transparency requirements. (6) Lastly, the evolution of LICs towards becoming Fiscal States has immense significance. By 2030, tax, not aid, will be the dominant source of public finance in most LICs. More governments should be able to finance their own development and take control of their development agenda. If there is a low learning trap (World Bank 2018, p. 199), it is in large part a low financing trap. It has been argued that ‘this year poses some real opportunities to unlock education for everyone – but only if we nail down exactly how we are going to do it and where the money is going to come from?’ (UNESCO 2020, p. 1). However, both these questions have answers located firmly within countries and determined by the national political economy of possibilities rather than in more aid. The We in question applies to governments and citizens, rather more than to development agencies.

Concluding remarks Understanding how systems are changing, and counting costs to make it more affordable to provide universal access to education, is essential if the next generation of aid is to have more impact than the last. The first part of this chapter provided analytic insight into four transitions that will shape how education systems in different groups of LICs and LMICs will develop. The second part detailed the evolution of aid to education and identified challenges going forward to 2030. Reflections on three decades of educational financing lead to the conclusion that a two-pronged strategy is needed to discourage history from repeating itself. The first prong is to support research and investment in education system planning and management that leads to durable gains in efficiency and effectiveness that make it possible to finance systems from domestic revenue. This approach can be used to develop methods to track costs and inequalities, identify sources of inefficiency in teacher deployment and utilisation, optimise school location, chart flows of students linked to costs, profile strategies to promote learning, assess inequalities, monitor the development of private providers, assess the costs to households of shadow schooling and identify low-cost levers for change. This is the kind of approach used in the Department for International Development support for Rashtriya Madhyamik Shiksha Abhiyan in India (Lewin et al. 2015c). LICs and LMICs in sub-Saharan Africa spend relatively more on education and get relatively less in terms of access and learning outcomes than in most

Unequal access to education 53 other parts of the world. In many countries twice as many students could attend secondary school if teacher deployment was as efficient as in the most effective systems. Efficiency and effectiveness gains could generate many billions of dollars of savings and reduce educational financing gaps. Conversely, many billions of dollars of additional funding without enhanced efficiency and effectiveness is unlikely to produce sustainable educational development that is worth financing. Three generations of aid to education since the 1960s have not yet succeeded in catalysing global transitions to more efficient, effective and equitable systems. The task now is to do better the fourth time around. The second prong of the new approach is to invest in supporting fiscal reforms that can increase domestic revenue. There is plenty of scope to raise more revenue and reduce the need for aid to fill financing gaps. The fundamental point is that a 1% increase in collection of revenue in sub-Saharan Africa would be roughly equivalent to all the aid to education from DAC countries. It would not have to be replenished on an ad hoc basis. It would be complemented by real economic growth that would result in more revenue to fund more services. It would reduce the need to contract new loans and risk a debt crisis. More domestic revenue will be generated as economies grow and revenue collection becomes more efficient, evasion more difficult and money transfers more transparent. The uncertainty is more about how additional revenue will be spent rather than whether more will be collected. The critical shift in perspective is to realise that supporting fiscal reform is a kind of aid to education. The analysis in this chapter should not be misunderstood. More aid is needed but not of the finance gap-filling kind that failed to result in sustainable educational development in the past. Gaps in educational financing are generated by aspirations and by failure to match these with the political economy of good governance that balances resources with spending and ambition with financial and other kinds of accountability. Aid and external assistance should never be a substitute for domestic political will. A goal of external financing should be to reduce the need for more external financing. Time may be running out on gap-filling aid and as Akufo Addo, President of Ghana argued: We cannot depend on other people to finance the education on our continent. I am saying that not to turn my back or to be ungrateful to all these important or noble people who have committed themselves to help, no . . . But, if we make our policy dependent on other people when their policy changes, we will suffer. But, if we make the policy for ourselves, then it means that, at all times, we will be in control of our own destiny. (Akufo Addo, President of Ghana, GPE Replenishment, Dakar, Feb 2018)5 The next decade will tell if this refreshing rhetoric is matched by a willingness to curate educational aid towards a new agenda that seeks to reduce the need for more aid. The purpose of counting the costs of unequal access to education is

54  Keith M. Lewin to find a solution to how to pay costs not once, but once and forever, through fiscal reforms rather than the well-intentioned but volatile benevolence of aid. Development is something that can be accelerated by aid but not caused by it (Seers 1969, 1981). Dependency theory provides a reminder that to develop should be an intransitive rather than transitive verb with the onus on countries to develop themselves albeit with measured and complementary support to accelerate progress. There is no solution to closing educational financing gaps that does not depend on the development of states that can fund public goods from domestic revenue. There is no solution that does not also work to promote efficiency and effectiveness and mobilise resources to best effect and minimise negative effects on the physical and social environment. Enduring solutions will be endogenously determined and financed by fiscal states.

Notes 1 Christopher Colclough and I first worked together in the Human Resource Group at the Institute of Development Studies at Sussex in the 1970s. This led to our work with the Bellagio Group of donors, later the Association for the Development of Education in Africa (ADEA) that catalysed the shift to aid for basic education (Lewin et al. 1982). Subsequently, we worked together for UNICEF on the financing Round Table at the World Conference on Education for All (WCEFA) (UNESCO 1990a) and helped develop the first Education for All programme. Fifteen years later Chris and I directed sister Research Consortia on Education for the Department for International Development (DFID) exploring educational access and outcomes in low-income countries. 2 This chapter was written before the COVID-19 global pandemic which will have far-reaching effects on the delivery of educational services and on how they can be financed. It is too early to predict what these will be with any precision. The basic arithmetic of educational financing will remain the same after the pandemic passes as it is structurally determined, so the analysis in the chapter will remain valid albeit with changes in the values of some of the parameters. 3 World Bank definitions are used for LICs and LMICs. 4 Developing countries as defined by EFA Global Monitoring Report 2002, pp. 19. 5 YouTube of speech available at www.africanvibes.com/ghanas-president-akufoaddo-speech-at-gpe/. Reported at https://www.myjoyonline.com/news/africa-hasresources-to-finance-its-citizens-education-akufo-addo/

References Canning, C., Raja, S., & Yazbeck, A. (2015) Africa’s Demographic Transition: Dividend or disaster? Washington, DC: World Bank. Colclough, C. (1977) Formal education systems and poverty-focused planning. Journal of Modern African Studies, 15(4), pp. 569–589. Colclough, C. (1997) Education, health and the market: An Introduction. In Colclough, C.L. (Ed.) Marketizing Education and Health in Developing Countries: Miracle or Mirage. Oxford: Clarendon Press. Colclough, C., Al-Samarrai, S., Rose, P., & Tembon, M. (2003) Achieving Schooling for All in Africa: Costs, Commitment and Gender. Ashgate: Aldershot.

Unequal access to education 55 Colclough, C., & Lewin, K.M. (1990) Educating All the Children: Meeting the Economic Challenges of the 1990s. New York: UNICEF. Colclough, C., & Lewin, K.M. (1993) Educating All the Children: Strategies for Primary Schooling in the South. Oxford: Clarendon Press. Colclough, C., & Manor, J. (Eds.) (1993) States or Markets? Neo-liberalism and the Development Policy Debate. Oxford: Clarendon Press. Easterly, W. (2013) The Tyranny of Experts: Economists, Dictators and the Forgotten Rights of the Poor. New York: William Easterly Paperback. Global Partnership for Education (2018) Unprecedented Worldwide Support for the Global Partnership for Education and Addressing the Global Learning Crisis. [Online] Global Education Monitoring Report, press release: Replenishment Conference Dakar 2018. Available from: www.globalpartnership.org/news-andmedia/news/unprecedented-worldwide-support-global-partnership-educationand-addressing-global-learning-crisis [Accessed 28 May 2020]. International Finance Facility for Education (IFFEd) (2016) The Learning Generation: Investing in Education for a Changing World. [Online] A  Report by the International Commission on Financing Global Education Opportunity, United Nations, New York. Available from: http://report.educationcommission.org/ report [Accessed 28 May 2020]. International Finance Facility for Education (IFFEd) (2019) The Strategic Case for IFFEd International Commission on Financing Global Education Opportunity. New York: United Nations. International Monetary Fund (2018) IMF Warns of Rising African Debt Despite Faster Economic Growth. [Online] Available from: www.reuters.com/article/ us-africa-imf/imf-warns-of-rising-african-debt-despite-faster-economic-growthidUSKBN1I9114 [Accessed 3 June 2020]. Learning Metrics Task Force (2015) Assessment for Learning (A4L): An International Platform to Support Assessment for Learning. [Online] Washington, DC: Brookings Institution. Available from: https://learningportal.iiep.unesco.org/en/ library/assessment-for-learning-a4l-an-international-platform-to-support-nationallearning [Accessed 1 June 2020]. Lensick, R.,  & White, H. (1999) Aid Dependence. Issues and Indicators. [Online] Report to the Expert Group on Development Issues. Almqvist  & Wiksell International, Stockholm. Available from: www.bistandsdebatten.se/wp-content/ uploads/2012/10/1999_2-Aid-Dependence.-Issues-and-Indicators.pdf [Accessed 28 May 2020]. Lewin, K.M. (1985) Quality in question; A  new agenda for curriculum reform in developing countries. Comparative Education, 21(2), pp. 117–133. Lewin, K.M. (1994) Cost Recovery and the Role of the State. Invited Paper. 12th Commonwealth Ministers of Education Conference, Pakistan. November. Commonwealth Secretariat, London, p. 30. Lewin, K.M. (2000) Knowledge Matters for Development. [Online] Professorial Lecture, University of Sussex. Available from: https://keithlewin.net/wp-content/ uploads/2019/01/Keith-Lewin-Knowledge-Matters-for-Development-.pdf [Accessed 30 June 2020]. Lewin, K.M. (2007) The Limits to Growth of Non-Government Private Schooling in Sub Saharan Africa. [Online] CREATE Pathways to Access Research Monograph No. 5, University of Sussex. Available from: www.create-rpc.org/pdf_documents/ PTA5.pdf [Accessed 30 June 2020].

56  Keith M. Lewin Lewin, K.M. (2008) Strategies for Sustainable Financing of Secondary Schooling in Sub Saharan Africa: Africa Human Development Series. [Online] World Bank, Working Paper No 136. Washington, DC. Available from: https://pdfs.semanticscholar.org/99c3/ffa4db01f6a0120ef1d9f6ca1b01d535864b.pdf [Accessed 28 May 2020]. Lewin, K.M. (2009) Access to education in Sub-Saharan Africa: Patterns, problems and possibilities. Comparative Education, 45(2), pp. 151–174. Lewin, K.M. (2015) Educational Access, Equity and Development: Planning to Make Rights Realities. [Online] Fundamentals of Educational Planning No. 97. International Institute for Educational Planning, UNESCO, Paris. Available from: http://unesdoc.unesco.org/images/0023/002350/235003e.pdf [Accessed 28 May 2020]. Lewin, K.M. (2017) The Educational Challenges of Transition: Key Issues for Low and Low Middle Income Countries and the GPE Towards 2030. [Online] Occasional Paper 2, Global Partnership for Education, Washington, DC. Available from: www. globalpartnership.org/content/educational-challenges-transition-key-issues-2030 [Accessed 28 May 2020]. Lewin, K.M. (2018) Learning Matters and the World Development Report 2018. [Online] UK Forum for International Education and Training (UKFIET) Blog. Available from: www.ukfiet.org/2018/learning-matters-and-the-world-development-report-2018/ [Accessed 1 June 2020]. Lewin, K.M., & Caillods, F. (2000) Financing Secondary Schools in Developing Countries: Strategies for Sustainable Growth. Paris: International Institute for Educational Planning, UNESCO. Lewin, K.M., Little, AW.,  & Colclough, C.L. (1982) Adjusting to 1980s; Taking stock of educational expenditure. In Financing Educational Development: Proceedings of an international seminar held in Mont Sainte Marie, Canada. Ottawa: International Development Research Committee, pp. 13–39. Lewin, K.M., & Sabates, R. (2012) Who gets what? Is improved access to basic education pro-poor in sub-Saharan Africa? International Journal of Education and Development, 32(4), pp. 517–528. Lewin, K.M., Siddhu, G., & Jha, S. (2015a) Demographic Change and Educational Planning. [Online] Rashtriya Madhyamik Shiksha Abhiyan-Technical Cooperation Agency, Ministry of Human Resources Development, Government of India. Available from: http://keithlewin.net/wp-content/uploads/2016/05/2.-Demographic-Transition-and-Education-Planning.pdf [Accessed 28 July 2020]. Lewin, K.M., Siddhu, G.,  & Jha, S. (2015b). Equity and Efficiency in Expansion of Secondary Schools. [Online] Rashtriya Madhyamik Shiksha Abhiyan- Technical Cooperation Agency, Ministry of Human Resources Development, Government of India. Available from: http://keithlewin.net/wp-content/uploads/2016/05/3.Equity-and-Efficiency-in-Expansion-of-Secondary-Schools.pdf [Accessed 28 July 2020]. Lewin, K.M., Siddhu, G., Jha, S., Härmä, J.,  & Parolin, B. (2015c). Equity in Access and Learning: A  Way Forward for Secondary Education. Research Report 0. [Online] Rashtriya Madhyamik Shiksha Abhiyan Technical Support Agency, Ministry of Human Resource Development, Government of India and DFID. Available from: https://keithlewin.net/wp-content/uploads/2016/05/0.-Synthesis-Equity-in-Access-and-Learning-in-India.pdf [Accessed 28 July 2020].

Unequal access to education 57 Little, A.W.,  & Green, A. (2009) Successful globalisation, education and sustainable development. International Journal of Educational Development, 29(2), pp. 166–174. Miller, E. (1991) Men at Risk. Kingston, Jamaica: Jamaica Publishing House. Moore, M., Prichard, W., & Fjeldstadt, O. (2018) Taxing Africa: Coercion, Reform and Development. London and New York: Zed Press. Seers, D. (1969) The meaning of development. International Development Review, 11(4), pp. 3–4. Seers, D. (1981) Dependency Theory: A Critical Reassessment. London: Francis Pinter Publishing. UNESCO (1990a) World Declaration on Education for All: A Framework for Action to Meet Basic Learning Needs. [Online] Paris: UNESCO. Available from: www. humanium.org/en/wp-content/uploads/2017/03/Education_for_all.pdf [Accessed 1 June 2020]. UNESCO (1990b) Meeting Basic Learning Needs: A Vision for the 1990s. Background Document. World Conference on Education for All March  5th-8th Jomtien, Thailand. [Online] Paris: UNESCO. Available from: https://unesdoc.unesco.org/ ark:/48223/pf0000097552 [Accessed 1 June 2020]. UNESCO (2000) The World Education Forum; The Dakar Framework for Action. [Online] Paris: UNESCO. Available from: https://sustainabledevelopment. un.org/content/documents/1681Dakar%20Framework%20for%20Action.pdf [Accessed 1 June 2020]. UNESCO (2015a) Education 2030: Incheon Declaration and Framework for Action for the Implementation of Sustainable Development Goal 4. [Online] Paris: UNESCO. Available from: http://uis.unesco.org/sites/default/files/documents/education-2030-incheon-framework-for-action-implementation-of-sdg4-2016-en_2. pdf [Accessed 1 June 2020]. UNESCO (2015b) World Education Forum, Final Report. [Online] Incheon, Republic of Korea. Available from: www.unesco.org/new/fileadmin/MULTIMEDIA/ HQ/ED/ED_new/pdf/WEF_report_E.pdf [Accessed 1 June 2020]. UNESCO (2016) UNESCO Examines Financial Shortfall Reaching Global Education Goals. [Online] Paris: UNESCO. Available from: https://en.unesco.org/ news/unesco-examines-financial-shortfall-reaching-global-education [Accessed June 2020]. UNESCO (2017a) Accountability in Education: Meeting Our Commitments. [Online] Global Education Monitoring Report No. 2017/18. UNESCO, Paris. Available from: https://unesdoc.unesco.org/ark:/48223/pf0000259338 [Accessed 1 June 2020]. UNESCO (2017b) Aid to Education Is Stagnating and Not Going to Countries Most in Need. [Online] Policy Paper No. 31. UNESCO, Paris. Available from: http://unesdoc.unesco.org/images/0024/002495/249568e.pdf [Accessed 1 June 2020]. UNESCO (2017c) Leaving No One Behind: How Far on the Way to Universal Primary and Secondary Education? [Online] Global Education Monitoring Report Policy Paper No. 27. UNESCO, Paris. Available from: http://unesdoc.unesco. org/images/0024/002452/245238E.pdf [Accessed 28 May 2020]. UNESCO (2018a) Raising Billions of Dollars for Education. [Online] World Education Blog. Global Education Monitoring Report. Available from: https://

58  Keith M. Lewin gemreportunesco.wordpress.com/2018/02/05/raising-billions-of-dollars-foreducation-in-dakar/ [Accessed 28 May 2020]. UNESCO (2018b) Did You Know There Are Just as Many Boys Out of School as Girls? [Online] World Education Blog. Global Education Monitoring Report. Available from: https://gemreportunesco.wordpress.com/2018/06/15/did-you-knowthere-are-just-as-many-boys-out-of-school-as-girls/ [Accessed 28 May 2020]. UNESCO (2020) Unlock Education for Everyone. [Online] World Education Blog. Global Education Monitoring Report. Available from: https://gemreportunesco. wordpress.com/ [Accessed 28 May 2020]. UNESCO Institute for Statistics (2017a) Combining Data on Out-of-School Children, Completion and Learning to Offer a More Comprehensive View on SDG 4. [Online] Available from: http://uis.unesco.org/sites/default/files/documents/ip61-combining-data-out-of-school-children-completion-learning-offer-more-comprehensive-view-sdg4.pdf [Accessed 1 June 2020]. UNESCO Institute for Statistics (2017b) Counting the Number of Children Not Learning: Methods for Global Composite Indicators for Education. [Online] Available from: https://unesdoc.unesco.org/ark:/48223/pf0000261560 [Accessed 1 June 2020]. UNESCO Institute of Statistics (2018) One in Five Children, Adolescence and Young People Is Out of School. [Online] Fact Sheet No. 48 February 2018 UIS/FS/2018/ ED/48. Available from: http://uis.unesco.org/sites/default/files/documents/ fs48-one-five-children-adolescents-youth-out-school-2018-en.pdf [Accessed 1 June 2020]. UNICEF (2015) The Investment Case for Education and Equity. [Online] New York: UNICEF. Available from: www.unicef.org/publications/index_78727.html [Accessed 1 June 2020]. United Nations (1948) Universal Declaration of Human Rights. [Online] New York. Available from: www.un.org/en/universal-declaration-human-rights/#:~:text=The %20Universal%20Declaration%20of%20Human%20Rights%2C%20which%20 was%20adopted%20by,of%20the%20Second%20World%20War [Accessed 30 June 2020]. Usher, A.D. (1990) North challenges to foot $10  billion bill. The Nation, Friday March 9, Jomtien, Thailand p. 8. Verspoor, A. (2008) At the Crossroads: Choices for Secondary Education in Sub-Saharan Africa. Washington, DC: World Bank. World Bank (2018) World Development Report 2018: Learning to Realise Education’s Promise. Washington, DC: World Bank. World Inequality Database on Education (2017) World Inequality Database on Education. [Online] Global Education Monitoring Report. UNESCO, Paris. Available from: www.education-inequalities.org [Accessed 1 June 2020]. Zubairi, A., & Rose, P. (2019) Equitable Financing of Secondary Education in SubSaharan Africa. Background Paper for the Mastercard Foundation. [Online] Research for Equitable Access and Learning (REAL) Centre, University of Cambridge. Available from: https://mastercardfdn.org/wp-content/uploads/2019/03/SEAFinance-Equity_REAL_Final-Version_Feb-2019-1.pdf [Accessed 1 June 2020].

4 Education for All in India and Sri Lanka The drivers and interests shaping egalitarian reforms Angela W. Little This chapter considers the limits and possibilities of education policy from an historical perspective with respect to inequalities in accessing educational opportunities. The story of Education for All (EFA) cannot be divorced from the history of education in particular countries. Much education history has been either ignored or forgotten in educational discourse. This chapter is written in the spirit of Learning from the Past (Benavot & Resnick 2007) and Learning from Developing Countries (Little 1988) in the belief that comparative lessons from the past broaden the conceptual basis on which alternative policies and intervention strategies may be evaluated in the present and the future. EFA was a term coined by the World Conference on ‘Education for All: Meeting Basic Needs’, held in Jomtien, Thailand in 1990. At its heart was the universalisation of primary education (UPE) of good quality worldwide. This goal had many precedents, names and slogans. Soon after its foundation in 1945, UNESCO launched movements for an expansion of fundamental education, building on the historical experiences of countries. Regional UNESCO conferences in the 1960s set out goals for the achievement of UPE. Following Jomtien, the Dakar Framework for Education, the Millennium Development Goals and, more recently, the Sustainable Development Goals have continued to set worldwide goals for, inter alia, primary and junior secondary education. By employing a comparative perspective in this chapter, I  examine a range of historical forces that have shaped primary and junior secondary education in India and Sri Lanka.1 India and Sri Lanka are geographically contiguous countries, separated by a narrow stretch of water. They have many characteristics in common: colonial relations with the Portuguese, the Dutch and the British; political independence from the British at around the same time (India, 1947; Sri Lanka 1948); democratic political regimes; lower middle-income economic status and multilingual, multi-ethnic and multi-religious societies. Sri Lanka’s main languages, religions and cultural practices derive from India. Yet, despite these similarities and interrelations, educational achievements of India and Sri Lanka differ markedly. In 2016, the literacy rate in India was 75% and in Sri Lanka 97%. In 1950, around the time of independence in both countries, the rates were 19% and 65% respectively. The historical reasons for these differences form the main foci of this chapter.

60  Angela W. Little The United Nations did not invent the idea of universalising access to fundamental education. Both India and Sri Lanka had experience of policies and schemes for elementary education, mass education and primary education long before the establishment of the United Nations in 1945. Indian and Sri Lankan experience demonstrates that any analysis of progress towards EFA must involve an understanding of the history of such schemes before 1945 and since. Moreover, analysis must go beyond the description of educational conditions and legislation and address, inter alia, the underlying political drivers of changes in national policy and the range of interests that surround, promote and resist their introduction and implementation over time.2 That is as true today of progress towards the UN Sustainable Development Goal for Education as it was yesterday of progress towards UNESCO’s goal of fundamental education. This chapter therefore focuses attention on access to the first eight years of formal education in India, known as elementary education (or lower and upper primary school, consisting of 5 and 3 grades, respectively) and the first nine years in Sri Lanka, known as the primary and junior stages of education (5 and 4 grades, respectively). Access to these grades is treated as an aspect of social equality. Social inequalities are reflected in inequalities of access to education since social inequality is often a major source of competition and conflict between interest groups which in turn drive policies intended to reduce or maintain inequality. Social inequality is thus both a cause and a consequence of education inequality. This chapter tells the story of such inequalities across four time periods: (1) preindependence (2) 1948–70 (3) 1970–90 and (4) 1990 to the present.3 However, before analysing these four periods, let us first note that any comparison of India and Sri Lanka involves the question of size. A simple answer to the question about differences in educational achievement might point to India’s size and population. India is a vast country of 3.287 million square kilometres with a population of 1.3 billion in 2019. Sri Lanka is an island nation located to the South East of India with a land area of just 65,610 square kilometres and a population in 2019 of 21.4 million, a tiny fraction of that of India. Sri Lanka’s population is similar to that of a small state in India. The experience of an even larger country, China, shows that size, per se, does not explain differential rates of progress towards equality of access to education. In 1950, India’s population was estimated to be 376 million and China’s 544 million. Their reported literacy rates were reportedly similar  – 19% and 20% for India and China respectively. By 2015 the rates were 72% and 96% respectively, with large differences between males and females in India and small differences in China (Zhang 2005). While a detailed discussion of differences between India and China is beyond the scope of this chapter, part of the explanation must lie in variations in political regime and the role of central government, cultural and linguistic complexity, attitudes to girls and women and differential rates of urbanisation. Indeed, the political regimes of individual states have frequently been used to explain differences in educational progress within India.4 With this in mind, we turn now to the first historical period in education – preindependence from the British rule.

Education for All in India and Sri Lanka 61

Pre-independence: competing interests and traditions Religious traditions in education predated the influence of Western religions in education by many centuries in both India and Sri Lanka. Education was village based, oriented to religious content and mediated through oral, and later, written literacy. From the sixteenth-century European religious interests  – the Portuguese Catholics and the Dutch and British Protestants  – began to shape the structure and content of education in both the countries, especially in the education of the local elites for positions in the church, commerce and government. Education for the masses – usually in vernacular languages and oriented to religious conversion – was promoted to varying degrees through networks of schools linked to respective religions. Neither India nor Sri Lanka represented well-defined political entities before the British colonial period, the Portuguese and the Dutch having confined their activities to the maritime areas. Britain’s contact with India was established via the British East India Company which traded with India before ruling from 1757 to 1858 (Basu 2012). By contrast, the British East India Company administered Sri Lanka jointly with the British Government for only four years, from 1798. Sri Lanka became a British crown colony in 1802. Whilst India became a colony from 1858, Sri Lanka’s direct experience of British colonialism lasted much longer than India’s. From 1858, the British Raj ruled approximately half of India directly and the other half indirectly through the native or princely states. Recent evidence on the correlation between direct/indirect rule and literacy, both in 1908 and in the post-colonial era, suggests little difference between reported literacy rates, though middle school provision appears to have been higher in indirectly ruled states (Iyer 2008; Roy 2014). In Sri Lanka, colonial rule was confined initially to the maritime areas, but from 1833 a unitary administrative and judicial system covered the whole country. In the earliest years, the colonial government had established three preparatory schools for high caste, male Sinhalese, Tamils and those of European descent and the Colombo Academy with parallel streams for these three ethnic groups. The schools employed the English medium, imparted Christianity and prepared candidates for church and government service. Mass education was provided by the parish schools. Although the British had stopped the salaries of Dutch parish schoolteachers, those same teachers agitated successfully for the rehabilitation of the parish schools (Jayasuriya 1979, p. 42). This did not mean that there was a policy commitment to public provision of education. It was more of a political response to the agitation of teachers who had formerly taught in the Dutch parish schools and the need to promote the growth of a loyal Christian population in ‘a newly conquered territory that was bounded on the inside by an independent native Kingdom’ (Jayasuriya1979, p. 42). Unlike some of his predecessors and successors, Governor Brownrigg (1812–20) encouraged the work of Christian missionaries in education.

62  Angela W. Little In India, Wood’s Despatch (1854) (Moore 1965) would come to have some limited impact on mass education. The Despatch recognised the responsibility of government for elementary education in the vernacular medium as well as education through the medium of English at the post-elementary level. Grants-in-aid were offered to support the work of non-government bodies in promoting an expansion of educational opportunity. But the expansion of primary education for the masses was resisted by many, especially the high-caste Hindu professional classes. In 1882 an Indian Education Commission again promoted government responsibility for education and in 1911 Gokhale introduced an Elementary Education Bill to establish compulsory elementary education as a responsibility of individual states. The Bill met with resistance from those landowners who feared that mass elementary education would interfere with child labour and was defeated in the Imperial Assembly (Kumar 2005, pp. 118–120). A famous exception was the introduction of compulsory and free primary education in the state of Baroda by the Maharaja Sayajirao Gaekwad III, as early as 1906 and far in advance of the rest of the country. Although the Despatch had been written for the Indian context, it had some impact in Sri Lanka via the work of the Morgan Committee of 1869. An expansion of the grants-in-aid scheme stimulated an expansion of enrolment in Protestant missionary schools. But it also fuelled intense competition between them, one outcome of which was the opening up of educational opportunities for girls and women and those living in areas with poor or no education provision. Over time, Roman Catholic interests coalesced with those of the Buddhists and Hindus as they made common cause against the Protestants and demanded more grants-in-aid. Between 1869 and 1900 the number of government schools grew fourfold from 120 to 500 while the grant-aided schools grew more than 60-fold, from 21 to 1,328 (Jayasuriya 1979). By1901, the adult literacy rate was 26.4%. In the early twentieth-century nationalist political struggles in both countries included calls to redress social inequalities in education. In India the expansion of education opportunity was central to the nationalists’ demands, promoted by Lajpat Rai, Mahatma Gandhi, Tagore and others. In 1937 Gandhi convened the Wardha conference where a radical national plan for basic education recommended that children from all communities be educated side by side, with religious instruction barred from state-funded schools. Yet the implementation of the scheme was thwarted, this time by opposition from Muslim groups (Oesterheld 2007). In 1944 the Central Advisory Board of Education drew up a plan, known as the Sergeant Plan, setting a time target of 40 years for the achievement of UPE. Sri Lanka’s transition to independence was, by comparison with India’s, ‘an oasis of stability, peace and order’ (de Silva 1981, p.  489). Nonetheless, ‘the Indian nationalist movement was a source of strength for the Ceylonese, and Ceylon enjoyed the fruits of success of the Indian movement’ (Jayasuriya 1979, p.  426). The State Council in Sri Lanka granted universal franchise in 1931, 20 years before universal franchise was introduced in post-independence India. This was followed by the implementation of social welfare programmes, including education. The Minister of Education of the State Council, C.W.W. Kannangara

Education for All in India and Sri Lanka 63 promoted a scheme for fee-free access to English-medium post primary schools,5 teacher training colleges and university. Urging an acceptance of the scheme he said, ‘we shall be able to say that we found education . . . the patrimony of the rich and left it the inheritance of the poor’ (Jayasuriya 1979, pp. 471–472). Challenged by a range of colonial, social, political, economic and religious vested interests that saw it as undermining their privileges, the scheme was nevertheless passed by the State Council in 1945. A network of central schools was established in rural areas to enable able children to access high-quality post primary English-medium education. Financial assistance was available for the poorest groups. Simultaneously, the medium of instruction in all government primary schools became Sinhala or Tamil. By the time of independence in 1948 Sri Lanka could boast an adult literacy rate of 67%, an increase over five decades of around 45 percentage points, while India’s adult literacy rate was 19%, lower than the rate in Sri Lanka 50 years earlier.

Independence to 1970: the drivers for educational expansion With such different educational and administrative histories, it was not surprising that the drivers for education expansion differed after each country achieved independence. The 1950 Constitution established India as a Federal Republic and enacted universal adult franchise (Shani 2018). Prime Minister Nehru’s vision for an independent India focussed on industrial development, contrasting markedly with Gandhi’s concept of village-based development (Jodhka 2002). Although the distinction between directly and indirectly ruled states disappeared, education continued to be planned and administered by the states. Article 45 obliged the states ‘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 14 years’ (Government of India 1950). Thirteen years after independence and the implementation of a range of education programmes, the educator J.P. Naik bemoaned the slow progress in elementary education, commenting that although the 1944 Sergeant Plan to provide free and compulsory education for those aged 6–13  years within 40  years was regarded in its time as ‘anti-national’ and ‘fantastically slow’  .  .  . today, we have come to a stage where the implementation of the Sergeant Plan will be regarded as a ‘progressive, bold and ambitious target!’. Why is it, he asked, ‘that we are unable to implement the only directive principle of the state policy in education?’ (Naik, 1966) He attributed slow progress to ‘apathy’ to education of the illiterate masses, traditional resistance to girls’ education, the existence of ‘backward’ groups such as scheduled castes and tribes and nomads, household poverty and the need for child labour, small and scattered habitations, forests and inaccessible areas, a high

64  Angela W. Little birth rate and explosion of the school-age population, a shortage of finance and the absence of a machinery to enforce compulsory attendance (Naik 1966, p. 5). Kumar (2005, p.  192) suggested that the promotion of mass education was resisted by the owners of property and capital who saw in it an undermining of the supply of cheap child labour. Naik subsequently became the member-secretary of the 1964 Kothari Education Commission which called for a radical restructuring of education and an equalisation of educational opportunities. However, the Commission undertook this work during a period of growing unemployment, economic stagnation, food shortages and political dissent over official languages and public administration. For example, an attempt to end the use of English for official purposes was resisted by a number of states, especially those whose Dravidian languages are unrelated to Hindi. The resolution at this time was that the official language of the Union Government would be Hindi, with English used as an auxiliary official language and the 22 states determining their respective official languages. A major step forward in addressing inequality was the 1968 National Policy on Education which created a blueprint for a uniform pattern of primary, upper primary, secondary and tertiary education, with science and maths figuring in its content. The language issue was now resolved through the ‘three language formula’ in which the state language was to be taught, along with Hindi or English, and another modern Indian or foreign language. In most government primary schools, the medium of instruction was the vernacular. If Sri Lanka’s transition to independence had been a relative oasis of calm, its transition to a post-colonial society was anything but. Government education policy became embroiled in regular swings of power between the right- and leftof-centre political parties, punctuated with insurrections and rebellion as in India. The official language of public administration and the language of instruction in education became sources of political tension. The role of English education in conferring privilege, the special status of the grant-aided denominational schools and various educational discriminators were brought into sharp focus. By this time, and in contrast to many parts of India, access to primary education was widespread, and the education concerns of the rural poor focussed increasingly on access to the type of post primary education that provided access to wage and salaried employment, especially in government. In 1956, S.W.R.D. Bandaranaike’s left-wing coalition came to power through the support of the rural masses, rural teachers, ayurveda6 practitioners, Buddhist priests and agricultural labourers. Achievement-based selection to secondary education at 14  years, introduced by the previous government, was abolished. The Official Languages Act of 1956 made Sinhala the sole official language of public administration, essential for recruitment to government jobs. This political move enraged the Tamil-speaking minority and contradicted the justification advanced earlier in 1944 for the introduction of Sinhala and Tamil instruction in all schools: this had stated that both Sinhala and Tamil would become official languages after independence. Over time, English was relegated from its role as

Education for All in India and Sri Lanka 65 a medium of instruction. The medium of instruction was changed in all primary schools in 1948, in all secondary schools from 1953, in University Arts subjects from 1959 and University Science subjects in the 1970s. In India, by contrast, English has never been proscribed as a medium of instruction. The high-status, privileged, fee-charging grant-aided denominational schools survived the change of government in 1956, but they did not survive its gaze. Despite considerable resistance from Protestants and Catholics, two Acts of Parliament (1960 and 1961) vested almost all assisted schools and their properties7 in the state. Legislation was enacted that prevented the establishment of new private schools. Thus: By 1960. . . major colonial policies had been reversed. Language and religion ceased to be a barrier to educational opportunity. The education system was unified with only 63 schools out of over 9,000 schools functioning as private schools . . . The political pendulum swung from time to time and each government attempted to use the educational system to reinforce its policies and to further its political and social goals . . . All political parties, however, attempted to attack poverty through redistributive measures. (Jayaweera 1986, pp. 10–12) The 1961 legislation, with redistribution at its heart, contrasted sharply with the experience of Indian private education and explains in part why private sector schools and fee-charging grant-aided schools have continued to flourish in India (Kingdon 2007; Juneja 2010) but not in Sri Lanka (Aturupane & Little 2020). In Sri Lanka, the hitherto elite schools, most of them grant-aided denominational schools, became fee-free government schools and were required to teach through one of the two national languages, Sinhala or Tamil, not English. They found other ways of maintaining their high status.8 They maintained their privilege by offering education from primary to upper secondary within the same school and guaranteeing access to the secondary grades to those enrolled in primary grade 1. Hence, access to primary grade 1 became very competitive. Children enrolled in schools that offered only a primary education9 were able to seek admission to the higher status schools (known today as the Type 1AB schools) but only after passing a highly competitive grade 5 selection examination. Unsurprisingly, access to which government primary school became a major issue for Sri Lankan parents (Kotalawala 1993; Wijesundara & Perera 2000). Up until this point, ethnicity had not emerged as a source of major political dissent in education in Sri Lanka. Tamil- and Sinhala-speaking children were able to follow their primary and secondary education in their mother tongue. Access issues focussed on quality and schools differentiated by location (rural vs. urban) and socio-economic status. However, ethnicity would emerge as a major source of political tension in the late 1960s, as differential access of Sinhala-medium and Tamil-medium students to prestigious university courses became an issue of contention.

66  Angela W. Little

Instability and the struggle for universal education: 1970–90 From the early 1970s, rising political and economic stability slowed the implementation of egalitarian reforms in both countries. In India, Mrs Gandhi’s new Congress Party (I) was returned to power in 1971 with an increased majority and a mandate to pursue a series of socialist and egalitarian policies. However, a worsening economy, growing social problems and allegations of corruption led to political disorder and the declaration of a state of emergency in 1975. Constitutional change in 1976 moved the subject of education from the state list to the concurrent list, with the consequence that education now became a shared responsibility between the state governments and the Union (or central) government. After Mrs Gandhi’s defeat in 1977, political pressure was put on government by Citizens for Democracy, a civil society group, who called for ‘a radical reconstruction of education under the auspices of a social movement, which would remove the state monopoly and compel it to work towards reforming the system in favour of the poor and deprived’ (Mathur 2001, p. 231). Their report was used to draft a National Policy on Education in 1979. However, the government fell shortly afterwards and the draft policy was consigned to oblivion. Mrs  Gandhi and her Congress Party (I) returned to power in 1980 and set about modernising the economy and society and developing an education policy designed to enhance the role of technology in social transformation. But the government was plagued by insurgencies and communal violence that would lead eventually to the assassination of Mrs Gandhi and the appointment of her son, Rajiv, as Prime Minister. He called for a reconstruction of the national education system, significantly raising the issues of values, unequal opportunities in education, gender biases in education and the role of education in alleviating social tensions. The 1986 National Policy on Education (Government of India 1986a) attracted political goodwill at the highest level of the central government (Sarup 1986). The programme for Universal Elementary Education promoted education equality, a reorientation of content and pedagogy and the improvement of the physical environment of schools. The innovation here was to call for a childcentred approach and the establishment of Minimum Levels of Learning in an effort to encourage both equity and quality in primary teaching and learning (Raina 2002, p. 177). Unlike earlier national policies, the 1986 policy was accompanied by a Programme of Action (Programme of Action 1986b). Operation Blackboard (OB) became the flagship action programme for the improvement of primary education. It advocated that at a minimum, primary schools should have: (i) two reasonably large rooms that are usable in all weather; (ii) necessary toys and games; (iii) blackboards; (iv) maps; (v) charts and (vi) other learning materials (Government of India 1986b). Implementation was slow, with differences between the states in the speed and extent of implementation and the base from which they

Education for All in India and Sri Lanka 67 were developing. The implementation challenges in the face of limited resources were formidable. Across the country 200,000 new posts were created for additional teachers to be placed in one-teacher schools while more than half a million teachers needed to be trained each year. Some of the implementation challenges reflected tensions between the Union Government at the centre and the states. Dyer (2000) suggests that the central Ministry trod a fine line between ‘suggesting’ and ‘prescribing’ actions, in a constrained financial environment and one that imposed additional workloads on the states. The OB scheme reflected a dual and irreconcilable agenda. There was the overt agenda of educational changes; and there was also a more covert agenda of political manoeuvring (Dyer 2000, p. 162). For example, the State Governor of Gujarat regarded OB as a central government scheme and appointed an extra officer to be in charge of centrally sponsored schemes, rather than integrating the scheme within the state system. State officials resented a lack of consultation by the central government and its dictation of where new buildings should be sited. An underlying agenda of the state was to limit intrusion into its preserve (Dyer 2000, p. 163). In Sri Lanka, political instability during the period 1971–90 arose from three insurrections of educated youth, frustrated by inequalities of access to education and government jobs. The strong welfare orientation of government to both health and education had, from the late 1930s, successfully led to reductions in death and infant mortality rates, increases in the rate of population growth and growth in enrolment in primary and junior secondary education. But economic growth was very slow and by 1970 almost 80% of the labour force aged 15–24 was unemployed. Education had become a victim of its own success. The first insurrection, in 1971, was led by the Marxist Janatha Vimukthi Peramuna (JVP) that claimed to represent the interests of poor youth of all communities. The JVP’s opposition to current education policy was less to do with the type of education available to rural children and more to do with the continuing monopoly of the better-off high-status educational opportunities and high-status jobs. The authors of the Education Reforms of 1972 assembled a response to the insurrection in some haste. It involved a plan for a more ‘relevant curriculum’ that would, in principle, prepare the majority of school leavers for their economic futures which, in most cases, lay outside the formal sector of employment. The subject ‘pre-vocational studies’ was introduced in junior secondary education to orient students to practical and locally relevant skills. The GCE, O and A level secondary school examinations were disassociated from their British namesakes to strengthen the integrity of a national education system. Curriculum reforms in primary education were based on the ideas of child-centred and activity-based learning. Rural and urban parents resisted the reforms. Rural parents had little interest in seeing their children prepared for low-status and low-income jobs, while the urban middle classes did not want to see the prospects of their children undermined by the abolition of historical links to British qualification systems. Such

68  Angela W. Little widespread dissatisfaction contributed to the downfall of the government and the reversal of several of the reforms, though not the reforms in primary education (Lewin & Little 1982; Peiris 1983). The second insurrection arose in North and East Sri Lanka. Through the 1970s Tamil groups, concerned about discrimination in access to higher education, were calling for an independent state – Tamil Eelam. Armed combat between the Sri Lankan armed forces and Tamil groups was to last almost 30 years. In an attempt to end the war, the constitution was amended in 1987. This was an outcome of the Indo-Sri Lanka Peace Accord between India’s Prime Minister Rajiv Gandhi and Sri Lanka’s President J.R. Jayewardene that required the devolution of some powers from central government to each of the provinces, a surrender of arms by the Tamils to Indian Peace Keeping Force and a withdrawal of Sri Lankan troops from contested areas. Tamil was upgraded from a national language to an official language and English was deemed to be a ‘link language’. However, the accord did not lead to a cessation of hostilities, the Indian Peace Keeping Force left the country and the war continued. While the 13th amendment of 1987 did not stop the hostilities, it did lead to provincial devolution. Education now appeared on the ‘concurrent’ list, with responsibilities shared between the centre and the province. At first sight this shift appears to be similar to that ushered in by the 1976 constitutional reforms in India, but the direction of travel was in the opposite direction. In India, the shift was from the state governments to the centre; in Sri Lanka, the shift was from the centre to the provinces. In India a constitutional amendment in 1992 introduced a counterbalancing shift in power at the local level, through the panchayati raj institutions which created a new dynamic for educational planning and management in many areas (Govinda & Bandyopadhyay 2008). The third insurrection in 1987 was, like that in 1971, led by the Marxist JVP in the South of the country. Where earlier the JVP claimed to represent the education and job frustrations of Tamil, as well as Sinhala, youth, it was now an increasingly Sinhalese nationalist organisation opposed to the Tamil insurgency in the North and the East. The government, faced now with insurgency on two fronts, met the JVP insurgency with brutal force. Once the Sinhalese unrest in the South was contained in 1989, President Premadasa’s government quickly established a Presidential Commission on Youth to investigate the causes of the insurrection and to create a national education policy unaffected ‘by the vagaries of transient political majorities’ (Government of Sri Lanka 1990). The Commission attributed much of the cause of the insurrection to an ‘irrelevant’ education system, rather than to the state of the economy.

1990  to the present: closing the gap The final phase of this comparative history brings us to 1990 and the gradual closing of the gap in primary and junior secondary enrolments between India and Sri Lanka and the evolving international agenda on EFA, alluded to at the beginning of the chapter.

Education for All in India and Sri Lanka 69 By the early 1990s the massive social mobilisation by the National Literacy Mission in India, launched in 1988, had increased the demand for elementary education and enhanced the role of non-state actors in the provision of elementary education (Govinda  & Bandyopadhyay 2008). The 1992 India National Policy on Education reaffirmed most of the content of the 1986 National Policy. Major national development plans and programmes were launched, with some financial support from external partners. Inter alia, these were the District Primary Education Programme from 1994, the Midday Meal scheme from 1995, Sarva Shiksha Abhiyan from 2000, the Tenth Five Year Development Plan from 2002, the National Programme for Education of Girls at Elementary Education from 2003 (Sarva Shiksha Abhiyan 2010a, 2010b) and the Activity Based Learning programme (ABL) from 2003 (Education Initiatives 2015). The authors of the Indian EFA 2000 Assessment Report and the EFA MidDecade Review described significant growth in access to elementary education since the time of the EFA conference held in Jomtien a decade earlier. At the same time, it acknowledged continuing disparities between the states, between rural and urban schools, caste groups and males and females (Singh 2000; National University of Educational Planning and Administration 2008). The Public Report on Basic Education in India (PROBE) report (1999) offered a more damning picture of educational progress, especially in the four northern states with historically poor records of provision: Bihar, Madhya Pradesh, Rajasthan and Uttar Pradesh. It highlighted large and continuing disparities by region, socioeconomic status, caste, tribe and gender and slow growth in literacy rates. Others pointed to the continuing impact of household resources, parental motivation, returns to child labour, school quality and caste membership on primary school participation, especially in rural areas and among girls (Drèze & Kingdon 2003). In 2002, a further constitutional amendment provided for free and compulsory education of all children in the age group 6–14 years (Government of India 2002). Although compulsory education acts had been on the statutes of many states for many years, some from before independence, these acts had not been formulated in a way that rendered them ‘justiciable’, that is, no one could be prosecuted if those rights were not met (Kabeer et  al. 2003). The 2002 constitutional amendment resulted in a movement for a Right to Education Bill (RTE) led by civil society groups, intellectuals and a small number of bureaucrats working in the central Ministry of Education. Resistance ranged from the private schools who objected to the requirement they reserve places for the poorest to the purists who insisted that all key ingredients for quality education were in place before the Bill should be passed.10 The RTE Bill was eventually passed (Government of India 2009). It enshrined in law the resolution for compulsory and free education that had been debated some 100 years earlier in the Imperial Legislative Council, but was rejected. The United Progressive Alliance held power over two terms, 2004–14 and oversaw a social welfare programme designed to meet the needs of the rural poor excluded from the benefits of the post-1990 programme of economic liberalisation (Deshpande et al. 2017). This rights-based programme included rural

70  Angela W. Little employment guarantees, food subsidies, health insurance as well as education. In 2009, the government launched Rashtriya Madyamik Shiksha Abhiyan (RMSA), in partnership with state governments, in recognition of increasing demands for the universalisation of secondary education in grades 9 and 10. This resonated with the Sri Lankan reform, some 37  years earlier, which had extended nonselective access to grade 9. In Sri Lanka, President Chandrika Kumaratunge came to power in 1994 and quickly established Education Task Forces to take forward the recommendations of the National Education Commission set up in the wake of the 1987–89 youth insurgency. A set of 19 reforms across primary and secondary education, known collectively as ‘The 1997 Education Reforms’ (Government of Sri Lanka 1997) sought, inter alia, to improve curriculum and pedagogy in primary education and to distribute educational opportunities at the junior and senior secondary levels more equitably. Compulsory education between the ages of 5 and 14 was finally enacted, even though grades 1–8 enrolment was near universal.11 The legislation required schools to establish school attendance monitoring committees to ensure not only universal enrolment but also increased rates of attendance. The production of the Reforms text involved a number of tensions between interest groups. First, there was the role of the private sector. The anti-private sector views prevailed and ensured that no recommendations on the expansion of the private sector at the primary or secondary level appeared in the 1997 policy text. Second, there was the role of English. As we saw earlier, post-independence policies had relegated English to a subject of study and had led in turn to the segregation of the education system along language lines, which, de facto, meant ethnic lines (Sinhala and Tamil). Some, including the President, wished to strengthen the role of English as a medium of instruction, while others again appealed to nationalist sentiments and the kaduwa [sword] of language that had cut society into shreds in the colonial past. In India, by contrast, there has been an expansion of private school enrolment. Between 2010 and 2014 the proportion of students in secondary schools grew from 24.4% to 31.4%. Parents believe that private schools provide access to learning English, to high class and caste peers and networks and better job prospects. The growth in private sector enrolment has been uneven, with several states having percentages of private school enrolments in the single digits while others have more than 35% (Lewin 2020). Wealth status is the factor of greatest significance for school choice in favour of private schools. Attendance at a private school is negatively correlated with low caste or tribe, rural residence, poorly educated and employed parents and being a girl. About 80% of boys and 62% of girls in private schools are from the richest two quintiles (i.e. top 40% of income) households (Härmä et al. 2016). By 2000, the Sri Lanka EFA 2000 Assessment reported growth in the provision of early childcare and development programmes, continuing high gross and net enrolment ratios in primary education, a reduction in repetition rates and increases in survival rates to grade 5 and a gender parity literacy ratio of 0.96 (Sri Lanka Ministry of Education and Higher Education 1999).

Education for All in India and Sri Lanka 71 Since the mid-2000s, schemes designed to improve the equitable distribution of educational opportunities have been embedded within an Education Sector Development Framework and Program (ESDFP) prepared in consultation with the Provincial Ministries and Departments of Education, the Ministry of Finance and the Finance Commission. To date, there have been three plans – 2006–10, 2012–16 and 2018–25. Currently, fee-free universal access to the age of 18 years (grade 13, age 17+) is opening up. This stands in contrast to India, where, since 2009, the RMSA programme has aimed to universalise access to grades 9 and 10 (ages 14+ and 15+ respectively) and has yet to extend to grades 11 and 12. As we saw earlier, the establishment of new private schools was proscribed in Sri Lanka from 1961. Currently the percentage of students participating in the government and private sectors of education in India and Sri Lanka vary markedly. Available data suggest that in Sri Lanka only 4.4% of children are enrolled in government-approved private and international schools (grades 1–13); in contrast, in India 33% of enrolments in grade 1 and 34% in grade 10 are in the private unaided sector, albeit with huge variations by state.12 In Sri Lanka the prestigious popular schools are government schools. This unitary system has now extended fee-free universal access to the end of secondary school, grade 13. In India, there is a large social gulf between those who access elementary education through the public system and those who access it through the private system. In 2000, India’s president, Dr K.R. Narayan (2000), described the contradictions of the ‘two Indias’ – the world’s largest pool of highly educated technical personnel and the world’s largest number of illiterates; the world’s largest middle class and the largest number of people below the poverty line. The middle classes disproportionately attend private schools and the poor government schools. In comparison, the majority of Sri Lanka’s children participate in the state-run system of education. Income inequalities are reflected in India’s education system to a much greater extent than in Sri Lanka.13 From the 1990s, and in the wake of the Jomtien EFA conference, both countries attracted increased levels of support from external agencies for the development of primary and junior secondary education. Already by 1990 the Swedish and British governments and UNICEF were involved in primary education programmes at state-level in India. Following the Jomtien conference, more bilateral and multilateral partners, including the World Bank, became involved (Colclough & De 2010; Ward 2011). The experience of state-level programmes was drawn on in the design of the District Primary Education Programme (DPEP) and the nationwide Sarva Shiksha Abhiyan programme, with substantial involvement at the national level. In Sri Lanka too the Swedish government and UNICEF had been involved in primary education programmes since before the time of the Jomtien conference. After Jomtien, however, the British government and others supplemented the efforts of the Swedes, UNICEF supported an array of primary education programmes and the World Bank and the Asian Development Bank supported General Education to the end of junior secondary (Ranaweera 2000; McGillvray et al. 2012).

72  Angela W. Little There was some resistance to the involvement of foreign agencies in both countries, especially the World Bank, but in general the financial and technical support enhanced ongoing work on access and quality (Colclough & De 2010; Wikramanayake 2009). In neither case did the external support appear to determine national policy directly. National policy continued to be driven endogenously. The gap between India and Sri Lanka in terms of enrolments in primary and junior secondary education has closed since 1990. Recent figures indicate universal enrolment in primary education in India and Sri Lanka and adjusted net enrolment ratios in lower secondary of 85% and 95% respectively (UNESCO 2016).

The drivers and interests shaping egalitarian reforms The history of educational equality in primary and junior secondary education in India and Sri Lanka is long. Legislative measures – in both colonial and postcolonial times – have certainly played their part in shaping policies and realities on the ground. But they did not arise in a vacuum. There have been underlying drivers of egalitarian policies and practices which have evolved over time while constellations of interests have appeared and reappeared as the contours of social inequality have changed. During the colonial period, governments had little need to actively promote the education of the poorest social groups. Their main interest was the creation of local English-speaking educated elites to serve government and the church. To varying degrees, they tolerated the work of religious missionaries in providing a religiously oriented curriculum in vernacular languages. In the early twentieth century, the nationalists’ calls for independence and preparation for self-government drove the demand for and design of education schemes for the poor. Calls for and commitments to universal franchise  – in Sri Lanka from 1931, but in India only from 1950 – drove an awareness of the need for education of those who would vote governments in and out of office. At independence, both countries adopted democratic political regimes, with periodic shifts in power between centre right and centre left parties. These shifts in power drove changes in education policy in both countries, with left-leaning regimes focussing much of their policy rhetoric on equalising education opportunities to a greater degree than the right-leaning regimes. These have been some of the underlying drivers for the expansion of education. Political and economic instability have played their part in driving and inhibiting egalitarian reforms. Underlying both have been inequalities between social groups marked, inter alia, by ethnicity, language, caste and economic status. The 1970s were a very turbulent time for India and education reforms were seen by the national government as central for national economic growth and communal integration. The 1970s and 1980s were also turbulent times in Sri Lanka. Three insurrections in the 1970s and 1980s arose out of the frustration of youth inequalities in access to the kind of the education that would guarantee them jobs and generated the imperative for the reforms of 1971 and 1997. In short, in both

Education for All in India and Sri Lanka 73 countries frustration with education and social inequalities drove social conflict which in turn drove education reform. Various interest groups have resisted and promoted egalitarian education reforms. In both countries, landowners resisted improvements in access for children on whose labour they depended, especially in the nineteenth and early twentieth century. Other interest groups who have promoted and resisted egalitarian reforms, to varying degrees at varying times, have included princes, politicians, bureaucrats, elites, groups defined by caste, class, ethnicity, religion and gender, landowners, rural masses, teachers, trade unions, language groups, political parties, religion affiliated schools, private schools and civil society organisations. Interest groups exogenous to the country have also promoted and resisted egalitarian reforms at the primary and junior secondary levels, including, inter alia, colonial government official and religious missionaries. More recently, bilateral and multilateral organisations, and a range of private organisations based in other countries, have played a role in primary and junior secondary education. Different social groups have been influenced to varying degrees by these drivers and interests. The case of the children of plantation workers in Sri Lanka illustrates the point. Historically, the education status of the tea and rubber plantation community has lagged far behind that of children living in surrounding rural areas and the towns.14 A long view of educational progress among this social group indicates the varying dominance of economic, political and social drivers and interests over time. In the mid-nineteenth century the religious goals of missionary organisations and philanthropic pressures on a colonial government drove the establishment of a network of schools offering rudimentary education. In the early to mid-twentieth century political interests in both the UK and India called for improvements in labour conditions in the plantations, including the education of labourers’ children. More general political improvements, including the growth of franchise and the trade unions and the expansion of education in urban and rural education, also played their part. In the late twentieth century the politics of trade unions, the economic decline of the plantation industry, the generation of a plantation labour surplus and the inter-ethnic politics of the civil war all played their part in increasing the expansion of education opportunity among the children of plantation workers (Little 1999). The stories of educational change among other social groups in India and Sri Lanka will, in similar fashion, display a blend of local, national and international drivers and interests.

Into the future Despite enormous progress over the past century, social inequalities remain in both countries at the primary and junior secondary level. In India much more needs to be done to reduce the education inequalities experienced by, inter alia, Dalits, Adivasis,15 child labourers, girls and children with disabilities (Brahme et al. 2018; Singal 2016; Ramachandran 2020). In Sri Lanka more needs to be done for those living in disadvantaged households in estates, rural and urban areas, those attending Type 1C, 2 and 3 schools, those studying in the Tamil

74  Angela W. Little medium, boys and those with disabilities (Jayaweera et al. 2013; Aturupane & Little 2020). In the introduction to this chapter I described both countries as multilingual, multi-ethnic and multi-religious. However, the degree of diversity between them varies. India has 23 official languages; Sri Lanka has two. India has many hundreds of mother tongue languages; Sri Lanka has six. India has at least nine main religions, with the majority of the population following Hinduism. Sri Lanka has four main religions with the majority of the population following Buddhism. India’s democratic regime has had to contend with a greater degree of diversity than Sri Lanka’s. The question arises: has the greater degree of diversity in India, resulted in a greater degree of competition and conflict between a larger number of interest groups and slowed the ability of her governments at all levels to introduce and implement policies designed to reduce inequalities in education? And if so, will this continue to be the case in the future? In 2016, India and Sri Lanka recorded near universal enrolment in primary education and adjusted net enrolment ratios in lower secondary of 85% and 95% respectively (UNESCO 2016). We would expect these figures to increase adult literacy rates in the future. Two questions arise: as enrolment at the base of the system expands, will the education level that determines the best life chances simply escalate up the educational ladder in both countries, rather than opening up good life chances for all? If so, will the social composition of state and private enrolments have promoted or thwarted the ability of the education systems in each country to increase social equality in the longer term?

Notes 1 Formerly Ceylon, the name was changed in 1972 to Sri Lanka, which will be used throughout. 2 For examples of underlying drivers and interests, in various contexts, see Grindle 2004; Jayasuriya 1979; de Silva 1981; Kumar 2005; Leftwich 2006; Little 1999, 2008, 2010a, 2010b; Kingdon et al. 2014; Deshpande et al. 2017. 3 This chapter does not address the impact of reductions of inequalities of education access on reductions in economic, social and political inequalities in the longer term. Nor does it address policies at the sub-national level and policies for the education of elites and the middle classes. 4 For example, Isenman 1980; Sen 1981; Weiner 1991; Kingdon 2007. Several commentators note the differences in education progress between the left-leaning states of Kerala, Tamil Nadu and West Bengal on the one hand and the BIMARU states of Bihar, Madhya Pradesh, Rajasthan and Uttar Pradesh on the other. 5 Sinhala and Tamil medium primary schools were fee-free already. 6 Ayurveda is a system of medicine with historical roots in India. 7 A small number opted to become fully private or unaided and non-fee-levying. 8 Many today are known as the Type 1AB schools. 9 At this time primary education spanned grades 1–6. Later it was grades 1–5. 10 Interviews with the author 2008–09. 11 The 1939 Education Ordinance that prescribed legislation on compulsory education had never been enshrined in law. 12 For Sri Lanka see Aturupane & Little 2020; for India see Härmä et al. 2016. The India figure excludes enrolments in the private grant-aided schools.

Education for All in India and Sri Lanka 75 13 Although much of the recent growth in private education in India has been in the low-cost sector, enrolments in primary and lower secondary government schools in India are biased heavily towards the poorest groups. In Sri Lanka enrolments at these levels are well distributed across all income groups. 14 In 1986–87, the adult literacy rate of Indian Tamils in the plantations was 69%, compared with 89% in rural areas and 88% in the country as a whole, these differential rates reflecting differential education access over the previous generations. In contrast with rural rates across India, these rates were high. 15 Dalits are disadvantaged persons from Scheduled Castes: Adivasis are disadvantaged persons from Scheduled Tribes.

References Aturupane, H., & Little, A.W. (2020) General education in Sri Lanka. In Sarangapani, P.M.,  & Pappu, R. (Eds.) Handbook of Education Systems in South Asia. Global Education Systems. Singapore: Springer. Basu, B.D. (2012) History of Education in India Under the Rule of the East India Company. London: Forgotten Books. Benavot, A., & Resnick, J. (2007) Lessons from the past: A comparative socio-historical analysis of primary and secondary education. In Cohen, J.E., Bloom, D.E., & Malin, M.B. (Eds.) Educating All Children a Global Agenda. Cambridge, MA: The MIT Press. Brahme, M., Babu, M.S.,  & Műller, T. 2018 (Eds.) Inclusive Education in India: Concepts, Methods and Practice. New Delhi: Mosaic Books. Colclough, C.,  & De, A. (2010) The impact of aid on education policy in India. International Journal of Educational Development, 30(5), pp. 497–507. Deshpande, R., Kailash, K.K., & Tillin, L. (2017) States as laboratories: The politics of social welfare policies in India. India Review, 16(1), pp. 85–105. De Silva, K.M. (1981) A History of Sri Lanka. London, Berkeley and Los Angeles: C. Hurst and Company and University of California Press. Drèze, J,  & Kingdon, G.G. (2003) School participation in rural India. Review of Development Economics, 5(1), pp. 1–24. Dyer, C. (2000) Operation Blackboard: Policy Implementation in Indian Elementary Education. Oxford: Symposium Books. Education Initiatives (2015) Evaluation of Activity Based Learning as a Means of Child Friendly Education. [Online] Final Report to UNICEF, New Delhi. Available from: www.unicef.org/evaldatabase/files/EI-UNICEF_Evaluation_of_ABL_as_a_ means_of_child-friendly_education-Final_Report_India_2015-024.pdf [Accessed 2 June 2020]. Government of India (1950) The Constitution of India. [Online] New Delhi. Available from: www.india.gov.in/my-government/constitution-india/constitutionindia-full-text [Accessed 30 July 2020]. Government of India (1986a) National Policy on Education 1986. [Online] New Delhi: Ministry of Human Resource Development, Government of India. Available from: https://mhrd.gov.in/sites/upload_files/mhrd/files/upload_document/npe.pdf [Accessed 2 June 2020]. Government of India (1986b) National Policy on Education 1986: Programme of Action. New Delhi: Ministry of Human Resource Development, Government of India

76  Angela W. Little Government of India (2002) The Constitution (Eighty Sixth Amendment Act) Article 21-A. [Online] Available from: www.mhrd,gov.in/rte [Accessed 28 November 2019]. Government of India (2009) The Right of Children to Free and Compulsory Education (RTE) Act. [Online] Government of India, New Delhi. Available from: https:// mhrd.gov.in/sites/upload_files/mhrd/files/upload_document/rte.pdf [Accessed 1 June 2020]. Government of Sri Lanka (1990) Report of the Youth Commission. Colombo: Government of Sri Lanka. Government of Sri Lanka (1997) The 1997 Education Reforms. Maharagama: Presidential Task Force on Education. Govinda, R., & Bandyopadhyay, M. (2008) Access to Elementary Education in India: Country Analytical Review. [Online] Falmer: CREATE. Available from: https:// assets.publishing.service.gov.uk/media/57a08bb7ed915d3cfd000ec0/India_ CAR.pdf [Accessed 2 June 2020]. Grindle, M. (2004) Despite the Odds: The Contentious Politics of Education Reform. Princeton, NJ: Princeton University Press. Härmä, J., Lewin, K., Siddhu, G.,  & Jha, S. (2016) The Shifting Terrain of Public and Private Provision. [Online] Rashtriya Madhyamik Skiksha Abhiyan Technical Cooperation Agency Research Report VI, Ministry of Human Resource Development, New Delhi. Available from: https://keithlewin.net/wp-content/ uploads/2016/05/6.-The-Shifting-Terrain-of-Public-and-Private-Provision.pdf [Accessed 2 June 2020]. Isenman, P. (1980) Basic needs: The case of Sri Lanka. World Development, 8, pp. 237–258. Iyer, L. (2008) Direct versus Indirect Colonial Rule in India: Long Term Consequences. [Online] Harvard Business School Working Paper No. 05–041, Boston, MA. Available from: www.hbs.edu/faculty/Publication%20Files/05-041_1feff996-f50e-4e5ab057-e0119cd19a62.pdf [Accessed 2 June 2020]. Jayasuriya, J.E. (1979) Educational Policies and Progress. Colombo: Associated Educational Publishers. Jayaweera, S. (1986) Educational Policies and Change – from the Mid-Nineteenth Century to 1977. Maharagama: National Institute of Education. Jayaweera, S., Gunawardene, C., Shanmugam, T., & Colombage, S. (2013) Out of School Children in Sri Lanka. Colombo: UNICEF. Jodhka, S.S. (2002) Nation and village: Images of rural India in Gandhi, Nehru and Ambedkar. Economic and Political Weekly, 37(32), pp. 3343–3353. Juneja, N. (2010) Access to What? Access, Diversity and Participation in India’s Schools. CREATE Research Monograph No 32. Delhi/Sussex: CREATE. Kabeer, N., Nambissan, G., & Subrahmanian, R. (Eds.) (2003) Child Labour and the Right to Education in South Asia: Needs versus Rights? New Delhi: Sage. Kingdon, G.G. (2007) The progress of school education in India. Oxford Review of Economics, 23(2), pp. 168–195. Kingdon, G.G., Little, A.W., Aslam, M., Rawal, S., Moe, T., Patrinos, H., Beteille, T., Banerji, R., Parton, B., & Sharma, S K. (2014) A Rigorous Review of the Political Economy of Education Systems in Developing Countries. [Online] EPPI Centre Report No. 2203. Available from: https://eppi.ioe.ac.uk/cms/Portals/0/ PDF%20reviews%20and%20summaries/Political%20economy%202014Kingdon. pdf?ver=2014-04-24-141259-443 [Accessed 2 June 2020].

Education for All in India and Sri Lanka 77 Kotalawala, D.E.M. (1993) Research on Year Five Scholarship Examination. Maharagama: National Institute of Education. Kumar, K. (2005) Political Agenda of Education: A Study of Colonialist and Nationalist Ideas. 2nd ed. New Delhi: Sage. Leftwich, A. (2006) From Drivers of Change to the Politics of Development: Part 3. London: DFID. Lewin, K.M. (2020) Realising rights to education: Is privatising educational services equitable, democratic, inclusive and sustainable? In Varghese, N.V., & Bandyopadhyay, M. (Eds.) Education, Democracy and Development. New Delhi: Shipra Publications. Lewin, K.M., & Little, A.W. (1982) Examination reform and educational change. In Sri Lanka, 1972–1982: Modernisation or Dependent Underdevelopment? [Online] Institute of Development Studies, Discussion Paper 180 Brighton. Available from: http://angelawlittle.net/wp-content/uploads/2012/06/ExaminationReformandEducationalChange1984.pdf [Accessed 2 June 2020]. Little, A.W. (1988) Learning from Developing Countries. Inaugural Professorial Lecture, Institute of Education, University of London. Little, A.W. (1999) Labouring to Learn: Towards a Political Economy of Plantations, People and Education in Sri Lanka. London: Palgrave Macmillan. Little, A.W. (2008) EFA Politics, Policies and Progress. CREATE Research Monograph No. 13. London and Sussex. Little, A.W. (2010a) The Politics, Policies and Progress of Basic Education in Sri Lanka. [Online] Consortium for Research on Education Access, Transition and Equity (CREATE), Research Monograph No. 38. Available from: www.create-rpc.org/ pdf_documents/PTA38.pdf [Accessed 2 June 2020]. Little, A.W. (2010b) Access to Elementary Education in India: Politics, Policies and Progress. [Online] Consortium for Research on Education Access, Transition and Equity (CREATE), Research Monograph No. 44. Available from: https://files. eric.ed.gov/fulltext/ED513316.pdf [Accessed 2 June 2020]. Mathur, K. (2001) Does performance matter? Policy struggles in education. Journal of Educational Planning and Administration, 15(2), pp. 225–240. McGillvray, M., Carpenter, D.,  & Norup, S. (2012) Evaluation Study of Longterm Development Cooperation between Sri Lanka and Sweden. [Online] Swedish International Development Cooperation Agency (SIDA) Evaluation 2012:1. Available from: www.sida.se/English/publications/110570/evaluation-study-oflong-term-development-co-operation-between-sri-lanka-and-sweden/ [Accessed 2 June 2020]. Moore, R.J. (1965) The composition of ‘Wood’s Education Despatch’. The English Historical Review, 80(314), pp.  70–85. Available from JSTOR: www.jstor.org/ stable/560738. [Accessed 19 November 2020]. Naik, J.P. (1966) Elementary Education in India: The Unfinished Business. London: Asia Publishing House. Narayan, K.R. (2000) The President Speaks. [Online] President K.R. Narayanan’s address to the nation on the eve of Republic Day, January 25. Available from: www. india-seminar.com/2000/487/487%20narayanan.htm [Accessed 2 June 2020]. National University of Educational Planning and Administration (2008) The MidDecade Review of EFA. New Delhi: UNESCO. Oesterheld, J. (2007) National education as a community issue: The Muslim response to the Wardha scheme. In Kumar, K., & Oesterheld, J. (Eds.) Education and Social Change in South Asia. New Delhi: Orient Longman.

78  Angela W. Little Peiris, K.M. (1983) Tiny Sapling: Sturdy Tree. Oslo: Universitetsforlaget. Public Report on Basic Education in India (PROBE) (1999) Public Report on Basic Education in India. Oxford: Oxford University Press. Raina, V. (2002) Decentralization of education. In Govinda, R. (Ed.) India Education Report: A Profile of Basic Education. Oxford: Oxford University Press. Ramachandran, V. (2020) Democratic inequalities: The dilemma of elementary education in India. In Varghese, N.V., & Bandyopadhyay, M. (Eds.) Education, Democracy and Development. New Delhi: Shipra Publications. Ranaweera, M. (2000) Donors and primary education. In Little, A.W. (Ed.) Primary Education Reform in Sri Lanka. Colombo: Educational Publications Department. Roy, T. (2014) Geography or politics? Regional inequality in colonial India. European Review of Economic History, 18(3), pp. 324–348. Sarup, A. (1986) A retrospective view of formulating a new education policy. Bulletin of the UNESCO Regional Office for Education in Asia and the Pacific, 27, pp. 121–134. Sarva Shiksha Abhiyan (2010a) A Programme for Universal Elementary Education Framework for Implementation. [Online] Available from: www.educationforallinindia.com/ssa.htm [Accessed 2 June 2020]. Sarva Shiksha Abhiyan (2010b) National Programme for Education of Girls at Elementary Level (NPEGEL). Available from: http://interactions.eldis.org/programme/ national-programme-education-girls-elementary-education-npegel [Accessed 19 November 2020]. Sen, A. (1981) Public action and the quality of life in developing countries. Oxford Bulletin of Economics and Statistics, 43(4), pp. 287–319. Shani, O. (2018) How India Became Democratic: Citizenship and the Making of the Universal Franchise. Cambridge: Cambridge University Press. Singal, N. (2016) Education of children with disabilities in India and Pakistan: Critical analysis of developments in the last 15 years. Prospects, 46, pp. 171–183. Singh, A. (2000) The Education for All 2000 Assessment. National Institute of Educational Planning and Administration India, Delhi. Available from: https://unesdoc. unesco.org/ark:/48223/pf0000220053 [Accessed 30 July 2020]. Sri Lanka Ministry of Education and Higher Education (1999) Education for All: The Year 2000 Assessment Country Report. [Online] Available from: https://unesdoc. unesco.org/ark:/48223/pf0000219519 [Accessed 2 June 2020]. UNESCO (2016) Education for People and Planet. Global Education Monitoring Report, 2016. [Online] Available from: https://unesdoc.unesco.org/ark:/48223/ pf0000245752 [Accessed 2 June 2020]. Ward, M. (2011) Aid to education – the case of Sarva Shiksha Abhiyan in India and the role of development partners. Journal of Educational Policy, 26(4), pp. 543–556. Weiner, M. (1991) The Child and the State in India. Princeton, NJ: Princeton University Press. Wijesundara, S., & Perera, W.J. (2000) Teacher perceptions of the national policy on primary education. In Little, A.W. (Ed.) Primary Education Reform in Sri Lanka. Battaramulla: Educational Publications Department. Wikramanayake, D.H. (2009) Education and Economic Globalisation: An Analysis of National Policy Formation in Sri Lanka. Unpublished PhD dissertation, University of Sydney. Zhang, T. (2005) Literacy Education in China. [Online] Background Paper prepared for the Education for All Monitoring Report. UNESCO. Available from: https:// unesdoc.unesco.org/ark:/48223/pf0000146208 [Accessed 2 June 2020].

5 Public–private partnerships in education Do they offer an equitable solution to education in India and Pakistan? Monazza Aslam and Geeta Gandhi Kingdon1 In a global context where world leaders have adopted ambitious plans to ensure ‘inclusive and equitable quality education and (promoting) lifelong learning opportunities for all’ (Sustainable Development Goal (SDG) 4),2 it is noticeable that large numbers of children in many countries continue to face numerous challenges in accessing education. Despite expansion of access, some children continue to be denied access to school and, for those who do, large numbers of children face substandard teaching and poorly provisioned classrooms. As a result, more than half the children and adolescents worldwide (58%) are failing to meet the minimum proficiency standards in reading and mathematics.3 The South Asia region houses a large number of out-of-school children. Kingdon (2017), for example, notes in India that, whilst the percentage of out-ofschool 6–10-year-olds has declined from 3.7% in 2009 to 2.8% in 2014, these India-wide percentages mask wide disparities across regions in access to schooling. Whilst there have been large declines in out-of-school children in Uttar Pradesh, for example, 3.8% of all 6–10-year-olds continued to be out of school in 2014. The various Annual Status of Education Report (ASER) surveys in Pakistan conducted annually by a local NGO4 since 2009 also consistently report striking numbers of out-of-school children. Some children are more vulnerable than others: in the South Asia region these tend to include girls, children with disabilities, those living in more remote or rural geographies or children belonging to more disadvantaged backgrounds (i.e. from lower socio-economic status or from lower castes or from religious and ethnic minorities). ASER Pakistan (2018) findings, for instance, show that more than 16% of the children aged 6–16  years across rural Pakistan are out of school or have dropped out. Amongst these children, 10.8% of those aged 6–10 years have never enrolled in school. The largest number of out-of-school girls are reportedly in rural Balochistan and newly merged districts in Khyber Pakhtunkhwa (KP) where 16.8% and 17.4% of girls aged 6–16 years are out of school. Data also reveal that almost a quarter of all children aged 6–16  years reporting some form of disability in Pakistan have either never been enrolled in school or have dropped out. These large inequalities in access, especially for the most vulnerable and marginalised children, continue to pose huge challenges to these two countries.

80  Monazza Aslam and Geeta Gandhi Kingdon The ultimate responsibility for ensuring that children – from all socio-economic backgrounds, from all religious or ethnic denominations, of all genders and ability groups and whether they are from rural or urban locations – have access to education lies with national governments. However, financially constrained governments faced with ever-increasing school-age populations have faced numerous challenges in achieving these goals.5 The simultaneous rise of the private sector as well as the emergence of numerous non-state providers in many parts of the world, including in South Asia, has presented an opportunity to severely constrained governments to partner with these providers to reach more students, often at substantially lower costs. Amongst the various types of providers, marketoriented (for profit) schools tend to be dependent on user fees for some or all of their running costs and tend to have a degree of independence from the state. Their business models, therefore, rely on successfully attracting and retaining feepaying students. Arguably, such models can engender more efficiency with many low-fee private schools keeping their operational costs substantially lower (by paying very low teacher salaries and using other resources frugally). At the same time, the evidence from a range of different contexts appears to suggest that they can provide the same levels of learning outcomes, if not better, than those offered by government schools (see summary of evidence in Day Ashley et  al. 2014). However, critics of private provision argue that as profit is a key motive for the provision of this type of education, partnership models (in which governments partner with private providers) may result in ‘non-profitable’ students, typically those who are the most marginalised and served only by the government sector to be further excluded. This so-called cream-skimming of more able students in partnership models may further exacerbate existing inequalities in education systems. In light of these issues, a key question this chapter aims to answer is whether government partnerships with private providers in some form of Public–Private Partnership (PPP) arrangement to improve access to education or to achieve better quality education is reinforcing or even exacerbating inequalities that exist in the system. Colclough’s (1996) research exploring education and the market was a seminal piece that brought the aforementioned questions to the forefront of discussion and debate. In exploring the role of the market and specifically the role of the private sector (and user fees) in education – the neoliberal solution – Colclough came to the conclusion that private schooling can provide a solution for financially constrained governments but only under tightly defined circumstances. However, in the two and a half decades since this work was published, the debate has certainly progressed. The objective of this chapter is to shed light on some of these new pieces of evidence using India and Pakistan as case studies. Our aim in this chapter6 is to provide first a critical exploration of key evidence on the nature and evolution of private and non-state provision and PPP arrangements between governments and different types of non-state and private providers. We do this in the section that follows. This section seeks more insight into the national and international debates associated with such governments partnering

Public–private partnerships in education 81 with private and non-state actors. It also discusses the roles and responsibilities of various actors entering partnerships and the extent to which the provision of an enabling environment for such arrangements is needed for the arrangements to work. After this, we use exemplary evidence from India and Pakistan to discuss the evolution of the private sector and various types of PPP arrangements in these two countries. The subsequent section summarises key pieces of evidence from these two countries to discuss whether PPP arrangements offer an equitable solution to the countries’ education problems. The final section reflects on what we learn about PPPs in relation to equity in school education, the contested role of private schools and PPPs and the factors which affect the effectiveness of PPPs in education.

Private provision and PPP arrangements globally: the state of the evidence Education provision by private and non-state actors has shown a dramatic increase in the last few decades. Taken as a whole, of the children enrolled in primary schools in the world, 9.7%were reportedly enrolled in private schools in 1970. This figure had increased to 18% by 2015, a doubling over during this time period.7 Data also suggest that private and non-state actors cater to ever-increasing populations in contexts as varied as South and West Asia, Latin America and the Caribbean and sub-Saharan Africa and the Arab states (Elacqua et  al. 2018). Therefore, the rise of the private and non-state sector is a worldwide phenomenon and not one limited to Western and more developed economies. For example, the rapid rise of private enrolments across Africa that has most recently been documented noted that 21% of children and youth are currently being educated in the private sector with this predicted to increase to 25% by 2021 (CAERUS 2016).8 These are not small numbers by any means – millions of children are studying in fee-charging schools or other forms of non-state provision (e.g. those run by NGOs, trusts or religious organisations) across the globe. Though often collated into an umbrella term of non-state or private, these providers are by no means homogenous. Non-state provision encompasses a variety of models and recent literature has noted that not only are there numerous hybrids of non-state provision but also blurred boundaries across the various categorisations (Day Ashley et al. 2014). In particular, recent reviews of literature summarising the evidence on non-state schooling also agree that there is typically a lack of an agreed set of definitions and limited information on providers (Day Ashley & Wales 2015). Providers differ in terms of their scale and scope of education provision as well as in terms of their management structures, financing arrangements, their relationship with the government and the extent of regulation and independence they face, etc. (Aslam 2017). It is, therefore, fair to say that the term non-state encompasses a wide spectrum of providers with varying modalities, diverse objectives and ultimately with a varied ability to impact on equity in access and learning (ibid.).

82  Monazza Aslam and Geeta Gandhi Kingdon This rise of the non-state sector has also been coupled with another phenomenon – the decision by many governments around the world to partner with the non-state sector to deliver education in the form of PPPs. Rising child populations, resource constraints and pressures to provide quality education to all have resulted in the increasing use of such arrangements across the world. These partnerships are premised on the belief that PPPs maximise the advantage offered by each sector and, in doing so, are able to cater to ever-increasing school populations in the most efficient manner possible. It is argued that increasing access, improving quality and delivering education in the most cost-effective manner are among the goals that governments which enter into such relationships aim to achieve (Patrinos et al. 2009). These partnership arrangements, therefore, offer a theoretical opportunity for governments to combine the potential reductions in inequality offered by the public financing of education with the efficiencies of private schooling (Barrera-Osorio & Raju 2015). The educational space in any given context can range from 100% public provision (with the government providing, financing and regulating education fully) to 100% private provision (i.e. where the private sector provides, finances and regulates education fully). In reality, most education systems in the world combine different types of contractual arrangements, ownership structures and funding arrangements. Table  5.1 summarises the different models of education provision that can theoretically exist in educational spaces including both

Table 5.1  Typologies of state and non-state providers A • • • • • • C • • • • • • •

Private provision and regulation; B private finance Private schools • Affordable private schools Home schooling • Non-subsidised non-governmental organisation schools/learning centres • Non-subsidised community schools • Non-subsidised religious schools

State provision; private finance – state regulation on varied forms School fees or tuition fees in state schools Individual philanthropy to support state schools Corporate social responsibility Private sponsorship of state schools

Private provision; public finance – D State provision and regulation; state regulation in varied forms public finance Vouchers for private schools • State schools, without fees State subsidies or scholarships for private schools Education service contracts/charter schools Private management of public schools State-subsidised non-governmental organisation schools/learning centres State-subsidised community schools State-subsidised religious schools

Source: Aslam (2017, p. 4)

Public–private partnerships in education 83 public only and private/non-state provision only as extremes but also different types of PPP arrangement that might emerge as a result of partnership between the state and the private sector. Amongst the most emblematic of PPPs are perhaps the education voucher schemes, loans and scholarships that aim to provide funding directly to children in order to increase their choice of schooling. Alternative PPP models exist in the form of contract or charter schools (schools that are owned and funded by the public sector but managed by the private sector). Another partnership arrangement can take the form of a government subsidy to a private or non-state provider with the expectation that this might take differing forms depending on the programme. This latter arrangement can also differ in whether the government provides a per-student subsidy or an unconditional grant (e.g. a block grant which remains flat and constant irrespective of changes in enrolment). The evidence now confirms that PPPs are widespread across Southern and Northern contexts (see Aslam et  al. 2017). The most common forms of educational PPPs have ranged from contracts relating to infrastructure, construction and management of schools to the provision of educational services and operations as a whole, for example, through voucher schemes or charter schools. These contracts typically outline how the government will fund non-state providers to supply an educational service of a defined quantity and quality for a specific period of time. The terms share the risk across the two sectors and also tend to include specified performance targets as well as sanctions for non- or poor performance (ibid.). Theoretically, the state partnering with the private sector offers the potential for increases in enrolment, greater equity of access to quality education as well as improvements in learning outcomes. In reality, these partnerships are complex, require important enabling conditions to be in place for them to be effective and have, sometimes, generated unintended consequences. The extent to which any partnership scheme will deliver on its theoretical promise is also limited by the availability of places as well as the quality of instruction within the schools (Patrinos et al. 2009). Proponents nevertheless note that different types of PPP arrangements offer the potential for reducing education inequalities at substantially lower costs, provided they are well targeted and well designed (ibid.). The rise of private schooling and government partnerships with the private sector globally has generated intense debate on the extent to which the private/ non-state sector offers an equitable solution to growing enrolment numbers as well as better quality education than purely government provision. The more recent evidence has been usefully summarised by Day Ashley et al. (2014) (on private schools), Wales et  al. (2015) (on religious and philanthropic schools), Aslam et al. (2017) (on PPP arrangements globally) and Aslam and Rawal (2018) (on private schooling and PPP arrangements in sub-Saharan Africa at the secondary level). A review of the evidence by Results for Development (R4D) (2017) focuses specifically on the role of the non-state and private sector in conflict and crisis settings. Overall, the evidence is indicative of improvements in learning outcomes in certain types of non-state provision, but this is caveated by the very low overall

84  Monazza Aslam and Geeta Gandhi Kingdon learning outcomes across education systems and, therefore, questions being raised about whether the so-called better learning outcomes amongst certain types of non-state providers mean much when the threshold of learning is so low (Day Ashley et al. 2014). Concerns have also been raised about the rise of non-state provision crippling the idea that education is a basic human right and in increasing marginalisation and inequality, the entire idea of privatisation (and any resultant partnering of the government with this sector) runs counter to all notions of human rights law (ibid.). There is evidence of certain types of nonstate education providers being able to reach the marginalised and disadvantaged more effectively but questions exist with regard to their sustainability (Wales et al. 2015). There is also evidence of certain types of non-state actors being able to achieve better learning outcomes than their state school counterparts (moderate evidence9 for low-fee private schools and philanthropic schools) and of certain types of arrangements with governments (subsidies to non-state actors) broadly suggestive of the positive relationship these arrangements have on educational quality (also see Day Ashley et al. 2014). The evidence on equity in access to quality education is, however, more ambiguous and mixed and differs depending on the type of non-state provider. Whilst there is some evidence that girls are less likely to access fee-charging private schools across a variety of contexts (Day Ashley et al. 2014; Datta & Kingdon 2019), the evidence is more ambiguous on fee-charging private schools being able to reach the poor and on the poor’s ability to pay for them (ibid.). The evidence of philanthropic schools (and to some extent religious schools) reaching the poor and marginalised is strong, with the evidence being more moderate with respect to their ability to target girls and achieve gender parity. Evidence on the affordability of these providers for parents is weaker and more inconclusive (Wales et al. 2015). Whilst different types of arrangements may work in different contexts, the critical factor remains the government’s ability both to foster an enabling environment and also combine it with effective legislation, monitoring and regulation to ensure quality education provision that reaches the most marginalised (e.g. the disabled, street children, etc.). The evidence globally on different types of PPP arrangements which are intended to improve access and learning outcomes across developing contexts has been reviewed by Aslam et al. (2017) and by Aslam and Rawal (2018) focusing specifically on secondary education in sub-Saharan Africa. The former review found mixed evidence on the ability of one type of PPP arrangement – charter schools (which combine private management with public funding) – to improve learning outcomes with only emerging evidence that they are able to reach the more disadvantaged in the contexts in which there are robust studies evaluating such arrangements. However, the review suggests that there are: [s]ome indications of the advantages of this type of arrangement, not only in terms of improved learning outcomes but also with respect to other educational aspects, such as increased enrolment and better management practices. Robust evidence on whether these schools directly benefit the poorer

Public–private partnerships in education 85 quintiles is very limited, but emerging evidence appears to suggest that contract schools may be able to reach the more disadvantaged in certain contexts (e.g. Colombia). (Aslam et al. 2017, p. 20) The evidence of PPP arrangements where the government subsidises a private or faith-based organisation is more ‘weakly positive’ with suggestions of ‘potential benefits in a government subsidising private schools to improve outcomes and reach the more disadvantaged’ (Aslam et  al. 2017, p.  24). However, the authors acknowledge the methodological drawbacks of many studies that form the basis for this conclusion. Aslam et al. (2017) also note the important role that subsidised schools can often play in providing education in contexts of conflict and crisis. This is particularly true in contexts of acute crisis and armed conflict where basic needs are not met and where non-state schools have been found to play an important role by filling gaps in education provision (R4D 2017). The authors’ key conclusion of their review is that state-subsidised schools which tend to serve more disadvantaged children and focus on poor rural areas, since they have a large market at the primary education level, tend to perform at least as well as government schools (in terms of learning outcomes) in such contexts. Hence, there is a strong rationale for this form of partnership arrangement (Aslam et al. 2017). Voucher arrangements provide an alternative PPP model which typically involves a government-funded tuition coupon being provided directly to students’ parents. In voucher schemes, a student’s parents receive a governmentfunded tuition coupon which can be redeemed at eligible public or private schools of their own choice. In theory, these schemes have the potential significantly to improve enrolment by expanding private choice, especially to those who might otherwise not be able to afford these spaces – and so are sometimes argued to be a means to tackle inequities that have been observed in who gains access to private schools. In reality, some have argued that voucher schemes may simply result in a reallocation of enrolment between private and public sector spaces with little gain in overall enrolment for the disadvantaged (Patrinos et al. 2009, p.  32). The dominant rationale for voucher schemes is their supposed impact on improving learning outcomes through the creation of competition between schools to attract and retain children and their greater accountability towards parents (since voucher parents in essence behave like fee-paying parents). Voucher programmes have been used across various contexts. Chile presents an example of a context where a universal voucher programme at the elementary and secondary levels of education has been operational since 1981 and has been extensively researched. Aslam et al.’s (2017) review, for example, notes that much of the evidence of voucher programmes comes from this Chilean context and this evidence on whether vouchers improve learning outcomes remains mixed and controversial. In some instances, the Chilean evidence has shown these schemes to have increased social stratification and inequities in the education system. Evidence from Pakistan and India on the impact of voucher schemes is also mostly

86  Monazza Aslam and Geeta Gandhi Kingdon mixed and only weakly positive in terms of measured improvements in learning outcomes (see the following section). In summary, the evidence on the role of private schools and non-state providers in improving outcomes and doing so in an equitable manner is mixed.

The evolution of private schooling and PPPs in India and Pakistan Having reviewed the global trends, we now focus in more detail on the experience in India and Pakistan. These are important countries because of their size. The non-state sectors in India and Pakistan are both large and seem to be here to stay. To illustrate, evidence from India’s official data from the District Information System on Education (DISE) shows that over the short six-year period from 2010–11 to 2016–17, enrolment in public schools fell by 18.3 million students and enrolment in private unaided schools rose by 17 million students (Kingdon 2017). The remaining 1.3  million students who left government schools but did not join the recognised private unaided schools could have joined either the unrecognised private unaided schools which are mostly not captured in DISE data, or they could conceivably have gone to private aided schools, though the latter is unlikely as their enrolments have not changed significantly. In several Indian states, nearly 50% or more of the elementary school-age children are attending private unaided schools (Kingdon 2017). Pakistan has seen a similar expansion of the non-state sector, particularly the rise of low-fee private schools (Andrabi et al. 2008; ASER reports 2010–2018). It is clear that the non-state schooling sector is not an urban elite phenomenon; data support the view that these schools exist also in rural, often even remote, areas in both countries. In examining the size of the private schooling sector in India, Kingdon (2017) has shown a steep growth in private schooling and a rapid shrinkage in the size of the government school sector, something she terms ‘parental abandonment’ of government schools. Her analysis of the latest India data shows that a large majority of private schools in most states are ‘low-fee’ when judged in relation to state per capita income, per-pupil expenditure in government schools and the officially stipulated minimum wage for daily wage labour. She argues that this suggests that affordability is an important factor behind the migration of children from government to private schools  – the apparent exodus from government schools and the dash to private schools might be partly to do with increasing affluence and partly to do with the desire for ‘English-medium’ education, but it may also reflect parental perception that the private schooling sector provides better quality education. Such a perception may be based on the observation that private school teachers display higher effort (lower teacher absence rate) than in public schools and the observation that they apparently produce higher learning levels. This is borne out in the Annual Status of Education Report (ASER) surveys. In both India and Pakistan, these have consistently shown that the raw learning achievement levels of children are substantially higher in private schools and that the learning gap between private and public schools has grown over time. Controlling for the generally superior home backgrounds of private school

Public–private partnerships in education 87 attendees (which could otherwise be biasing the results and generating the higher achievement levels) greatly reduces the raw achievement gap between private and public school students. Nevertheless, in several studies about one-third of the raw achievement gap remains even after statistically controlling for student background (e.g. Kingdon 1996 in India; Aslam 2009 in Pakistan). From an equity perspective this is an important consideration. For example, using ASER data from India and Pakistan, Alcott and Rose (2015) demonstrate that socio-economic status and gender are both important determinants of whether children are in school, the type of school they attend and whether or not they are learning. In particular, they note that whilst learning varies across the type of school a child attends, socio-economic disparities prevail with more disadvantaged children in private schools learning less than the more advantaged children in government schools. They also note that gender disparities in Pakistan are more pronounced amongst the more socio-economically disadvantaged. Within this burgeoning and thriving private sector, both countries have also seen the evolution of several innovative public–private partnership arrangements in the education sector. India has two dominant forms of public–private partnerships in education. The first of these has historically taken the form of grant-in-aid schools (generally referred to as aided schools) which were inherited from the British at Independence in 1947. These schools cater to a substantial portion of children in the country, especially at the middle and secondary levels of education. Second, India has recently converted every private unaided school into a PPP via its Right to Education Act (RTE) 2009, which necessitates by law that every private school give 25% of its seats to designated children from poor and disadvantaged backgrounds in return for reimbursement from the government. Aslam et al. (2017, p. 92) note that: Unlike the British system that went through multiple reforms, over time the aided schools of India became ossified in the same state as at the time of Independence; indeed, the environment for their running became more hostile in the early 1970s when the aided schools’ autonomy was seriously reduced through centralising legislation such as the Direct Payment Agreement of 1972 in Kerala and the Salary Disbursement Act of Uttar Pradesh in 1971, and in other states of India, which mandated that the salaries of aided school teachers would be paid directly into their bank accounts from the government treasury, rather than going as a grant to the private managers of their schools. This, the authors note, has proved to be ‘counterthetical to effectiveness and increased accountability’ (Aslam et al. 2017, p. 92) because it removed a key element from the private model – the need for teachers to be locally accountable to the private management of the school – and this, in turn, altered the incentives within the local educational structures. This PPP model functions increasingly like a government school – the aided school teachers are now recruited by the government’s Education Public Service Commission (exactly like their government school counterparts) and aided

88  Monazza Aslam and Geeta Gandhi Kingdon schools are mandated to charge no fees from their pupils (exactly like government schools). This loss of autonomy in important respects has ensured that the outcomes of aided school children are no better than those of children from government schools (Kingdon 1996). This model of aided schools in India has also evolved in an environment that has not enabled the PPP to thrive. Aslam et al. (2017) note that, by receiving a block grant (a flat amount of government subsidy which does not vary with changes in enrolled students), the incentives for efficiency are not built into the grant formula adopted by the government. There is anecdotal evidence that whilst student enrolment in these schools has declined, the number of teachers appointed in the schools has remained unchanged because teachers remain established in emptying schools as they are supported by powerful unions and teacher politicians (ibid., p. 92). Moreover, the 2009 RTE Act has created the largest PPP in the country by making it legally binding for all private schools to give at least 25% of their seats to children from disadvantaged backgrounds for which the government is required to reimburse them. However, not all agree that this form of PPP has created the relevant enabling environment for such a partnership to flourish. India also has a rich tradition of experimenting with other forms of PPP arrangements such as the provision of school vouchers to children. Like India, Pakistan also has a rich heritage of PPPs in education, inherited from the British in the form of grant-in-aid schools. Unlike India, these were temporarily stopped in 1972 due to the nationalisation of the entire country’s education system. It was resumed in the 1990s largely indirectly through the education foundations (such as the Punjab Education Foundation in Punjab) through the formation of semi-autonomous bodies and also directly through the department and ministries in order to achieve the universal education targets set by the international community. Since 2001–02, the country has embarked on a formal mission to embrace PPPs as a public policy strategy aiming to address resource and quality constraints identified by stakeholders within the country and partners in the international community and overcome through partnerships. Aslam et al. (2017, p. 84) note that: The year 2010 was a landmark year for education and PPPs in Pakistan, in which, on the one hand, education was elevated as a fundamental constitutional right under Article 25a for all children aged five to 16 years of age, and, on the other, the provinces of Punjab and Sindh passed their provincial PPP Acts, which were largely infrastructure-focused. Subsequently both provinces issued new acts and amendments called the Punjab PPP Partnership Act 2014 and the Sindh PPP (Amendment) Act 2015 to include services beyond infrastructure across all sectors and providing a cover for the public financing of services through transparently procured partnerships. Various PPP models have evolved in Pakistan over the years and have taken diverse forms with different modes of engagement and varying types of financing arrangements. Table 5.2 illustrates the variety of current PPP models in existence

Public–private partnerships in education 89 Table 5.2  Typologies of PPP arrangements in Pakistan Type I

Type III

On public sector government-owned sites and schools initiated by the education departments in provinces.

Schemes under semi-autonomous bodies, Education foundation programmes on private owners’ sites and schools and sometimes failed or underperforming public sector school sites managed by non-state partners.

Financing: mixed; some government or user charges and fees. Governing body examples are cadet and public schools.

Financing: vouchers, subsidies per child for targeted schools and agreed outcomes or key performance indicators.

Type II

Type IV

On public sector governmentowned sites initiated by private sector philanthropy – Corporate Social Responsibility-Civil Society Organisations through memorandums of understanding.

Procured through PPPs or advertised under PPP Acts 2010 (Punjab and Sindh).

Financing: public sector resources, supplemented by CSR, philanthropy and donors’ funds through CSOs (Type II may be switched to Types 1 and IV).

Financing: majority public sector finance that may be topped up by private sector resources.

Source: Aslam et al. (2017, p. 85)

in Pakistan and shows that they range from demand-side ones that offer vouchers to parents for attendance in private schools of choice to supply-side efforts that subsidise low-fee private schools and those that privately manage state schools (Partnership for management models). This section has highlighted the various forms and types of arrangements that exist across both India and Pakistan in which governments partner with the private/non-state sector to reach education goals. The following section discusses some of the key evidence across both countries to ask whether different types of partnership arrangements offer equitable solutions to the countries’ educational challenges.

PPP arrangements in India and Pakistan – do they offer an equitable solution to educational woes? There is emerging evidence across both India and Pakistan on various forms of PPP arrangements and the extent to which they provide educational opportunities to students from various backgrounds and improve learning outcomes. This section builds on evidence that has been presented by Aslam et al. (2017)

90  Monazza Aslam and Geeta Gandhi Kingdon on different types of PPP arrangements in India and Pakistan by introducing new evidence that has emerged in 2018 and 2019. As discussed earlier, the PPP arrangements existing in the two countries have taken various forms – ranging from contract schools to state subsidies to non-state providers to the provision of vouchers directly to students. Some high-quality evidence is beginning to emerge from both contexts that explore these different types of arrangements. Having said that, only one study looking at charter schools in Pakistan has evaluated the Partnership for Management arrangement in which the state contracts private providers to manage government schools, in that the private provider ‘adopts’ a government school. Malik et al. (2015) found that, whilst in Punjab adopted government schools are associated with better learning outcomes, the impact of such adoption on pupils’ learning outcomes in Sindh appears to be more ambiguous. The evidence on subsidy-type PPP arrangements in Pakistan includes two studies which evaluate the Foundation Assisted Schools programme initiated by the Punjab Education Foundation. This programme provided financial and technical support to low-cost private schools for each child enrolled in the programme (a per-student subsidy of Rs. 350, equivalent to USD 4, per month per enrolled child). Malik (2010) and Barrera-Osorio and Raju (2015) and another study of the Sindh Education Foundation’s Promoting Low-Cost Private Schooling in Rural Sindh programme (Barerra-Osorio et  al. 2017a, 2017b) also generally tended to find positive outcomes in terms of increased enrolments especially for the more disadvantaged, as well as improved learning outcomes. For example, Barerra-Osorio et al. (2017a, 2017b), specifically evaluated the delivery of education to underserved children (i.e. those residing in marginalised areas and particularly girls) using a PPP arrangement. For local private entrepreneurs who were qualified to participate, the programme offered a number of benefits (including a per-student subsidy, school leadership and teacher training and teaching and learning materials) to establish and run fee-free, coeducational primary schools in villages that were especially underserved in rural Sindh. Under this arrangement, the provincial government provided the School Education Foundation with full discretion over the regulation of programme schools, and programme-school operators enjoyed some flexibility to decide how to structure and run schools around the guidelines provided by SEF (ibid., p. 1). In their analysis of this programme, Barerra-Osorio et  al. (2017b) found that when programme schools (i.e. those which formed part of the PPP arrangement) were provided with a per-student subsidy to provide tuition-free primary education (with half of the treated villages receiving a higher subsidy for female students), enrolment increased by 30 percentage points and test scores improved by 0.63 standard deviations. There was also evidence of improved aspirations for marginalised boys and girls studying in programme schools. The evidence on vouchers (another form of PPP arrangement) in Pakistan is more limited. Malik’s (2010) mainly descriptive study explores the Education Voucher Scheme which aimed to promote choice, efficiency, equity and social cohesion in slums in Punjab. Through the Education Voucher Scheme, the Punjab Education Foundation aimed to deliver education vouchers (redeemable

Public–private partnerships in education 91 against tuition payments at participating private schools) to every household with children aged 5–13 years. Starting in 2006 as a pilot in Lahore, the Education Voucher Scheme has since expanded significantly across the province of Punjab (reaching almost 500,000 students aged 5–16  years across over 1,600 partner schools across the province).10 Malik’s (2010) early, mainly descriptive study, held positive views about the scheme from an equity perspective (noting that it allowed individuals from low-income backgrounds to send their children to school in the first instance and to choose the type of school in the second). Ansari’s (2012) study of the Education Voucher Scheme in Lahore also notes that it is associated with greater choice for families as they move from little or no schooling to having options in schooling choice. He also notes that this scheme has improved equity by providing lower-income families access to private schools. However, the author notes that student achievement data on children enrolled in this scheme is limited, but the sample on the whole is indicative of this voucher scheme’s students doing no worse than their non-voucher peers. As we have seen, India has converted every single private school into a PPP through the 2009 Right to Education Act which necessitates that every private school give 25% of its seats to designated poor and disadvantaged children in return for reimbursement from the government. If the Right to Education Act were to be implemented as planned, it would result in India having probably the largest number of children attending private schools through public funding as well as the largest attempts at school integration anywhere in the world (Aslam et al. 2017). However, there is some scepticism in the literature about the capability of the 25% reservation to target the most disadvantaged mainly because of financial barriers experienced by more disadvantaged families. There is also concern that several criteria have to be met for a poor child to be eligible for studying for free in a private school within this 25% provision under the RTE Act, and many applicants do not meet the eligibility criteria. In these cases, poor children may be excluded. Conversely, due to corruption in the system, well-off children who are ineligible may take up the places reserved for socio-economically disadvantaged children (Srivastava & Noronha 2016; Sucharita & Sujatha 2018; and the evidence summarised in Day Ashley et al. 2019).

The potential of voucher programmes Empirical studies of voucher programmes in India have noted the potential for such PPPs to improve student retention and provide access to more disadvantaged children (Crawfurd et al. 2019; Wolf et al. 2015; Muralidharan & Sundararaman 2015). Kumari (2016) studied a contract school arrangement  – the Rashtriya Adarsh Vidyalaya schools programme  – where private entities are responsible for providing infrastructure and managing the schools and the government supports 1,000 selected deprived students who will be charged nominal fees (and an additional 1,500 students may be enrolled and charged fees by the private entity). However, Kumari was critical of the high anticipated costs to the more disadvantaged students of this particular model (see Day Ashley et  al. 2019).

92  Monazza Aslam and Geeta Gandhi Kingdon The evidence on whether these PPP school models offer a higher quality schooling experience to children in India is mixed. Muralidharan and Sundararaman’s (2015) study of the voucher experiment in primary schools in Andhra Pradesh found that students who won the lottery/voucher to attend private schools had better outcomes in Hindi (only taught in private schools) and similar outcomes in other subjects, despite the fact that private schools are found to spend a third less funds per student than the government sector. This, according to the authors, was indicative of a more productive private sector that is able to achieve similar results with significantly less instructional time and significantly lower per student cost (mainly due to substantially lower teacher salaries). Wolf et al.’s (2015) study of another voucher programme, this time in a Delhi urban slum, also offers similar findings – children who won vouchers to attend private schools had better learning outcomes, particularly girls. Three years later however when Crawfurd et al. (2019) re-examined the same voucher scheme, they found no evidence that access to private schooling had generated improvements in learning outcomes. These contradictory findings of the same programme raise questions about the ability of the voucher programme PPP to improve learning of children from disadvantaged background (Day Ashley et al. 2019).

Discussion and conclusion Clearly, PPPs do not provide a total solution to all efficiency and equity problems in school education. It has been suggested by their advocates that PPPs can enhance both efficiency and equity. With regard to equity, PPPs which involve the public funding of private production of education do ease the financial constraint for poor families by permitting them to attend a private school without paying fees, via for example a government-given voucher, direct benefit transfer or scholarship. In these ways, PPPs in school education do improve equity in access to private schooling which, otherwise, would be accessible only to the rich. If the private schools are deemed to offer higher quality education than public schools in a country, then by enabling some poor children access to private schools, PPPs can promote equity in access to good quality schools. It is important to recognise that while some types of PPPs can result in some disadvantaged individuals accessing private schooling, they will not bring complete equality in access to good quality education. The reason is that the betteroff children can supplement the value of the government-given voucher or direct benefit transfer and thus can attend the higher fee private schools compared to the lower fee private schools that the voucher amount will permit the poorer children to attend. Whilst in theory it is possible to design vouchers to get perfect equality of access to education by making the voucher value inversely proportional to family income (Nechyba 2004), in practice, this requires accurate information on family incomes, which is not available. Thus, we can conclude that PPP arrangements can improve on the equity situation by easing the economic constraint for the poor and enabling them to access fee-charging private schools that are often perceived to be of higher quality.

Public–private partnerships in education 93

The contested role of private schools and of PPPs Whilst both India and Pakistan have adopted PPPs in education, there is a notable tension in the literature and rhetoric about such arrangements in both countries. On the one hand, their emergence is testament to their high public demand and the need for highly resource-constrained governments to provide schooling to all through partnerships, but, on the other hand, in public discourse the ‘privatisation of education’ tends to be lamented in both countries. While there has been public concern in recent years in both India and Pakistan about elite private schools’ fee levels and fee hikes (leading to new legislation to regulate fees in several Indian states and similar rhetoric in Pakistan), at the same time there has also been a recent upsurge in the government-mandated closure of low-fee private schools in several Indian states as a result of their not fulfilling the numerous conditions of government recognition. The evidence based on different types of PPP arrangements is sparse but emerging. However, the experience of India and Pakistan appears to suggest that PPP arrangements may be able to improve access and enrolment though the extent to which they can do so for the very poor and disadvantaged is sometimes questioned. In terms of their ability to improve learning, the evidence is indicative of students in such arrangements not doing worse than students in government schools and given the ability of these arrangements to achieve learning at a substantially lower per-pupil cost, this is often offered as a reason why these arrangements are more cost-effective (an important aspect of promoting education in large population countries experiencing considerable poverty). However, this in turn raises the question of whether marginally better learning outcomes make private schools and PPP arrangements more effective than government schools.

The factors behind the effectiveness of PPPs in education It has also been recognised in the literature that the capacity of partnerships to succeed has been held back by the state’s limited recognition of their potential (Malik et al. 2015). Most PPP arrangements in Pakistan have, for example, emerged despite minimal support and in some instances even having to overcome disincentives from the state to work with their private partners (ibid.). A similar point is noted by Aslam et al. (2017) in their case study in India, namely that generating an enabling environment is crucial for the effectiveness of PPPs and the very first step of that requires consent of the private partner in entering that relationship. The 2009 RTE Act in India offers a counterexample of this – where private partners were forced into this arrangement in India without prior consent, through legislative force. By virtue of this, the PPP relationship does not have the enabling environment that would be conducive to its success. A key aspect of an enabling environment also relates to transparency and clarity in the policies adopted towards the partner – the private sector actor – and the adoption of a facilitative approach towards them. Another critical element of successful partnerships requires that all parties (private and public) fulfil their end

94  Monazza Aslam and Geeta Gandhi Kingdon of the agreement. For example, under the Right to Education Act, government is required to reimburse private schools for educating poor children but in many states of India, the governments have delayed payments or substantially reduced reimbursements compared to the mandated levels. Treating the private sector as an equal partner, rather than as a volunteer, is also another crucial aspect of an enabling environment. And finally, adopting a clear policy towards PPPs rather than an ad hoc approach remains important. It is challenging to assess whether a PPP arrangement has succeeded or failed if the conditions in which it was established were not conducive for a transparent, effective and productive partnership.

Notes 1 Among other affiliations and memberships, Geeta Gandhi Kingdon is President of the board of a K-12 Registered Society non-profit private school in India and also a member of the Secondary Education Exam Board of the Government of Uttar Pradesh. 2 https://unstats.un.org/sdgs/report/2017/Goal-04/ 3 https://sustainabledevelopment.un.org/sdg4 4 ASER surveys have been conducted by Idara-e-Taleem-o-Aagahi (ITA), Pakistan, a non-governmental organisation working on education issues since the early 2000s. 5 The South Asia region also faces the challenge of increasing populations of schoolage children. India, for instance, has witnessed a substantial increase (of 4.3%) in the absolute primary school-age population of 6–10-year-olds between 2009 and 2014 (IMRB 2014). 6 This chapter draws from various literature reviews that both of the authors or one of the authors has been extensively involved in. These include the following: Day Ashley et al. (2014), Wales et al. (2015), Aslam et al. (2017) and Aslam and Rawal (2018). These reviews were mainly commissioned by the UK Department for International Development (DFID) (except Aslam et  al. (2018) which was commissioned by the Mastercard Foundation). The reviews by Day Ashley et al. (2014), Wales et al. (2015) and Aslam et al. (2017) mainly focused on primary education in developing countries globally whilst the Aslam and Rawal (2018) review focused on secondary education in sub-Saharan Africa. These were all ‘rigorous reviews’, that is, high-quality literature reviews that scoped extensive databases based on key search terms and identified medium-high quality studies that answered the research questions posed in each review. They closely followed the guidelines described by DFID on assessing the quality and strength of education in the education sector. 7 https://data.worldbank.org/indicator/se.prm.priv.zs 8 Data sources on private school enrolment, especially from government sources, are often heavily under-reported due to the transitory nature of unregistered schools and weak data systems. 9 This term has been used across various rigorous reviews and evidence syntheses commissioned by DFID in a bid to quantify the strength of existing evidence in education research. See Day Ashley et al. (2014) for further details. 10 https://schooledupunjab.wordpress.com/

References Alcott, B., & Rose, P. (2015) Schools and learning in rural India and Pakistan: Who goes where, and how much are they learning? Prospects, 45(3), pp. 345–363.

Public–private partnerships in education 95 Andrabi, T., Das, J., & Khwaja, A.I. (2008) A Dime a Day: The Possibilities and Limits of Private Schooling in Pakistan. [Online] World Bank Policy Research Working Paper No. 4066. Available from: http://citeseerx.ist.psu.edu/viewdoc/download ?doi=10.1.1.207.833&rep=rep1&type=pdf [Accessed 8 June 2020]. Annual Status of Education Report (ASER) (2010–2018) Annual Status of Education Report ASER Pakistan 2018: National. [Online] Available from: \http://aserpakistan.org/document/aser/2018/reports/national/ASER_National_2018.pdf [Accessed 6 June 2020]. Ansari, A. (2012) Educational Voucher Scheme in Lahore: Serving the Underserved. [Online] Teachers College Columbia, Working Paper. Available from: https:// ncspe.tc.columbia.edu/working-papers/OP203.pdf [Accessed 30 July 2020]. Aslam, M. (2009) The relative effectiveness of government and private schools in Pakistan: Are girls worse off? Education Economics, 17(3), pp. 329–354. Aslam, M. (2017) The Effectiveness of Non-state Provision to Raise Educational Access and Quality for the Marginalised in Developing Contexts: A Review of the Evidence. Knowledge, Evidence and Learning for Development, HEART. (Unpublished paper). Aslam, M., & Rawal, S. (2018) Public-private Partnerships and Private Actors in Secondary Education in Sub-Saharan Africa. [Online] Background report prepared for the Mastercard Foundation. Available from: https://mastercardfdn.org/wpcontent/uploads/2019/03/Public-Private-Partnerships-FINAL.pdf [Accessed 8 June 2020]. Aslam, M., Rawal, S., & Saeed, S. (2017) Public-private Partnerships in Education: A  Rigorous Review of the Evidence. [Online] Report commissioned by Ark Education Partnerships Group. Available from: https://arkonline.org/sites/default/ files/ArkEPG_PPP_report.pdf [Accessed 8 June 2020]. Barrera-Osorio, F., Blakeslee, D.S., Hoover, M., Linden, L.L., Raju, D.,  & Ryan, S.P. (2017a) Delivering Education to the Underserved through a Public-private Partnership Program in Pakistan. [Online] World Bank Policy Research Working Paper, WPS-8177. Available from: http://documents.worldbank.org/curated/ en/868011504015520701/pdf/WPS8177.pdf [Accessed 8 June 20220]. Barrera-Osorio, F., Blakeslee, D.S., Hoover, M., Linden, L.L., Raju, D., & Ryan, S. (2017b) Leveraging the Private Sector to Improve Primary School Enrolment: Evidence from a Randomised Controlled Trial in Pakistan. [Online] World Bank. Available from: https://pdfs.semanticscholar.org/a424/70508cbb28f664cd0ff47b5dd f18126b6512.pdf [Accessed 8 June 2020]. Barrera-Osorio, F., & Raju, D. (2015) Evaluating the impact of public student subsidies on low-cost private schools in Pakistan. Journal of Development Studies, 51(7), pp. 808–825. CAERUS Capital Report (2016) The Business of Education in Africa. [Online] Available from: https://edafricareport.caeruscapital.co/thebusinessofeducationinafricasummary.pdf [Accessed 8 June 2020]. Colclough, C. (1996) Education and the market: Which parts of the neo-liberal solution are correct? World Development, 24(4), pp. 589–610. Crawfurd, L, Patel, D., & Sandefur, J. (2019) Low Returns to Low-cost Private Schools: Experimental Evidence from Delhi. Working Paper, Centre for Global Development. Datta, S., & Kingdon, G. (2019) Gender Bias in Intra-household Allocation of Education in India: Has It Fallen over Time? [Online] IZA Discussion Paper No. 12671, IZA Institute of Labor Economics, Bonn. Available from: http://ftp.iza.org/dp 12671.pdf [Accessed October 2020].

96  Monazza Aslam and Geeta Gandhi Kingdon Day Ashley, L., Aslam, M., Meyer, A., & Skinner, R. (2019) Private Schools and Publicprivate Partnerships in India: Can They Provide Inclusive and Equitable Quality Education to the Most Deprived? Paper prepared for the 2019 UK Forum for International Education and Training (UKFIET) Conference, Oxford. Day Ashley, L., McLoughlin, C., Aslam, M., Engel, J., Wales, J., Rawal, S., Batley, R., Kingdon, G., Nicolai, S., & Rose, P. (2014) The Role and Impact of Private Schools in Developing Countries. [Online] London: Department for International Development (DFID). Available from: www.odi.org/sites/odi.org.uk/files/odi-assets/ publications-opinion-files/10145.pdf [Accessed 8 June 2020]. Day Ashley, L.,  & Wales, J. (2015) The Impact of non-State Schools in Developing Countries: A Synthesis of the Evidence from Two Rigorous Reviews. [Online] DFID Education Rigorous Literature Review. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/486417/imact-non-state-schools-dev-countries.pdf [Accessed 30 July 2020]. Elacqua, G., Iribarren, M.L., & Santos, H. (2018) Private Schooling in Latin America: Trends and Public Policies. [Online] Technical Note, Inter-American Development Bank. Available from: https://publications.iadb.org/publications/english/ document/Private-Schooling-in-Latin-America-Trends-and-Public-Policies.pdf [Accessed 8 June 2020]. IMRB (2014) National Sample Survey on Estimation of Out-of-School Children in the Age 6–13 in India. New Delhi: Social Research Institute, IMRB. Kingdon, G. (1996) The quality and efficiency of private and public education: A case study of urban India. Oxford Bulletin of Economics and Statistics, (58), pp. 57–82. Kingdon, G. (2017) The Private Schooling Phenomenon in India: A Review. [Online] Working Paper No. 1706, Department of Quantitative Social Science, Institute of Education, University College London. Available from: https://ideas.repec. org/p/qss/dqsswp/1706.html [Accessed: 9 June 2020]. Kumari, J. (2016) Public – private partnerships in education: An analysis with special reference to Indian school education system. International Journal of Educational Development, 47, pp. 47–53. Malik, A. (2010) Public-Private Partnerships in Education: Lessons Learned from Punjab Education Foundation. Philippines: Asian Development Bank. Malik, R., Bari, F., Muzaffar, I., Mashhood, T., Mansoor, M., & Ali, A. (2015) Partnerships for Management in Education: Evidence from Punjab and Sindh. [Online] Institute of Development and Economic Alternatives, Lahore, Pakistan. Available from: www.academia.edu/15328083/Partnerships_for_Management_Evidence_ from_Punjab_and_Sindh [Accessed 9 June 2020]. Muralidharan, K., & Sundararaman, V. (2015) The aggregate effect of school choice: Evidence from a two-stage experiment in India. Quarterly Journal of Economics, 130(3), pp. 1011–1066. Nechyba, T. (2004) Income and peer quality sorting in public and private schools. In Hanushek, E., & Welch, F. (Eds.) Handbook of Economics of Education. North Holland: Elsevier. Patrinos, H.A., Barrera-Osorio, F., & Guáqueta, J. (2009). The Role and Impact of Public-Private Partnerships in Education. [Online] World Bank. Available from: www.ungei.org/Role_Impact_PPP_Education.pdf [Accessed 9 June 2020]. Results for Development (R4D) (2017) Affordable non-State Schools in Contexts of Conflict and Crisis: A Literature Review. [Online] Prepared by Results for Development (R4D) for Education Development Center to submit to the United States

Public–private partnerships in education 97 Agency for International Development. Available from: www.r4d.org/wp-content/uploads/ANSS-Final-Version3_May24_2018.pdf [Accessed 9 June 2020]. Srivastava, P., & Noronha, C. (2016) The myth of free and barrier-free access: India’s Right to Education Act-private schooling costs and household experiences. Oxford Review of Education, 42(5), pp. 561–578. Sucharita, V.,  & Sujatha, K. (2018) Engaging with social inclusion through RTE: A case study of two private schools in Delhi, India. International Journal of Inclusive Education, pp. 1–15. Wales, J., Aslam, M., Hine, S., Rawal, S., & Wild, L. (2015) The Role and Impact of Philanthropic and Religious Schools in Developing Countries: A Rigorous Review of the Evidence. [Online] London: Department for International Development. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/482337/role-impact-philanthropic-schools. pdf [Accessed 9 June 2020]. Wolf, P.J., Egalite, A.J.,  & Dixon, P. (2015) Private school choice in developing countries: Experimental results from Delhi, India. In Dixon, P., Humble, S.,  & Counihan, C. (Eds.) Handbook of International Development and Education. Cheltenham: Edward Elgar Publishing.

6 The influence of politics on girls’ education in Ethiopia Louise Yorke, Pauline Rose and Alula Pankhurst

In Ethiopia, as in many Southern country contexts, girls have benefitted least from the rapid expansion of education access and are more likely to be affected by the ‘learning crisis’ (UNESCO 2014; World Bank 2018). While more girls are in school than ever before, many still do not have the opportunity to access school. Girls who are in school often learn little while they are there and dropout of school early without acquiring even basic skills in maths and literacy. The poor quality of education is a significant factor contributing to students’ low learning levels. However, gender norms and practices, operating at multiple levels of society, produce additional barriers for girls and underpin the inequalities that they face (Colclough et al. 2000). While considerable commitment has been directed towards improving girls’ education in recent decades, more effort is needed to ensure that education systems are equitable for girls. A key question that we pose in this chapter is why the Ethiopian education system has not yet delivered equitable access and learning for all girls, despite highlevel government commitment. We argue that this question requires a deeper look at the politics of education systems (Hickey & Hossain 2019; Levy et al. 2019). Research in other related contexts has highlighted the value of understanding the gap between ambitious policy for girls’ education and disappointing progress in this area. Notably, in Kenya and South Africa during the period 2007–11, Unterhalter and North (2019) focused on policy actors and relationships, in what they term the middle space. They identify this space as an important site of negotiation for determining whether such policies are translated into practice. Building upon Unterhalter and North’s focus on the politics of implementation in education systems, we consider not only what is happening within the middle space but also how it is influenced by, and interacts with, the wider political context over time. In exploring the politics of girls’ education, we draw on Hickey and Hossain’s (2019) domains of power approach, which provides a framework for understanding the interaction between the political settlement and the education policy domain. The political settlement refers to ‘the balance or redistribution of power between contending social groups and social cases, on which any state is based’ (Di John  & Putzel 2009, p.  4). In their comparative analysis of the political economy of education quality reforms in six low- and middle-income countries, Hickey and Hossain (2019) demonstrate how different types of political contexts

Influence of politics on education in Ethiopia 99 interact with the education policy domain in ways that shape the design, uptake and implementation of reforms design to improve learning outcomes. Similarly, Levy et al. (2019) demonstrate how, in South Africa, policy and implementation are shaped by the political and institutional context. Yet the role of politics in understanding patterns of progress in girls’ education specifically has received very limited attention. In this chapter, we address this gap by exploring the role of politics in relation to girls’ primary education in Ethiopia, drawing on findings from our research on a large-scale government quality education reform programme. Drawing on Hickey and Hossain’s (2019) domains of power framework, we consider progress in girls’ education and how this may be influenced by the wider political settlement. We explore how politics shapes the commitment and capacity of the government to promote reforms aimed at advancing gender equality in education. Given that the negotiation of gender equality is a profoundly political process (Goetz 2019), it is particularly pertinent to understand more about how politics and power are relevant for girls’ education. We focus in particular on dynamics within the education policy domain, which refers to the meso-level field of social relations within which actors promote competing agendas (Hickey  & Hossain 2019), which is similar to Unterhalter and North’s (2019) concept of the middle space. We consider how policies are formulated and implemented, including how ideas in the wider political settlement shape what is possible within the education system in terms of strategies to promote girls’ education (Hickey & Hossain 2019). We concentrate on the role and agency of policy actors within the education system – noting how they may be influenced by wider political and social dynamics – and what this means for girls’ education. Incorporating a gender lens, we highlight the importance of considering the role of gender norms within an education system which can mistakenly be seen as a gender-neutral site. We conceptualise gender norms as informal institutions which, in contrast to formal institutions, are more likely to be unwritten and imparted outside of officially sanctioned channels. We explore how informal institutions permeate formal structures of the education system, through their inherent exclusion of women, before moving on to explore the interaction between formal and informal institutions and their effects. Others have shown the importance of informal institutions in understanding how behaviour and outcomes are shaped (e.g. Chappell  & Waylen 2013; Sen 2007). Broadly speaking, these authors have shown how the interaction between formal and informal institutions may work to complement, compromise or curtail progress in the area of girls’ education as we will further explore. Given the influence of gender norms on girls’ education in Ethiopia, exploring if and how gender norms operate within the education policy domain seems particularly salient. To set the context for this chapter, in the next section we provide an overview of the progress and challenges in girls’ education, before describing our research and providing a brief political background in the Ethiopian context.

100  Louise Yorke, Pauline Rose and Alula Pankhurst

Progress in female education and gender equality in Ethiopia Twenty years ago, Colclough et  al. (2000) demonstrated that while poverty limits the educational opportunities of both girls and boys in Ethiopia, gender norms and practices, working within the household, the school, labour market and society created and maintained the significant and additional inequalities that girls faced at that time. Since 2000, considerable progress has been made in getting more girls into primary school in Ethiopia.1 Government statistics show that girls’ primary school net enrolment has more than doubled from a mere 37% of the official school-age population in 1999–2000 to 96% in 2016/17 (Ministry of Education, 2017), with average differences between boys and girls also narrowing over this time (Figure 6.1). In addition, more girls now are staying in school, with 52% of girls completing the full cycle of primary school in 2016/17 (Ministry of Education 2017), compared with only 12% of girls in 2000 (Ministry of Education 2005). Profound differences in enrolment for boys and girls are found across regions (Figure 6.2). Within regions, rural students are even more disadvantaged, with poverty further compounding the disadvantages faced. As seen in Figure 6.1, gender gaps are starting to open up once again, while girls are also more likely to be affected by the learning crisis. According to Ethiopia’s National Learning Assessment from 2007 to 2010, while overall learning

140

Primary gross enrolment ratio

120 100 80 60 40 20 0

Male

Female

Figure 6.1 Gross enrolment rate at primary level (grades 1–8) by gender, 1995/ 96–2016/17 Source: Ministry of Education (2017)

G5-8 Male

77 71

77 68

64

A

L

A

O TI A N

ED

A

N

W

BA

IS D

IR

BA A

A H A

D

D

RI RA

A

A G

G . BE

N

M

U

BE

M

LL

U

Z

P N

LI A

SN

M

M SO

IY

A

RA

RO O

M

H

A

FA A

A

TI

G

RA

Y

R

30 28

37

45

59

69

70

70

78

81

91

89 92

101 99

111

120

133

156

Influence of politics on education in Ethiopia 101

G5-8 Female

Figure 6.2 Gross enrolment rate at upper primary level (grades 5–8) by gender and region, 2016/17 Source: Ministry of Education (2017)

levels are low, girls are even less likely to be learning than boys in all subjects, in all primary grades and across time (Woldetsadik 2013). Outside school, girls and women continue to face discrimination, harassment and significant domestic work burdens, and many of these challenges follow girls into the classroom, reminding us that girls’ education and gender equality are interlocked and progress in one domain requires progress in the other. In explaining these enduring challenges, gender norms and practices highlighted by Colclough et al. (2000) two decades ago are still a powerful influence and remain as some of the greatest barriers to girls’ education enrolment, progression and achievement. Gender norms refer to the socially constructed forms of masculinity and femininity that organise daily life and result in the dominant status of men and the lower status of women. Ethiopia continues to be a deeply patriarchal society, and traditional and restrictive norms and practices constrain the lives and experiences of many girls and women both directly and indirectly (Colclough et  al. 2000). A  strong sense of surveillance/monitoring by others helps to reinforce these ubiquitous, yet unspoken rules and norms (Chuta  & Crivello 2013). While not all women experience disadvantages a significant majority are not in the position to enforce their rights especially those living in rural areas and those living in poverty (Burgess 2012). As a result, practices such as early marriage and heavy domestic work burdens and limited formal employment

102  Louise Yorke, Pauline Rose and Alula Pankhurst opportunities continue to impact the education and opportunities of girls and women in Ethiopia. An illustrative example of the persistence of gender norms and resulting practices is the continued practice of early marriage. Ethiopia has one of the highest rates of early marriage in sub-Saharan Africa, despite the government introducing a law in 2000 setting the age of marriage at 18 (Stavropoulou et al. 2017). There is evidence that the age of marriage is increasing overall. However, progress is uneven and especially slow in rural areas (Central Statistical Authority 2016), where gender norms are often more entrenched and difficult to address. Strategies to address gender norms have mostly focused on the community level. However, while gender norms may be produced and reproduced within communities, parents and communities are often doing what they perceive to be in the best interests of girls and seek to protect rather than harm them (Boyden et al. 2012; Chuta & Morrow 2015; Pankhurst 2014). While within communities more sophisticated and nuanced approaches are likely to be needed, in this chapter we seek to highlight how it is important also to consider how gender norms are operating in the wider society and how these affect education systems. In the next section, we describe the Research on Improving Systems of Education (RISE) Ethiopia programme on which this chapter is based, for which we have gathered information on Ethiopian policy-makers’ views on girls’ schooling. Before reporting our findings, we give an overview of the political context in Ethiopia, with respect to its policies and progress in education and promoting gender equality.

Researching the politics of educational reform in Ethiopia We draw on work undertaken as part of from the Research for Improving Systems of Education (RISE) Ethiopia study.2 This study has been investigating the design, adoption, implementation and impact of the government’s largest education quality reform programme, which has an explicit focus on equity for marginalised groups, particularly girls. The data that we explore in this chapter come from our system diagnostic of the Ethiopian education system carried out in 2018, which included analysis of government documents (policy, plans and reports) and actor mapping. In addition, we undertook key informant interviews to gather the perspectives of 150 key government stakeholders, who had experience with the design and/or implementation of education reforms. In these interviews, we explored these processes aimed at improving education quality and equity. Our interviews took place between February and December 2018, at the federal, regional and woreda (district) level, across the seven of Ethiopia’s 11 regions and city administrations that are included in the RISE research (Table  6.1).3 The small number of women who we were able to interview is reflective of the gender distribution of staff in key government positions in the education system in Ethiopia.

Influence of politics on education in Ethiopia 103 Table 6.1  Government stakeholders included in the key informant interviews Level Stage 1, February 2018 Stage 2, May 2018 Stage 3, December 2018

Federal level government officials Regional and zonal level government officials Woreda (district) level government officials Total

Male

Female

26

3

62

4

50

5

138

12

The RISE Ethiopia research team members carried out the interviews.4 Informed consent was obtained for all interviews. These were undertaken in the preferred language of the participants where possible. We recorded, translated and transcribed all interviews. We encouraged participants to share their personal views in relation to their role and we assured them that their responses would remain anonymous. While the sensitivity of exploring the politics of government reforms was a challenge in our study, a major advantage was the experience of the RISE Ethiopia research team of the political and social context. We used thematic analysis to code the data, facilitated by NVivo software. Two stakeholder dialogue meetings provided an opportunity to validate key issues emerging from the analysis. Members of these workshops included a number of donors and government officials who had participated in the interviews. Drawing on the data from these interviews, we explore what was taking place both in the wider political context and within the education policy domain, with respect to approaches and strategies related to girls’ education. It is to these elements that we now turn.

Dominance of girls’ education in Ethiopian political priorities In Ethiopia, the political settlement has been described as dominant developmental, which refers to relatively cohesive intra-elite relations within the ruling coalition and a focus on the idea of a ‘developmental state’ (Clapham 2018; Zenawi 2012). After defeating the socialist ‘Derg’ regime (1974–91), the current government came to power in 1991 under the leadership of Meles Zenawi, who ruled until his death in 2012. Key objectives of the new ruling coalition, the Ethiopian People’s Revolutionary Democratic Front (EPRDF),5 were to establish peace and stability, to reduce poverty and to expand social and physical infrastructure. Economic growth and social inclusion became political priorities for achieving these overarching objectives, and education came to be seen as having a key role in achieving these objectives (Clapham 2019; Zewide 2018). The government embarked upon an ambitious education strategy and the importance

104  Louise Yorke, Pauline Rose and Alula Pankhurst of equity, including gender equity, in and through education was firmly set out. Published in 1994, the Education and Training Policy (ETP), which so far has remained the main education policy of the government, sets out a transformative agenda for the education sector. For girls, education has a key role in ‘reorienting society’s attitude and value pertaining to the role and contribution of women in development’ (Ministry of Education 1994, p. 11). In addition to education, gender equality and women’s empowerment have also been key priorities of the Ethiopian government since coming to power in 1991. It is well documented how women played an important role in the struggle against the Derg, and it has been suggested that this paved the way for a feminist space to emerge in the initial years of the post-1991 government (Blystad et  al. 2014; Semela et  al. 2019). Numerous policies intended to improve the lives of girls and women were developed, paying attention to international standards (e.g. the Beijing Platform for Action). Gender mainstreaming was taken up as a core approach to addressing gender equality across sectors in addition to strengthening the participation of women in the country’s economic development and political affairs (National Planning Commission 2016). Although committed to gender equality, ensuring that it has remained prominent on the government’s agenda was less straightforward for gender equality than for education. In the early years of the new EPRDF government, a number of influential female figures played pivotal roles in ensuring that the 1995 Constitution was gender-sensitive (Article 35) (Government of Ethiopia 1995).6 The Women’s Affairs Office was established in 1991 which later became a Ministry in its own right, and this Office ensured that women’s departments were founded in all government ministries, including the Ministry of Education. Additionally, a National Policy on Women was developed in 1993, and the implementation of this policy was supported by the introduction of the first National Action Plan on Gender Equality in 2000 and an updated version in 2006 (MoWA 2006). However, while the government established an enabling space for advancing social objectives including gender equality aided by the efforts of influential female figures, this space was created outside the central government structures. In coming to power, the government established a largely hierarchical and topdown style of governance giving the government considerable reach down to the local level, while also ensuring that they had tight control of the political space and authority over what was implemented (Hagmann & Abbink 2012; Vaughan 2011). Like other dominant developmental states (e.g. Rwanda), Ethiopia made rapid economic advancement in a relatively short time period. Yet the ‘downside of the dominance’ of the EPRDF was perhaps most clearly illustrated during the 2005 elections (Williams 2019). In response to opposition gaining greater ground than expected, the ruling coalition attempted to re-establish authority and secure power by reasserting their control in both civil and political space (Clapham 2019). This led to a stifling of civil society organisations including those working to advance gender equality, which had drastic consequences for the advancement of the gender equality agenda (Blystad et  al. 2014; Burgess

Influence of politics on education in Ethiopia 105 2012). The fact that these organisations were outside the core government structure may have enabled the swift manner in which they were dismantled. Justification for this increasing centralisation of power was made through a reenergised and accelerated focus on economic development (Zenawi 2012). The goal of reaching lower middle-income status by 2025 became the driving force of the government and ‘double-digit’ growth became a new source of legitimacy claims as set out in the government’s Growth and Transformation Plan (GTP) (Clapham 2018; Hagmann & Abbink 2012). While poverty reduction and social inclusion remained on the government’s agenda, the focus on economic growth and infrastructure expansion gained greater ascendency. Implicit was the assumption that economic growth would lead to greater social equality, without the need to address underlying structural disadvantages (Fiseha  & Gebresilassie 2019; Hagmann & Abbink 2012). The increasing focus on economic growth, in turn, had implications for the strategies in the domains of education and gender equality (Woldehanna & Araya 2019). In education, while the core policy remained, focus intensified on the rapid expansion of education to create the skilled labour force needed to move towards the goal of lower middle-income status. This had implications for how gender equality was envisioned, and women came to be seen as having a key role in advancing the government’s development agenda. At the same time, it was suggested that women’s education and economic empowerment would be a central mechanism for increasing gender equality and addressing the discrimination of women in society by ‘unleashing women’s potential through education, skill development and employment’ (National Planning Commission 2016, p.  92). Thus, the emphasis was on an instrumental approach to women’s participation in education and formal employment to achieve the goal of rapid economic growth and promote other positive externalities without addressing the root causes of women’s inequality (Goetz 1997). It seems that policies implemented by the government in education and gender equality were more concerned with improving the educational access, formal employment and participation, while doing little to redistribute power in the social relationships affecting the lives of women. This perhaps explains the growing gender gaps amid increasing enrolment rates described in the previous section. In the next section, we consider in more depth the interaction of the specific political context of Ethiopia with the education policy domain and, in particular, with girls’ education.

Policy legacies and strategies for girls’ education We find that strategies that were being pursued in relation to girls’ education have been very much shaped by the wider political context. In particular, ideas in the wider political settlement concerning how development happens in the context of the hierarchical mode of governance have influenced the perceived challenges and solutions for girls’ education. These strategies have mainly served to improve enrolment of some but not all girls, while doing much less to improve their learning outcomes.

106  Louise Yorke, Pauline Rose and Alula Pankhurst Since 1994, the ETP has been implemented through a series of Education Sector Development Plans (ESDPs) that outline goals and strategies for the education sector, with the government currently implementing its fifth plan (Ministry of Education 2015). The 1994 ETP identified a number reforms to improve girls’ education, including ensuring the curriculum is sensitive to gender issues; increasing the number of female teachers; providing educational inputs for girls; encouraging the participation of women in education organisation and management and providing financial support to raise the participation of women in education. While these are important strategies for improving girls’ education, it is evident that these strategies largely relate to issues of representation and increased inputs, reflecting more widely held ideas in the political domain concerning how progress happens and may partly explain why progress has been mostly limited to access rather than quality of education. Across the five ESDPs to date we can see that strategies to improve girls’ education have not deviated from the ETP (Ministry of Education 1994), focusing on increased inputs, increased participation of girls, increasing the number of female teachers and addressing gender dimensions in the learning environment (see Table 6.2). Characteristic of the hierarchical mode of governance, the link Table 6.2 Strategies for girls’ education outlined in the Education Sector Development Plans (I–V) ESDP II

ESDP III

ESDP IV

ESDP V

Access and retention

ESDP I

Strategies to improve girls’ education

Local and mass media campaign encouraging girls’ schooling

-





-

-

Increase provision of water supply



-







Girl-friendly facilities, including separate latrines





-













-





















Gender Remove gender bias in textbooks dimensions in and curricula the learning Gender awareness training (parents, environment teachers, managers and students) Increase number of female teachers

Improve the system

Guidance and counselling for girls





-

-

-

Tutorial support and assertiveness training for girls





-



-

Gender sensitive budgeting

-

-

-

Capacity of structure to promote girls’ education

-

-

-





Increase female administrators, supervisors and directors

-

-

-



-

Source: Ministry of Education (2015)



Influence of politics on education in Ethiopia 107 with the ETP has ensured tight control of the federal government over what is planned with limited scope for innovation with regard to the strategies pursued within regions of Ethiopia. Strategies for addressing cultural norms are also evident in some of the plans, including local mass media campaigns and skills training for girls. While the continued focus on girls’ education through the plans should be commended, it appears that the approach adopted has not been sufficiently transformative and has restricted the ability to make meaningful and sustainable progress in relation to girls’ education. Wider political dynamics may also influence stakeholder’s understandings on what they perceive as necessary to ensure progress in girls’ education. Ongoing difficulties with making progress are reported throughout the ESDPs, including with respect to engaging stakeholders in efforts to advance gender equality. The government has tried to address this by alternating between a mainstreaming approach in the earlier plans (ESDP I–III) and a targeted approach in the fourth plan (ESDP IV), returning once again to a mainstreaming approach (ESDP V) to ensure that cross-cutting issues become the responsibility of all implementing bodies. However, this has not sufficiently brought about the needed change in girls’ education. In the context of a top-down governance structure within the education system, the lack of motivation of stakeholders to implement these policies perhaps relates to their beliefs about what is needed to bring about change and also to the process for monitoring progress in girls’ education. Government targets related to girls’ education have consistently focused on enrolment and gender parity, with learning targets only introduced in the most recent plan (ESDP V), providing few incentives for stakeholders to engage with more hidden forms of gender inequality. Data on girls’ education is not disaggregated across rural–urban location or socio-economic status which erases important aspects of difference among girls. In summary, we find that within the education policy domain, strategies are often not sufficiently transformative and constrain responses to girls’ education. As we go on to discuss, some stakeholders may not be sufficiently motivated to implement them, which we argue is at least partly affected by the wider political context.

Informal institutions and the education policy domain In adopting a gender lens to our analysis, we take account of the influence and interaction of informal institutions within the education domain which may shape behaviours and outcomes. As such, in this section we explore how gender inequalities emerge in the education system in both nominal and substantive ways, including the power and representation of female actors in the education system, along with other more hidden gender biases.

Women’s power and representation in the education policy domain Key influential female figures not only played pivotal roles in advancing the gender equality agenda in the political realm in the early years of the government,

108  Louise Yorke, Pauline Rose and Alula Pankhurst but their influence was also an important factor within the education policy domain. Dr Gennet Zewide was one such figure who served as the Minister for Education from 1991 to 2006 and, until very recently was the only woman to have held this position.7 Through her leadership, Dr Zewide raised the visibility of women in this domain and played a central role in incorporating gender concerns into education policy and ensuring that strategies for greater gender equality were pursued, evident from the inclusion of issues on girls’ education from the first Education Plan (ESDP I) (Rose 2003). These achievements, however, were hard-won, and as a female Minister operating in a male-dominated space, Dr Zewide8 describes the continuing opposition she faced despite her high-level position: For a long time, they didn’t take me seriously. . . . I was the only Minister who had bad things written about them in the media. . . . I was the only one that, you know, who some parliamentarians disobeyed. Against such opposition, Dr  Zewide describes how she had to ‘struggle’ to encourage others working in the education system to recognise the importance of gender equality. She describes the opposition of male stakeholders within the education system: One of the directors, an elderly man who has been working in the Ministry [of Education], for a long time, he came up with this idea that, you know, ‘we don’t need a women’s department in the Ministry [of Education] as the government does not discriminate against women. This encounter, she explains, was emblematic of the attitudes of others in the education system at this time. Had it not been for the presence, commitment and persistence of Dr  Zewide, the issue of girls’ education would most likely been sidelined. Even to this day, the representation of women within the education system remains limited. Our mapping of key education stakeholders found that women are grossly under-represented in the education system, especially in decision-making positions, even though ensuring women’s participation has been a policy priority throughout the education plans. Only 12 of 150 of our interviewees were women, despite our efforts to include women whenever they were identified through our actor mapping. We found that women tended to occupy positions considered as ‘lower status’ by other stakeholders, often positions associated with advancing equity issues. The absence of women from decision-making positions is a glaring injustice in itself, but it also has consequences for policies that are pursued, particularly those related to girls’ education. Given that the design of reforms occurs in a top-down manner, with only those in high-level positions included in decisionmaking in any meaningful sense (Asgedom et  al. 2019), women are unlikely to have been involved in the design of these plans and programmes.

Influence of politics on education in Ethiopia 109 One male federal-level stakeholder described the difficulties in increasing the number of women in the education system. For example, he described how the introduction of gender quotas was met with resistance from some stakeholders who claimed that these positions were ‘already occupied’. A number of male interviewees also suggested that women were reluctant to enter into leadership positions, even though there was training available. No explanation for women’s hesitancy was provided, but we can speculate that this may be due to the fact that insufficient efforts were made to attract women, an issue also highlighted by Dr Zewide, or that women were not willing to enter into a male-dominated structure that did not accommodate their needs. Similar to Dr  Zewide, both male and female stakeholders included in our interviews working to advance gender equality in education described experiencing these same struggles. At the federal level, stakeholders tasked with the main responsibility for designing and implementing strategies for improving girls’ education reported having very little power within the structures of the Ministry of Education and described difficulties in mobilising support and coordinating action to improve girls’ education. At the regional level, stakeholders further described how the de-prioritisation of girls’ education resulted in inadequate funds being allocated. Particularly striking in these accounts are the similarities between the present-day reality captured during our interviews and that described by Dr Zewide more than 20 years ago. Women continue to be under-represented in decision-making powers in the education system and gender biases are still prevalent in the attitudes and behaviour of stakeholders, despite over 20 years of policy commitment. This signals the persistence of more hidden and intangible factors, to which we will now turn.

Stakeholders’ attitudes towards girls’ education While all stakeholders included in our analysis were committed to improving educational quality, a different perspective emerged in relation to stakeholders’ commitment to gender equality in education. Discrepancies were found amongst stakeholders concerning the challenges that stakeholders believed girls faced and what they considered as necessary to ensure progress in relation to girls’ education. In explaining these patterns of difference, we grouped the attitudes of stakeholders who we interviewed into three broad categories – progressive (gender equality requires attention to gender norms), narrow (gender equality is synonymous with gender parity and thus has been achieved) and mixed (gender equality is linked to gender parity and more inputs are needed) (Table 6.3). We found stakeholders’ views to be influenced by ideas related to how development happens, patterns of governance and more implicit gender biases, which interacted in different ways. Furthermore, we identified differences across regional locations in terms of which views stakeholders were more likely to hold, but not across gender with both male and females found in each category.

110  Louise Yorke, Pauline Rose and Alula Pankhurst Table 6.3  Categorisation of stakeholders’ attitudes towards girls’ education Attitudes towards girls’ education

Progressive

Narrow

Mixed

Progress

Gender parity has improved as a result of an increase in girls’ enrolment, but this is not enough to secure gender equality.

Progress in gender equality equals gender parity. As such gender equality is achieved.

Progress in gender equality is mostly linked to gender parity.

Challenges

There are multiple and intersecting challenges that different groups of girls face. Gender norms and practices have a significant impact. Lack of resources and attention are directed to girls’ education.

Girls do not face challenges and are not disadvantaged. Girls are seen as a homogenous group.

Girls face many challenges, but differences of experience are not emphasised. Gender norms and practices have a significant impact. Lack of resources and attention are directed to girls’ education.

Strategies

Increased and coordinated attention and support needed. Should be tailored to needs. Planning should be gender sensitive.

Strategies are not needed/were not suggested.

More inputs are needed. Girls should be supported to withstand challenges.

Who held these views?

Held by many stakeholders, at federal, regional and woreda levels, especially in Amhara, Oromia and Southern Nations, Nationalities and People’s Region.

Held by some stakeholders at regional and woreda levels, mainly in Addis Ababa and the emerging regions.

Held by most stakeholders at federal, regional and woreda levels, particularly in Oromia and Southern Nations, Nationalities and People’s Region.

Progressive views Almost one-third of the stakeholders we interviewed expressed progressive views on gender equality. These views were found particularly at the federal and woreda level in the northern regions of Amhara and Tigray. Individuals in this category

Influence of politics on education in Ethiopia 111 acknowledged that great progress had been made in getting more girls into school, but that this was not enough: In terms of gender, there is some improvement in the participation of females, but still several females need special support in order to stay in schools. (Male stakeholder, woreda level, Amhara) In this group, stakeholders reflected on the damaging effects of gender norms and practices and emphasised the intersecting challenges that meant girls living in rural areas were even more disadvantaged. They also emphasised girls’ agency, noting that when girls are supported educationally, they are capable of achieving at least as much as their male counterparts. At the same time, they recognised that girls’ progress in education was limited by structural factors that had not yet been addressed: As we know girls are not as free as boys. They are not given the chance to go out and go to library and study like their male counterparts . . . They also do work at home and girls carry the burden of the household. So, I feel that they are still lagging behind in many ways. (Male stakeholder, zonal level, Tigray) A crucial challenge described by the stakeholder in the progressive group was the lack of attention and support given to this issue by other stakeholders, echoing the issues we raise in relation to our assessment of the ESDPs. For example, one stakeholder at the federal level described how: very limited [support is provided]. That is our challenge. The reason girls’ dropout is high because we do not give [this issue] especial emphasis . . . The support is limited compared to the need. (Male stakeholder, Ministry of Education) The lack of prioritisation of girls’ education resulted in insufficient funds being allocated which in turn constrained the ability of stakeholders within this group to make progress. Recognising the inadequacy of current strategies, more holistic and targeted approaches were called for, which would need the commitment not only of all stakeholders in the education system but also across different sectors (such as health) noting that ‘the education office cannot solve this problem’ (Male stakeholder, woreda level, Oromia).

Narrow views We identified just over one-quarter of stakeholders as having narrow views towards girls’ education. In this group, gender equality was equated with gender parity and little attention was given to the challenges faced by girls. Differences between girls were not acknowledged and understanding of girls’ education was

112  Louise Yorke, Pauline Rose and Alula Pankhurst largely based on stakeholder’s interpretation of official government statistics. In some locations where government statistics indicated that gender parity had been achieved, or where girls outnumbered boys, stakeholders referred to these figures to support their view that gender equality had been achieved and this was sufficient for addressing the inequalities faced by girls. Reflecting patterns in official government statistics, stakeholders in Addis Ababa, at both at the regional and woreda level, frequently described how there were ‘no problems related to gender disparity’ (male stakeholder) and that ‘in the schools we have a greater number of girls than boys’ (male stakeholder). Progress was measured by achieving the governments’ targets without giving further thought to the real experiences of girls. Similar views were captured outside of the capital. In SNNP, for example, a woreda official stated that: In terms of access for education, there is improvement . . . The gender parity index is also one now. This indicates that there is equal number of male and female students within our schools. So now we have to look for affirmative action for male students. (Male stakeholder, woreda level, SNNP) Perhaps even more surprisingly, some stakeholders also held the view that gender parity had been achieved in the emerging regions of Benishangul Gumuz and Somali where overall enrolment remains low. For example, a view was expressed that girls were now performing educationally ‘on the same level as boys’ (Male stakeholder, regional level, Benishangul Gumuz) and that, while in the past girls had been disadvantaged, ‘nowadays, but they are supported by affirmative actions to compete with the men’ (ibid.). Overall, in some cases, these views appear to reflect the hidden biases of stakeholders (which may be both conscious and unconscious), and that appeals to enrolment figures were used to justify these biases. While it may be that special support is required for certain groups of boys in this region, the idea that progress in girls’ education has somehow undermined boys’ education indicates at best a lack of comprehension of the nature of disadvantage faced by girls.

Mixed views The category mixed was the largest category, accounting for nearly one-half of the stakeholders and reflects most closely ideas within the education policy domain as to how progress in girls’ education is conceptualised and pursued. Like those in the progressive category, these stakeholders recognised the intersecting nature of challenges faced by girls, including the persistence of gender norms and noted how different groups of girls experienced challenges. At the same time, progress in girls’ education was viewed mainly in terms of equal numbers of boys and girls enrolled. Proposed strategies for improving girls’ education included the provision of more inputs and resources, similar to strategies outlined in the

Influence of politics on education in Ethiopia 113 education plans. They also included helping girls to withstand the challenges they faced with little attention given to underlying structural disadvantages faced by girls, which perhaps explains the persistence of these challenges: The school has to play its own role through organising club[s] and student council[s]. This will help them to withstand traditional cultural influences. The teacher has to support them by arranging separate tutorial class for them. (Male stakeholder, woreda level, SNNP) It seems that stakeholders within this category have in some ways internalised ideas within the education policy domain as to what is needed for progress in girls’ education. In summary, considerable heterogeneity is evident in terms of the attitudes of the stakeholders that we interviewed, which we placed along a continuum ranging from progressive to narrow, with the majority of stakeholders falling somewhere in between. While the commitment of stakeholders to girls’ education is important, we find that their understanding of progress in girls’ education also matters, which in turn affects the strategies and approaches they believe are needed. Reflecting on the differences in stakeholders’ views towards gender equality in education, we find that ideas and institutions (formal and informal), and their interactions, have a role to play. Within the constraints of formal institutions, those with progressive attitudes identify some progress in increasing access but acknowledge that much is yet to be done, but that they are limited in many ways, including difficulties in engaging others and the shortage of resources available for girls’ education. For those with mixed attitudes, while they may also be committed to improving girls’ education, their understanding of what is needed to ensure such progress may compromise their ability to achieve this goal. In failing to take account of the more deep-seated inequalities faced by girls they may be left wondering why progress is difficult to achieve. We also noted the persistence of more hidden and underlying forms of gender biases held by a minority of stakeholders, such as those who were opposed to gender quotas or believed that progress in girls’ education undermined that of boys. Claims that gender equality has been achieved and targets have been met are often used to justify these hidden biases. In some cases, not only are strategies for girls’ education not sufficiently transformative, but stakeholders are not incentivised to implement them, even where formal strategies have been pursued to attempt to foster a commitment to girls’ education. Our analysis of these patterns over time has revealed how little change has taken place over the course of 20  years demonstrating how embedded gender bias (nominal and substantive) are. It is perhaps naïve therefore to think that the education system is equipped to address the intractable problem of gender norms and practices found at the local level without first addressing these biases in its own structures. In the next and final section, we reflect more on the insights offered through our analysis of the politics of girls’ education and consider the

114  Louise Yorke, Pauline Rose and Alula Pankhurst implications for addressing the gap between ambitious policies and meaningful and sustainable progress for all girls.

Discussion and conclusion In this chapter, we have shown how taking account of the politics of girls’ education helps to reveal important insights concerning the gap between high-level commitments to girls’ education and slow and uneven progress in this area. The domains of power framework (Hickey & Hossain 2019) has helped to reveal the importance of the interaction between the political settlement and the education policy domain and how this shapes the uptake and implementation of strategies for girls’ education. Yet, in understanding patterns of progress in relation to girls’ education, we have also demonstrated the need to take account of gender inequality within the education policy domain (Chappell  & Waylen 2013; Sen 2007). Based on our analysis a schematic representation of how we understand the politics of girls’ education is provided in Figure 6.3. Our analysis has shown how the political context has to some extent shaped what is possible in relation to girls’ education in the education policy domain. The dominant developmental state of Ethiopia has produced a specific political context, which over time has seen increasing centralisation of power and an ambitious development agenda that gives special importance to education and gender equality. Within this context, Ethiopia has made rapid gains in recent years driven by rapid economic growth and infrastructure expansion towards the goal

Ideas underpinning development e.g. rapid economic growth

Women’s representation in the education system

Policy, plans and strategies for girls’ education

Political settlement e.g. dominant developmental coalition

Information institutions e.g. gender norms and expectations

Stakeholders’ views of girls’ education e.g. progressive, mixed, narrow

Formal institutions e.g. top-down governance

Implementation of strategies to improve girls’ education

Progress in girls’ education

Figure 6.3  Categorisation of stakeholders’ attitudes towards girls’ education

Influence of politics on education in Ethiopia 115 of reaching lower middle-income status by 2025. Yet, in the midst of these gains, inequalities have increased, and growing fragmentation has taken place within the country. The assumption that progress in addressing underlying disadvantages and structural inequalities would occur once economic growth was complete has not borne fruit, and the absence of effective strategies for addressing underlying structural disadvantages has affected those who were the most marginalised. Widely held ideas concerning how development happens, have shaped policies, plans and strategies that are pursued in relation to girls’ education in the education policy domain and this has constrained the range of available choices for girls’ education (Hickey & Hossain 2019). Rapid progress has been made in terms of access and some progress has also been made in gender parity represented by the decreasing gender gaps in enrolment, although worryingly these gaps are opening once again. Yet beyond this visible progress, substantive progress is still lacking, including real improvements in the lives of those who are most disadvantaged and hidden aspects of gender equality have not been addressed. While understanding the wider political context undoubtedly provides important insights into patterns of progress in girls’ education, we have also highlighted the need to take account of informal institutions within the education system, which cannot be considered to be a gender-neutral site. Nominally, women continue to be absent from positions of power in the education system while substantively ideas about the nature of gender roles influence the attitudes, behaviour and action of stakeholders and limits the progress that can be made. Nevertheless, individual actors can make a difference and throughout the government’s trajectory, the importance of female figures committed to advancing gender equality has helped to ensure a continued focus on gender equality both in the education policy domain and in the wider political settlement. In more recent times, positive progress has been made in increasing women’s representation in the wider political context. The first female president of Ethiopia, Sahle-Work Zewde, appointed in 2018, promised to ensure that the empowerment of women remains firmly on the government’s agenda. This was followed by the appointment of Ethiopia’s first gender-balanced cabinet, along with the entry of women into a number of high-level positions9 many of whom have been strident advocates of gender equality offering outspoken and often critical views of the government throughout their careers. Their presence suggests a commitment on the part of the government to real change. Nevertheless, our analysis demonstrates how we cannot assume that this progress will automatically translate into benefits for girls and women, and there is an urgent need to also address more substantive forms of gender inequality, both in the wider society and within the education system. Drawing on these insights, we suggest that a more transformative approach to girls’ education is needed. It will be important to move beyond simplistic notions that increasing girls’ access and retention will lead to greater gender equality. In formulating such an approach, it will be important to recognise the diversity of girls’ experiences and to draw upon their lived experiences. Importantly, we have highlighted how engaging all stakeholders in ensuring gender equality within

116  Louise Yorke, Pauline Rose and Alula Pankhurst the education system may require a more nuanced approach than simply shifting between mainstreaming and targeted approaches. Finally, while understanding the politics of girls’ education has provided valuable insights into the progress in girls’ education and has helped to provide important insights for addressing gaps in relation to girls’ education, we recognise that this is only one aspect of a much wider picture on girls’ education. As such, there is a need for further work to understand how this links to what takes place at the school level, including girls’ lived experiences.

Notes 1 In Ethiopia, primary education currently consists of eight years of schooling, including lower primary education (grades 1–4) and upper primary education (grades 5–8). Children usually enter into grade 1 around the age of seven years old. 2 RISE Ethiopia has examined whether and how a large package of national reforms works to improve learning equitably. More information is available at www.riseprogramme.org/countries/ethiopia 3 Addis Ababa, Amhara, Benishangul Gumuz, Ethio-Somali, Oromia, SNNP and Tigray. 4 In addition to the authors of the chapter, this included RISE Ethiopia members at Addis Ababa University (Prof. Amare Asegdom, Dr Belay Hagos, Prof. Darge Wole, Prof. Girma Lemma, Prof. Tirussew Tefera) and the Ethiopian Policy Studies Institute (Mr Chanie Eligu), Cambridge University (Dr Padmini Iyer) and Harvard University (Ms Shelby Carvalho). 5 The EPRDF consists of four parties: the Tigray People’s Liberation Front (TPLF), the Amhara Democratic Party (ADP), the Oromo Democratic Party (ODP) and the Southern Ethiopian People’s Democratic Movement (SEPDM). Our chapter was written prior to December 2019 when Dr Abiy Ahmed disbanded the EPRDF and formed a new political party, the prosperity party. While our analysis focuses on the period before December 2019, this does serve to highlight the rapidly changing political context in Ethiopia. 6 Dr Gennet Zewide, personal correspondence. 7 The Ministry of Education has recently been divided into two sections: the Ministry of Education and Ministry of Science and Higher Education (MOSHE), Dr Hirut Woldemariam, a former professor at Addis Ababa University was appointed Minister of the MOSHE. 8 Given her role of prominence and easy identifiability, Dr Zewide granted us permission to identify her in this chapter. All other interviews are anonymised. 9 This included the President of the Federal Supreme Court of Ethiopia (Meaza Ashenafi) and Head of the Election Board (Birtukan Mideksa).

References Asgedom, A., Hagos, B., Lemma, G., Rose, P., Teferra, T., Wole, D.,  & Yorke, L. (2019) Whose Influence and Whose Priorities? Insights from Government and Donor Stakeholders on the Design of the Ethiopian General Education Quality Improvement for Equity (GEQIP-E) Programme. [Online] Research on Improving Systems of Education (RISE) Insight Note. Available from: www.riseprogramme.org/blog/ whose-influence-whose-priorities-ethiopias-geqip-e [Accessed 12 May 2020]. Blystad, A., Haukanes, H., & Zenebe, M. (2014) Mediating development? Exchanges on gender policies and development practices in Addis Ababa, Ethiopia. Africa Today, 60(4), pp. 25–45.

Influence of politics on education in Ethiopia 117 Boyden, J., Pankhurst, A., & Tafere, Y. (2012) Child protection and harmful traditional practices: Female early marriage and genital modification in Ethiopia. Development in Practice, 22(4), pp. 510–522. Burgess, G.L. (2012) When the personal becomes political: Using legal reform to combat violence against women in Ethiopia. Gender, Place and Culture, 19(2), pp. 153–174. Central Statistical Agency [Ethiopia] (2016) Ethiopia Demographic and Health Survey 2016. Addis Ababa, Ethiopia, and Rockville, MD: Central Statistical Authority and ICF. Chappell, L., & Waylen, G. (2013) Gender and the hidden life of institutions. Public Administration, 91(3), pp. 599–615. Chuta, N., & Crivello, G. (2013) Towards a ‘Bright Future’: Young People Overcoming Poverty and Risk in Two Ethiopian Communities. [Online] Young Lives Working Paper 107. Young Lives, Oxford. Available from: www.younglives.org.uk/content/towards-bright-future-young-people-overcoming-poverty-and-risk-ethiopia [Accessed 12 May 2020]. Chuta, N.,  & Morrow, V. (2015) Youth Trajectories through Work and Marriage in Rural Ethiopia. [Online] Young Lives Working Paper No. 135. Young Lives, Oxford. Available from: www.younglives.org.uk/content/youth-trajectoriesthrough-work-and-marriage-rural-ethiopia [Accessed 12 May 2020]. Clapham, C. (2018) The Ethiopian developmental state. Third World Quarterly, 39(6), pp. 1151–1165. Clapham, C. (2019) The political economy of Ethiopia from the imperial period to the present. In Cheru, F., Cramer, C., & Oqubay, A. (Eds.) The Oxford Handbook of the Ethiopian Economy. Oxford: Oxford University Press. Colclough, C., Rose, P.& Tembon, M. (2000) Gender inequalities in primary schooling: The roles of poverty and adverse cultural practice. International Journal of Educational Development, 20(1), pp. 5–27. Di John, J., & Putzel, J. (2009) Political Settlements: Issues Paper. [Online] Discussion Paper. University of Birmingham, Birmingham, UK. Available from: http:// epapers.bham.ac.uk/645/ [Accessed 12 May 2020]. Fiseha, A., & Gebresilassie, F.H. (2019) The interface between federalism and development in Ethiopia. In Cheru, F., Cramer, C., & Oqubay, A. (Eds.) The Oxford Handbook of the Ethiopian Economy. Oxford: Oxford University Press. Goetz, A.M. (1997) Getting Institutions Right for Women in Development. London: Zed Books. Goetz, A.M. (2019) From transformative policy to transforming political settlements. In Nazneen, S., Hickey, S., & Sifaki, E. (Eds.) Negotiating Gender Equity in the Global South: The Politics of Domestic Violence Policy. London and New York: Routledge. Government of Ethiopia (1995) Constitution of Federal Democratic Republic of Ethiopia. Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Hagmann, T., & Abbink, J. (2012) Twenty years of revolutionary democratic Ethiopia, 1991 to 2011. Journal of Eastern African Studies, 5(4), pp. 579–595. Hickey, S, & Hossain, N. (Eds.) (2019) The Politics of Education in Developing Countries: From Schooling to Learning. Oxford: Oxford University Press. Levy, B., Cameron, R., Hoadley, U., & Naidoo, V. (2019) Political transformation and education sector performance in South Africa. In Hickey, S., & Hossain, N. (Eds.) The Politics of Education in Developing Countries: From Schooling to Learning. Oxford: Oxford Scholarship Online, p. 105.

118  Louise Yorke, Pauline Rose and Alula Pankhurst Ministry of Education (1994) Education and Training Policy. Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Ministry of Education (2005) Education Statistics Annual Abstract 1997 E.C. (2004/05). Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Ministry of Education (2015) The Education Sector Development Programme V. Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Ministry of Education (2017) Education Statistics Annual Abstract 2009 E.C. (2016/17). Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Ministry of Women’s Affairs [MoWA] (2006) National Action Plan for Gender Equality (NAP-GE) 2006–2010. Addis Ababa: The Federal Democratic Republic Government of Ethiopia. National Planning Commission (2016) Growth and Transformation Plan 2015/16– 2019/20. Volume I: Main Text. Addis Ababa: The Federal Democratic Republic Government of Ethiopia. Pankhurst, A. (2014) Child Marriage and Female Circumcision (FGM/C): Evidence from Ethiopia. [Online] Young Lives Policy Brief. Young Lives. Available from: www.younglives.org.uk/content/child-marriage-and-fgm-evidence-ethiopia [Accessed 12 May 2020]. Rose, P. (2003) Can Gender Equality in Education Be Attained?: Evidence from Ethiopia. Background paper for 2003 UNESCO Global Monitoring Report. Semela, T., Bekele, H., & Abraham, R. (2019) Women and development in Ethiopia: A sociohistorical analysis. Journal of Developing Societies, 35(2), pp. 230–255. Sen, G. (2007) Informal institutions and gender equality. In Jütting, J., Drechsler, D., Bartsch, S., & de Soysa, I. (Eds.) Informal Institutions: How Social Norms Help or Hinder Development. Development Centre Studies. Paris: OECD Publishing. Stavropoulou, M., Gupta-Archer, N., & Marcus, R. (2017) Adolescent Girls’ Capabilities in Ethiopia: The State of the Evidence on Programme Effectiveness. [Online] Gender and Adolescence: Global Evidence (GAGE), Overseas Development Institute (ODI). Available from: http://docplayer.net/83106541-Adolescent-girls-capabilities-in-ethiopia-the-state-of-the-evidence-on-programme-effectiveness-maria-stavropoulou-and-nandini-gupta-archer.html [Accessed 12 May 2020]. UNESCO (2014) Teaching and Learning: Achieving Quality for All; EFA Global Monitoring Report, 2013–2014. Global Education Monitoring Report. UNESCO, Paris. Available from: https://unesdoc.unesco.org/ark:/48223/pf0000225660 [Accessed 18 May 2020]. Unterhalter, E., & North, A. (2019) Education, Poverty and Global Goals for Gender Equality: How People Make Policy Happen. London and New York: Routledge. Vaughan, S. (2011) Revolutionary democratic state-building: Party, state and people in the EPRDF’s Ethiopia. Journal of Eastern African Studies, 5(4), pp. 619–640. Williams, T.P. (2019) The downsides of dominance: Education quality reforms and Rwanda’s political settlement. In Hickey, S., & Hossain, N. (Eds.) The Politics of Education in Developing Countries: From Schooling to Learning. Oxford: Oxford University Press. Woldehanna, T., & Araya, M.W. (2019) Poverty and inequality in Ethiopia. In Cheru, F., Cramer, C., & Oqubay, A. (Eds.) The Oxford Handbook of the Ethiopian Economy. Oxford: Oxford University Press. Woldetsadik, G. (2013) National Learning Assessment in Ethiopia: Sharing Experiences and Lessons. [Online] World Bank. Available from: https://olc.worldbank.org/ sites/default/files/Session_6_Ethiopia_Woldetsadik.pdf [Accessed 12 May 2020].

Influence of politics on education in Ethiopia 119 World Bank (2018) World Development Report 2018: Learning to Realize Education’s Promise. Washington, DC: World Bank. Zenawi, M. (2012) States and Markets: Neoliberal limitations and the case for a developmental state. In Noman, A., Botchwey, K., Stein, H., & Stiglitz, J.E. (Eds.) Good Growth and Governance in Africa: Rethinking Development Strategies. Oxford: Oxford University Press. Zewide, G. (2018) No One Left Behind: Redesigning the Ethiopian Education System. Ethiopia: Mega Publishing and Distribution.

Part II

Challenges and opportunities in addressing inequalities through education

7 Overriding social inequality? Educational aspirations versus the material realities of rural families in Pakistan Arif Naveed This chapter illustrates how the rise of basic schooling in contexts like rural Punjab in Pakistan is experienced and mediated within already stratified social and cultural spaces. It argues that poverty and social inequality generate a range of differentiated values, meanings, aspirations and schooling decisions amongst families with complex, and often compounding, effects on their strategies to use schooling to improve their living conditions. For the efforts to universalise educational access to be effective, it is therefore necessary to take account of such cultural complexities by addressing the forms of social stratification that can be found both in and through schooling. This is particularly important since some 22.84 million school-going-age children in Pakistan were reported to be out of school (GOP 2018), with poverty often cited as the main reason for this pattern, despite the unprecedented expansion of the education system since the new millennium. Given persistent under-funding of education in Pakistan (Malik & Rose 2015), national reforms continue to grapple with the issue of improving access of the traditionally excluded in ‘supply-driven’ ways that often undermine the apparently agentic yet socially structured decision-making of the poor. I argue in this chapter that it is in the process of decision-making that social inequality is revealed and often reproduced through educational choices. The multifaceted social hierarchy in Pakistan embedded within the education system and the labour market is associated with differentiated private returns to schooling with potentially lower (expected and actual) returns to, and higher costs of, schooling for the socially disadvantaged, especially compared with those favourably positioned in the social order. Improving access to education under such conditions could reduce the private cost of schooling for the poor but cannot, on its own, increase the private economic returns of schooling in a socially shaped labour market. Neither does it necessarily address everyday subjugation often experienced by the poor. In the absence of wider social and economic reforms aimed at equalising economic opportunities, socially disadvantaged groups have little to gain from investing in schooling as the levels and quality realistically achievable for them do not necessarily offset their prior disadvantage. It is crucial to recognise that the educational strategies of the poor, like others, are guided by societal norms and wider ideas associated with progressing towards

124  Arif Naveed a better life (Appadurai 2004) which are formed in a complex web of social relations. Educational goals, thus formed, are revised in light of experiences within and outside the education system. Schooling is situated in families’ long-term intergenerational quest to achieve a life that offers their members opportunities to progress, collectively and individually. When the promised route out of poverty and towards dignified living through schooling is seen to break down, little is left for the disadvantaged to use to meet the targets of the global development agendas. At this point, no amount of supply-driven educational reform or even legislation for compulsory schooling can help universalise access to schooling; rather, advancing the interests of those already privileged is more likely. This chapter has five sections. The first provides a brief description of the theoretical framework that I  use to conceptualise the relationship between social inequality and schooling. I follow with a description of the longitudinal research I conducted in a village in Punjab where I attempted to tap the aspirations for, and the reality of, social mobility amongst rural families. My analysis of the second-round interview data I collected in 2016 asks what values and meanings are attached to the role of schooling in families’ pursuit of social mobility and to what extent these are differentiated by the different positioning of the families within rural social order. I then explore how such socially structured meanings and values shape differentiated aspirations for achieving social mobility through schooling and different family trajectories. The final section concludes with a discussion on the need for socio-economic reform as a condition, rather than an outcome, of educational expansion in the Global South.

Conceptualising the rural social structure Many factors restrict the uptake of schooling and mediate its impact on poverty. The day-to-day politics of schooling and the labour market shape the objective possibilities of the poor through their subjective realisations. Arjun Appadurai’s (2004) concept of the capacity to aspire encourages us to appreciate the engagement of the poor with the cultural regimes that govern their lives and thus to uncover this micro-level, cultural politics of schooling. Situating schooling within aspirational worlds reveals its relationship with beliefs about material conditions, social relations and general norms about what constitutes a ‘good life’, to which the ‘choices made’ and the ‘choices voiced’ by the poor are attached (ibid., p. 68). For Appadurai, those with greater resources, status and power have richer experiences that help them ‘in navigating the complex steps between “norms and specific wants and wishes” ’ (ibid., p. 68). The poor, in contrast, have ‘less easy archiving of alternative futures’ and a ‘weaker sense of the pathways from concrete wants to intermediate contexts to general norms and back again’ (ibid., p. 69). With ‘practice, repetition, exploration, conjecture, and refutation’ necessary for their capacity to aspire (ibid., p. 68), poor families are inherently disadvantaged as they tend to have too much at stake to make risky decisions (Wood 2003) for the sake of gaining experience.

Overriding social inequality? 125 Appadurai’s insights have been extended and complemented by many, including Debraj Ray (2003) whose notion of an aspiration window (formed from individuals’ cognitive worlds) enables us to see individuals drawing information by comparing themselves with others. Contexts that are characterised by higher perceived social mobility and less polarisation provide a broader aspiration window that can, in turn, inspire higher levels of aspirations. Here, the cultural analysis and the framing of capacity to aspire of both Appadurai and Ray points to the significance of relationality. After all, it is in the ‘thick of social life’ that individuals’ meanings, values, preferences, aspirations and strategies for a good life are shaped, and a consensus is formed (Appadurai 2004, p. 67). White’s (2017, p.  133) conception of well-being makes a case for adopting an ontological position that sees ‘relationality as logically prior to individuals’. This avoids the risk of placing greater responsibility on individuals of what is an outcome of the collective  – the social structure (Sointu 2005; Ahmed 2004). A relational ontology addresses an important omission in educational research by recognising the role of families (and by focusing on their stratified positioning in their communities) in educational processes, thus generating deeper insights into the cultural complexity of schooling. Such a focus broadens our understanding of the role of schooling in the lives of the rural poor by revealing the power relations that shape their subjectivities and aspirations and contribute towards the reproduction of inequality (Hart 2012). In this context, Bourdieu’s (1977) concept of habitus is appealing given its remarkable complementarity with Appadurai’s notion of capacity to aspire. Both concepts bring together the objective and subjective, structure and agency, helping us understand how a person’s position in the social order shapes their schemes of thought, perceptions and tendencies – the day-to-day practices in determining their life chances in and through schooling. Habitus, as a set of durable and transposable dispositions, refers to the embodied form of objective structures, internalised through socialisation (Bourdieu 1977, p.  87). It represents the ‘mental and cognitive structure through which individuals perceive, understand, appreciate and deal with the social world’ (p. 76), thus recognising the mechanisms through which their aspirations are formed and adjusted to their experiences of material reality. It acknowledges that the ‘practical knowledge’ of social actors is constitutive of the reality of the social world (p. 142). Using this argument, Madeleine Arnot and I developed a ‘Habitus Listening Guide’1 with which to explore how young people negotiate the relational dispositions, ambitions and values within their families and the implications for their consequent educational choices and work experiences (Naveed  & Arnot 2019; Naveed 2019a). Collecting and hearing deeper, dynamic, family narratives are crucial in understanding the relationship between schooling, poverty and social inequality. For Bourdieu, the social structure is highly significant because of individuals’ sense of limits or sense of reality. His so-called theory of practice suggests that individuals think and take positions based on their relative place in social structure, the field, depending upon the quantity of their economic, social and cultural

126  Arif Naveed capitals (see Hilgers  & Mangez 2015). Examining the relationship between schooling and rural social inequality through this framework therefore requires an understanding of the social structure and hence of some of the key (in)formal institutions. It is to these I now briefly turn.

A pentagonal rural social structure By engaging with literature on the patterns of social inequality in rural Pakistan (specifically rural central Punjab), I identified what I call a pentagonal social structure consisting of five sets of power relations: (1) landownership; (2) kinship and caste; (3) religion; (4) patriarchy; and (5) the politics of patronage (Naveed 2019b). Whilst Pakistan is highly diverse with rich cultural traditions varying across its geography (Mohmand  & Gazdar 2007), my brief overview of these five elements helps us recognise the cultural complexity that were likely to have shaped the engagements with schooling of the rural families I met. (1) Landownership: The profound impact of landownership on educational outcomes is demonstrated in the highest upward educational mobility found amongst the landowners and lowest amongst the landless (cf. Cheema  & Naseer 2013). Such patterns reflect the fact that landownership, as a primary factor of production and social differentiation in the rural economy, has historically been the most prominent aspect of the social structure in rural Pakistan (Elgar 1960; Akhtar 2008). Landlords have historically positioned themselves as ‘middlemen’ between the villagers and the state, avoided successive political/land reforms and maintained their dominance of rural social life, despite the emergence of a range of intermediary occupational classes (Alavi 1974; Akhtar 2008; Javid 2012). (2) Kinship and caste structures: Early theorising of kinship in rural Pakistan (e.g. Alavi 1972 p. 25), suggested that kinship ‘embodies the primordial loyalties, structuring its social organisation’ and presents itself as ‘the key . . . dimension of economic, social and political interaction’ (Gazdar 2007, p.  87), affecting educational experiences and economic outcomes. In rural Punjab, structural inequalities of social status and power are associated with the ranking of one’s caste in the village (ibid.). Caste hierarchies permeate schools; teachers may discourage students from low-caste backgrounds with parents having little voice in teachers’ accountability (Tamim & Tariq 2015). Where schools are in a community dominated by high-caste groups, low-caste parents are less likely to school their daughters (Jacoby & Mansuri 2011). Overall caste identities strongly differentiate the increasing educational mobility over three generations of men in rural Punjab (Cheema & Naseer 2013). (3) Religion: Religion is central to the social organisation and hierarchy in Pakistan. Religious elites have a long history of dominating social, cultural and political life particularly in rural Pakistan, maintaining social order and resisting attempts to uplift the poor, socio-economically and educationally (Darling 1928; Malik & Mirza 2015).2 The communal division of colonial

Overriding social inequality? 127 Punjab in 1947, the resulting riots and the representation of religio-nationalism in educational discourses encouraged pejorative attitudes towards ‘religious others’. Not only can the history and image of religious minorities be distorted in educational content, but children from minority backgrounds can also face a hostile schooling environment (Hussain et al. 2011), limiting their prospects for social mobility. (4) Patriarchy: The patriarchal nature of the Pakistani society is manifested in all dimensions of the social structure, as power is greatly concentrated amongst men. Women benefit from their families’ advantaged position in the social hierarchy but also suffer disproportionally if their family is disadvantaged. Hence Pakistan is ranked at the bottom of various global indices of gender equality including education.3 While gender relations are likely to change over time, the traditional three-generational patrilocal households common across rural Pakistan are known to be sites which reproduce gender inequality (Kandiyoti 1988), mediating female life chances in and through education. (5) Relationships of patronage: In contexts where governments often fail to ensure the basic rights and the welfare of the poor, and markets fail to ensure equal competition for everyone, the poor rely on their personal relationships and community networks to ensure security in their lives and livelihoods (Wood 2003; Shami 2010). Such dependencies affect access to basic services since those who are less able to reciprocate favours are excluded (Chaudhry  & Vyborny 2013). Such relationships have had a detrimental effect on the education system; for example, in Khyber Pakhtunkhwa province, officials have adapted the ‘unofficial’ way of doing official business by exchanging favours with political elites (Khan 2012). These relationships are likely to embed labour markets, thus mediating the use of education for economic gain. This complex constellation of mutually reinforcing social hierarchies grant different resources and opportunities to different families and their members. It determines how far those in disadvantaged positions can aspire to a change in their circumstances through schooling. Families and their members with differentiated positionings in this rural social structure experience different processes of socialisation, shaping their habitus (Bourdieu 1977, p. 87). Consequently, deepseated differences resulting from such power relations affect how individuals perceive, understand, aspire, strategise and act (McNay 1999; Arnot & Naveed 2014) when using schooling to improve their lives. We can see here how far the reality of social inequality confronts the potentiality of education to reduce its effects.

Researching generational and gender worlds Given this theoretical framework, my empirical work focused on the relationships within (across gender and generations) and between families (differently located in village social hierarchies). In August–September 2016, I revisited the village,

128  Arif Naveed Chak Nagri (a pseudonym) in Central Punjab where I conducted fieldwork in 2009–10 for the Research Consortium on Educational Outcomes and Poverty (RECOUP).4 The village consisted of landowners, farmers, landless tenants and agricultural/non-agricultural labourers, some government employees and the owners of small businesses within and outside the village.5 In 2016, I re-interviewed the members (a father, mother, son and daughter aged 15–25 in 2010) of 10 families previously interviewed in 2010 (see Naveed & Arnot 2019). By 2016, most sons and daughters were married, and most daughters lived in other villages with their in-laws. Several sons were also away, living in various cities for work and returning home only periodically. However, I was able to re-interview the available members of seven of the families still living in the village or nearby (see Table 7.1).6 In 2016, I designed a new semi-structured interview schedule to explore the aspirations for, and experiences of, social mobility and how these were mediated by the rural social structure.8 Since social mobility is a social-scientific, Westerninspired concept largely based on urban social structures, I used the more familiar concept of taraqqi, an Urdu word of Arabic origin, commonly understood in nearly all local languages in Pakistan to uncover participants’ own core expressions and subjectivities and the cultural discourses underlying their interpretations of social mobility. Taraqqi captures several implied meanings of social mobility and can be translated literally as ‘progress’ or ‘improved social and economic status over time’. The conversations about taraqqi revealed the subtle dynamic underlying processes associated with social inequality that constrain social mobility and what contributions schooling could make to reduce inequality in families’ experiences. All interviews were audio-recorded in Urdu and transcribed. I coded the data in NVivo using a thematic frame. The selected quotes were then translated into English. All names were anonymised. Interviewees have been given a pseudonym, followed by the acronym indicating their status (M = mother, F = father, D = daughter and S = son) and by their years of schooling. The remainder of this chapter focuses on the pattern of responses and profiles of three of these families, first on their meanings, norms and values and then outlining how these shaped their aspirations through schooling. Table 7.1  The 2016 interview sample by the highest educational level achieved7 Category

Total interviewed

No/low schooling 0–2 years

Primary completed 5 years

Secondary and above 10+ years

Father Mother Sons Daughters

7 7 5 6

1 4 1 0

4 2 2 4

2 1 2 2

Overriding social inequality? 129

Socially structured meanings, norms and values An important insight offered by all the family members I interviewed was that schooling was seen as part of their long-term struggle to attain taraqqi. Whatever their circumstances, these families had a clear understanding of what they wanted to gain collectively and for their members by investing in schooling. In the wake of economic precarity, attaining secure and sustainable livelihoods and decent jobs was a crucial expected gain from schooling. For mothers like Nazima, Kiran and Irfana who had not been schooled, ‘when children get education and government gives them jobs, their taraqqi happens then’. Alternatively, for those unable to secure salaried employment, schooling was considered to help them run a karobar (i.e. all forms of economic activity with growth potential) as it gave them the required awareness and attitude: ‘one should be educated, have a functioning mind, have money, can buy things, this is what is taraqqi, that comes with hard work’ (Farheen, M, 5). Schooling was expected to provide employable skills, attitudes, values and knowledge that would be rewarded in labour market. A closer listening to participants’ voices revealed that the meanings given to taraqqi were structured along the lines of social status, economic position and gender. At one end of the spectrum were those who engaged in hard labour and considered having any stable source of income as an indication of taraqqi.9 Instability caused by consumption shocks, for example, resulting from illness or wedding expenses, often pushed the poor into indebtedness, or even bonded labour, thus postponing their taraqqi: ‘The lenders don’t let you eat away their loans. They ask to pay back first and then give wages’ (Irfana, M, 0). For such families, starting any karobar was a sign of economic independence thus an indication of upward social mobility as fathers Bakht and Khadim argued: ‘if one has a karobar then taraqqi happens’. No matter how small, independent karobar liberated unskilled workers from exploitative, seasonally fluctuating and uncertain casual work which often came with social control by employers and did not necessarily lead to socio-economic uplift. At the other end of the spectrum were those who had better schooling, and taraqqi for them meant an improvement in their social status and prestige along with economic progress. For those with relatively greater economic capital coming from landowning, privileged caste families, taraqqi did not mean merely securing livelihoods. It also meant ‘having power and authority’ to influence others preferably through high-status public sector jobs (Rahim, S, 12) which also granted them access to state authority. These different values and norms, and their meaning and understanding of a good life, had far-reaching implications for the educational possibilities for their young family members. Beyond occupational gains, the ability to read and write in certain ways was considered relevant to taraqqi. Schooling was seen to create a sharp distinction between the parrha-likha (schooled) and the unparrh (without schooling), with different habitus and trajectories for taraqqi. Schooling could thus potentially restratify the very social order depending on who could access it. Those who were better schooled could draw upon the embodied forms of cultural capital – thus

130  Arif Naveed distinguishing them from those with no schooling and positioning themselves as on a taraqqi track: The more there is education, the more a person would be refined from the inside; the more education you have in your society or at home, the more you will head towards taraqqi. (Shuja, S, 16) Those with religious dispositions had other commitments that they wanted to fulfil through schooling. Shuja, for example, described how schooling improved the ‘morals’ of society alongside peoples’ ‘ways to communicate’; it improved their saqafat (culture) and tehzeeb (civilization) and their maeeshat (economy), thus contributing to success in the broadest sense. Schooling was thus seen as both intrinsic to and instrumental in the process of taraqqi – both a means and an end in the pursuit of a better life. The centrality of family in defining each individual’s goals and values is seen in parents such as Akhtar for whom taraqqi was manifested in domestic life, such as in the quality of familial relationships, parenting and the household environment. Another crucial dimension of the socially structured meaning of taraqqi related to the gendered expectations placed on the schooling of sons and daughters in ways that reproduced the sexual division of labour. Investments in sons’ schooling were seen as an integral part of the intergenerational bargain (McGregor et al. 2000) as they were meant to ensure the welfare of their parents in their old age in addition to improving their own economic and social status. Under such norms, schooling was meant to prepare boys and girls separately for their respective roles in the family, community and economy. The utmost necessity for sons was viewed as their being able to keep the ‘household system well-functioning’ (Akhtar, F, 5). A good job would add to a family’s standing in the community and increase its ‘goodwill’ as people came to ‘greet them’, thus adding to their prestige. A son’s education was thus crucial for the whole family’s taraqqi. Amongst the sons with more years of schooling, those with higher and professional education were considered better placed to secure high-status jobs: ‘If a guy is becoming a doctor . . . [or] if someone is becoming an engineer, one knows he is doing taraqqi’ (Akhtar, F., 5). Taraqqi was associated, therefore, with urban professional work, salaried employment and skilled labour. Schooling was the best strategy for sons to fulfil these expectations: ‘Taraqqi for a young man is the completion of his education, getting a job, a permanent job and parents are also at peace that their son has [a] job’ (Rahila, M, 5). Those sons who failed to secure sustainable livelihoods were subject to temporary, hazardous work and a disappointment to ageing parents by failing to take up the provider’s role. Daughters, in contrast, had far fewer (in)formal paid employment opportunities in the rural economy; their schooling was essentially linked to unpaid work of providing care, raising children and contributing to the economic success of

Overriding social inequality? 131 their menfolk and children, whilst also adding to family honour by adhering to the accepted gender norms (see Arnot & Naveed 2014). Some parents wanted their daughters to get a ‘good education’ so as to get ‘good jobs’ – for example, to ‘become a teacher’ at a public school that offered good pay and job security or even at a low-paying private school. For economically aspiring families, women’s earnings – whether from a job or a karobar like ‘stitching clothes’ or ‘some kind of work at their homes like making dresses or tutoring children’ – were an important, often welcome contribution to the household economy. At the same time, the social and gender order embedded in the labour market severely limited such thoughts. By and large, the family and the social structure placed a woman’s taraqqi largely within gendered norms of domesticity, expecting them to be the primary provider of care to their current and future family members. Their schooling was meant to prepare them for such a role, which would create a ‘good household environment’ and demonstrate their family’s taraqqi: ‘One such educated woman in a family changes the entire environment of the family’ (Akhtar, F, 5). The normative situating of women within domesticity inevitably invoked the value of the reproductive and maternal roles that their schooling was expected to prepare them for. Like Akhtar, most parents believed that by schooling their daughters ‘their future generation will do taraqqi’. Marriage therefore offered a daughter a chance to move up socially while also setting the parents free from having to support her. A ‘good marriage’ could bring a young woman high social status, economic uplift and personal stability for the rest of her life and prestige and peace of mind for her parents, while also providing families with new social networks or strengthening existing ones. In contrast, a ‘wrong marriage’ could lower a daughter’s social and economic status and lead to her continued economic dependence on her parents. Families had to avoid such a fate for their daughters and their strategies usually protected the boundaries of caste, kinship and socio-economic status. On occasion, potential matches could come from the outside of these close-knit social sphere, opening up the possibility for change in social relations. An educated young woman would have better life chances by marrying into a well-off family, as she could offer a rich home environment, take good care of her husband’s family, raise the children intelligently and educating them better. The chances of upward mobility would also increase dramatically if an educated woman was lucky enough to find culturally appropriate employment, such as becoming a schoolteacher or a doctor. The proud declaration of father Khadim (a brick-kiln worker with no schooling) reflected this: ‘All my daughters are educated and now married . . . one still at home has BA; if she had found a job, she would have done taraqqi’. Such social structuring of values, meanings and norms, differentiated as they are for sons and daughters, underlie the aspirations and strategies for schooling. How, and if so whether, such meanings and values help perpetuate inequalities in and through schooling is addressed in the next section.

132  Arif Naveed

A sense of limits and realities Families that are hierarchically positioned in the pentagonal rural social structure have different educational and economic goals to aspire to. Their varied sense of limits and reality (Bourdieu 1977, p.  36), of feasibility, risks and opportunity costs are inscribed in the habitus of their members. Even the most marginalised families started with high desires for taraqqi through schooling – after all, the rise of mass schooling in rural communities hinges on the promise of an escape out of poverty and an improvement of life chances (Jeffery et al. 2008). The education system in such a context, however, may not be prepared to equalise and capitalise on such increased aspirations for taraqqi through schooling by offering equitable learning experiences to students from unequal social backgrounds. Any failure of schools attended by the poor in providing economically valuable skills and credentials (perpetuated by the increasing privatisation of schooling) and any discrimination in the labour market even when poor young people acquire these skills can result in narrowing of the aspirational window. Inequality in schooling and work transitions displays the material realities that familial aspirations may not easily escape. Through these dynamic processes, one can see limits to the potential of schooling in improving the life chances of all. Using a relational ontology and the Habitus Listening Guide for the analysis of my family interviews (Naveed & Arnot 2019), I was able to compare three Punjab families,10 revealing the dynamic processes in which their aspirations were shaped by their positioning in the pentagonal rural social structure and their intra-family cross-gender and generation perspectives which collided and reshaped the possibilities they saw of social change.

Rehmat’s family’s limited expectations and low educational outcomes My first example of this complex cultural tension between the capacity to aspire and the reality of social inequality is the family of Rehmat who was 47 at the time of my second interview. Rehmat came from a kinship group that had a low status in the caste hierarchy of the village. He was a landless village butcher who traded livestock on the side, an activity which was prone to seasonal fluctuations and losses.

Owing to a childhood living in poverty, Rehmat had not attended school nor had his wife Latifan. They had three sons and two daughters, but none had reached secondary school. Rehmat, Latifan and their daughter Itrat (who lived with her in-laws in a neighbouring village) revealed their relatively modest but also frustrated familial aspirations and the low outcomes of the schooling they had received.

Overriding social inequality? 133

Father Rehmat I tried hard to educate my children. The oldest son left school from 7th Grade. He was not going to school and started working with me. The one younger than him studied till 8th, then he also left studies and is also working with me now. The youngest one is studying in 6th Grade; we are trying that he continues his studies. . . . I wanted my sons to get education so that they understand themselves, understand worldly matters, [and] do their own good karobar. That’s what all parents desire for their children.  .  .  . Our poverty stopped us from fulfilling my dreams. When there are not enough resources, one can’t spend on education and then just quietly pulls [children] out of school. If I were educated, would have known what education is. My children would also have got education. When I myself am not educated, what could I tell them about education?

Mother Latifan We didn’t have education, what dreams could I have had then? I did think of a good home, a good karobar so as to have a few days of comfort but nothing happened. . . . Not a single desire has been fulfilled. . . . If children came home and told me that they have completed their lessons, how could I  know whether or not they did? Many times, I  have thrown out their important papers by mistake. One is blind from the eyes if one is not schooled  .  .  . I have spent [my] own time, it is gone now.  .  .  . My dreams about the daughters were that they should get married into beautiful homes, well-to-do families. . . . For sons, I want them to have good karobar, a nice home, that’s all I desire. I can only pray to Allah. What else can I do? I pray 5 times and ask Allah that only He can help me now, none else can do anything. Ah, what to feel about it all now? . . . I want that Allah may help in my sons’ karobar so that they buy some land for their house and I find some peace in my heart. I don’t have big dreams. I just want my children to have their own space, their own home in my lifetime.

Daughter Itrat (5 years of schooling) I wanted to become a teacher after my studies but . . . there was no system at home, how could they have educated me? There was a lack of money and the business was not good either. Whatever my father earned, we would spend on food by the end of the day. How could they have educated us then? . . . I don’t think much about the dreams that remained unfulfilled, I don’t feel good about those. . . . When I go [to village] and meet my friends and they tell me that they completed their education, I feel angry

134  Arif Naveed about myself.  .  .  . I  had thought of better than how my family is now. Their circumstances may change as you never know when something good happens. If anyone at home gets a job, my parents’ circumstances will improve. . . . Who were I to think about my in-laws [and marriage]? My family decided that it was good. My husband is son of my maternal aunt . . . one has many desires but whatever one gets is alright.

Rehmat’s case illustrates that even poor, landless parents who had never been to school start with hopes of schooling their children so as to achieve taraqqi. At the same time, these aspirations are limited to achieving karobar for sons and decent marriages for daughters offering a narrow aspirational window to parents as well as a ‘less easy archiving of the alternative futures’ (Appadurai 2004, p. 69). Given the fragile economy of Rehmat’s family, educational aspirations had to dovetail with the need to draw sons frequently into the family’s economic activity at a young age. It is in the material realities as well as social relations of the family that the struggle between short-term economic survival and the long-term possibilities through a better education occurs, taking sons out of school ‘against their parental will’ – a habitus involving a voluntarily rushing towards the inevitable future, as Bourdieu tells us. The intersection of family’s disadvantaged position in the social structure and its gendered meaning of taraqqi and values also implies the disruption of daughters’ schooling as soon as a reasonable marriage proposal arrives for them. They had to give up their high educational and occupational ambitions for marital arrangements in ways that offered them little or no agency in making crucial decisions about their lives. Another aspect of educational expansion in a context of major social inequality is the deepening of disadvantage through the internalisation of failure amongst the poor. This is evident in Itrat’s self-directed anger over her lack of educational success in comparison to her peers. An outcome of structural inequality is portrayed as disadvantage resulting from their personal incapacity. The discursive framing of schooling also caused symbolic violence (Bourdieu 1977, p. 237) on this family where those with low/no schooling believed their subjugation to be legitimate – as mother Latifan believed about herself ‘One is blind from the eyes if one is not schooled’. Such views misplace the responsibility of failure on the disadvantage rather than the social structure.

Bakht’s family’s high aspirations and poor outcomes Bakht was a brick-kiln worker. His family belonged to a religious minority; with a caste identity associated with refuse work, the lowest in the kinship hierarchy of the village (although the family had not engaged in this work for at least three generations). In the following, we see a marked gap between the family’s high aspirations for social mobility and their poor outcomes of schooling.

Overriding social inequality? 135 Bakht was schooled to primary level whilst his wife Irfana had not. They had five sons and three daughters. Their oldest son had not been to school and the second one had 10  years of schooling, a requirement for many low-level public sector jobs at that time. However, he could not find a job. The third son did not go to school and the fourth one only attended school for two years. Both started working with Bakht at the brick-kiln at an early age. Their youngest son was intelligent but dropped out of school from 7th grade. None of their daughters received any schooling; all were married at the time of the second interview. The educational aspirations of family members and their experiences are poignantly described in their own words.

Father Bakht My own father took care of my [second] son. He would take him to school and bring him back and would keep in contact with his teachers. That is why he could study. . . . He went to the city some 15 or 20 years ago [with] 10 years of schooling, and at that time, 10 years was a lot. I got tired of running around for his job but couldn’t find any. They were asking for Rs. 100,000 to 150,000 for a job.11 We are unable to earn our meals, where could we bring a lakh from? . . . Kamzor admi ko rishwat ke beghair job nahe miltee (a powerless man does not get a job without bribes). He is now running a donkey cart in the city like his oldest brother. The other [son] got 7 years of schooling. . . . He was sharp minded, always stood first in his class, scored better than sons of the rich people . . . looking at his older brother unable to get job, he left school . . . he learnt repairing cars, he could also work on tractors and motorcycles, whatever you name, . . . He went abroad, and has helped us a lot in constructing our house, but went out of contact for the last 3 years. Now I have asked my sons to start their own karobar. If that happens then some taraqqi is possible. This labour work doesn’t feed us properly, what kind of taraqqi is possible with it? But setting up [karobar] of our own would cost at least one or two lakh, we shall arrange this by working hard.

Mother Irfana When I was young, I had many dreams. I wished to marry into a prosperous family so that I could also have some comfort and do taraqqi . . . Had thought that there would be enough money, children should be educated and we should feel freed and experience taraqqi . . . nothing happened. My children are also like this, going in circles, doing daily labour, what taraqqi can they do? The plan was that sons would have education and

136  Arif Naveed have their own karobar. Had thought that we should work hard and save some money, to improve the lives of children . . . When my children were young and my husband was working hard at brick kiln . . . I used to work at people’s homes, and children started going with their father to the brick kiln. Their father could not educate them. He could not get them out of brick kiln, and taraqqi could not happen. What taraqqi can they do now? None of my dreams are fulfilled. Sons could not find better paying work so no dream was fulfilled . . . when dreams are not fulfilled, children could not do taraqqi, what can we do?

Son Imran (2 years of schooling, brick-kiln worker) I had many dreams when I was young. I thought of doing a lot of taraqqi, wanted to earn and feed parents. It could not happen due to poverty. I just kept going to work with my father. Doing one’s own karobar requires money, so it cannot be done because of poverty. We could only fulfil our dreams if we had any money, could buy cattle raise and sell them and earn some money. When we don’t have money, we cannot do that . . . our poverty is increasing over time.

The relational dynamics of schooling are most evident in Bakht’s case as two generations took it as their responsibility to school the third generation – a desperate strategy to break out of exclusion from the web of oppressive social relations in the village. The consequences of these dynamics became all the more evident as the family began to navigate the working of the labour market when even the secondary schooling of their son could not offset the family’s overlapping disadvantage of poverty, landlessness and low-caste status. The personal archiving of the experienced inability of schooling in helping the poor achieve taraqqi could be seen playing a crucial role in narrowing the educational aspirations for younger family members. When a secondary schooled older sibling ended up running a donkey cart like his eldest brother who had not been to school, it is not surprising that the younger, high-achieving brother chose to drop out of school opting for an alternative track to achieve taraqqi. It was not the lack of aspirations for taraqqi or schooling per say, it was the oppressive social relations governing educational outcomes that limited the ability of the poor to make schooling work for them. There are limits to what the expansion of mass schooling can contribute to the lives of socially disadvantaged like Bakht’s family apart from deepening the sense of betrayal, social exclusion and a widening of the gap from those favourably positioned in the local social order.

Aslam’s family’s high aspirations and high educational outcomes Aslam, 61, and his wife Kinza were both secondary schooled, retired primary school teachers. Kinza was one of the few women in the village who had had

Overriding social inequality? 137 formal employment. Both belonged to the privileged caste in the village. Kinza’s family owned more land than her husband. After retirement, the parents set up a successful fertiliser business.

Aslam and Kinza had two sons and two daughters, all of whom had completed secondary schooling and progressed to tertiary education. One daughter had earned a master’s degree. The aspirations of mother, son and daughter about education and taraqqi were relatively high, allowing them to reflect on a frustrating range of educational outcomes in the family (Aslam was not available for a second interview).

Kinza, Mother When I  was young, I  thought of doing the teacher training course after 10th grade and this dream was fulfilled. Then I became a teacher. . . . My parents helped me a lot. I  went to [a city name] for the teachers’ training and they used to visit me there. It was 1972, and very few people had education at that time . . . I take pride in all that I could achieve . . . I had wished my children could get a good education. I got the youngest daughter admission at the Degree College and then at the University, and fulfilled her desire to do a Master’s degree. One son did BA and the other did higher secondary only. . . . They themselves did not study further, there was no other hurdle. The oldest son went abroad and the younger also followed him. I had a lot more aspirations for taraqqi than what my family has achieved so far. I wanted my children to get higher education and do jobs. . . . They couldn’t get a public-sector job, but otherwise, Allah has his blessings on us.

Rahim, Son (13 years of schooling) My dream was to join the police. Like one becomes SHO (head of the police station) after completing an education and the whole area knows that there is an SHO, who works carefully, and does not take bribes. There is a fear about him in the area and everyone thinks he is an honest person, doesn’t misuse his office, and also provides security to everyone. That was my dream for a while. Then I started thinking about going abroad. I went abroad and spent some 3–4 years working there. Then it came to my mind that whatever hard work I am doing there, I could do the same efforts back home and do my own karobar and that would be much better, so why facing the worries of being out of home and the country. I just could not pay sufficient attention to my education as I was more involved in the family business. When one has to work on the business, then studies become only a formality. The age of 20 or 22 is not for running

138  Arif Naveed business, it is for studies. If one starts doing other work, it is not good for the studies. One should have a strong [social] background for having high ambitions and for pursuing them . . . I did not have any problems. My brother had gone abroad, and I had to take care of many things so could not focus on my studies. I could not go to university, and attended the college in the neighbouring village that did not have good educational standards. If I were in the city, in addition to attending university, I could also have joined some academy for private coaching and done better at my studies. If I had a slightly better education, I could have thought of looking for a job. There is nothing particular that I  wished and did not achieve. I  have worked very hard. My education is a bit low, but I  have paid full attention to my karobar Masha Allah. . . . We have done good taraqqi from the previous generation. Will take better care of our children so that they do further taraqqi.

Samina, Daughter, (12 years schooled) Some of my dreams are fulfilled and some remain unfulfilled. I think about those dreams and say thanks to Allah for whatever He has given me. Sometimes I feel I should have studied more. In my BA, I failed in English and I feel sad about it. I had desired for more and more taraqqi. I wished all our family needs were fulfilled by my husband. But that did not happen. The main reason is that my husband has only 10 years of schooling and these days, this much schooling has no value. If he had better education, he could have applied for some other job. I think about my children and wish they study well. I have said to their father that for the next 5 years, we should put them in good schools so that they have good educational foundation. Afterwards, whatever school he likes, he can send them to.

The striking feature here is the relative ease with which Kinza’s dream of becoming a schoolteacher, at a time when fewer women attended school was fulfilled. The privileged economic and social status resulting from landownership and higher caste identity of her parents meant that Kinza’s family could provide for the resources required for her teacher training in a different city and the foundations not only for a successful career as a schoolteacher but also widening the aspirational window for her family’s taraqqi. The relative shift in gender norms in Kinza’s own experience was evidently transmitted to the next generation. A sense of experimentation was reflected in her son Rahim’s ability to choose between options. The educational success of the older siblings, as Ray (2003) noted, provided compelling information for his own educational and economic aspirations.

Overriding social inequality? 139 Clearly, social change also raises aspirations. Whilst all children had achieved more than secondary school, their mother’s desire for their education was even higher and remained unfulfilled. A wide range of experiences, whether personal or those of other family members, helped Rahim to navigate ‘the complex steps between . . . norms and specific wants and wishes’ (Appadurai 2004, p. 69). The social structure which had worked against Bakht and Rehmat’s families instead offered resources to Aslam’s children in pursuit of their ambitions. Their failure to meet the educational aspirations of Kinza did not limit their aspirations for cultural and material resources to achieve taraqqi. Rahim’s complex experience of exploring a wide range of possibilities – education, employment opportunities in Pakistan, overseas and setting up a business in Pakistan and his disposition to aim for power, prestige, authority and economic success appear to have been driven by the privileged family position offering economic, social and cultural resources. At the same time, significant gender differences existed within Aslam’s relatively successful family. While sons went on to pursue economic opportunities overseas and set up successful businesses in Pakistan, his daughters lived as housewives after their marriages, unlike their mother who was a working woman. The marriage of the youngest university educated daughter into a major city-based well-off family was presented by the family as a marker of her taraqqi. The life chances of the other daughter were tied to the status of her husband who had less schooling than her. Marriage was crucial for daughters to realise their aspirations for taraqqi. Samina felt she had attained less taraqqi than she or her family wished for.

Discussion Common across all these case studies is that rural families, no matter how poor, start off with high aspirations for achieving a good life, collectively and for their individual members, and schooling is seen as integral to achieving this life goal. Those who are relatively privileged, however, have a better navigational capacity and better understanding of the social norms that they use ‘to explore the future more frequently and more realistically’(Appadurai 2004, p. 68) in achieving taraqqi in and through schooling that help them realise and improvise these aspirations. The desires and wants of those disadvantaged by the rural social structure, however, frequently confronted unfavourable material realities. Their experiences can force them to lower their aspirations down to the levels suitable for people like them. The insights in these data question Dreze and Sen’s (1995) rather optimistic view of schooling improving substantive freedoms suggesting that, in highly economically, socially and politically unequal contexts, schooling may instead perpetuate prevalent social hierarchies of caste, religion, class and gender and widen economic inequalities (see also Jeffrey et al. (2008) in the case of northern India). The expansion of basic schooling in the Global South as a gendered, classed project may well groom rural girls for a better version of docile domesticity, instead of empowering them economically (Jeffery et al. 2005).

140  Arif Naveed My aim here was to offer a new way of looking at and into the relationship between schooling and social inequality. By analysing the educational experiences of these families through the lens offered by Appadurai and Bourdieu, I  have illustrated the ways in which the cultural capacity to aspire is entangled with one’s material realities and place in social structure. After identifying the shape of what I  call a pentagonal rural social structure, I  have shown some of its inner working and dominance in the subjectivities of three rural families’ values, meanings, aspirations and their economic outcomes. I highlighted how the ideas of progress and a good life that provide the moral basis for individual and familial educational aspirations, strategies and practices are culturally constructed and reflect (although often not explicitly or intentionally) the power relations of the local social order. The complex interplay between structure and agency, and between cultural, economic and social resources can leave schooling as a mechanism for social reproduction, capable of perpetuating existing inequalities rather than encouraging social opening and mobility. The expansion of schooling is not therefore unproblematic in an unchanged social structure. It raises expectations with its promise of a better life. At the same time, in the absence of cultural and economic resources to realise these aspirations, it leads to considerable disappointment, low outcomes and frustrations and potentially greater social distance between not just those variably positioned in social structure but also between the schooled and those who had not been schooled (Arnot & Naveed 2014). For schooling to be a viable investment strategy, it must offer these poor, landless, socially excluded families a credible route towards improvement in their economic conditions and a dignified social status. The case of Bakht’s family makes clear that, even if very poor, some families are willing to take a leap into investing in education in the wake of economic hardships, even if the economic returns to the level and the quality of schooling they could access are uncertain. Quantitative analyses of chronic poverty across South Asian contexts indicate that, although schooling was important in preventing families from falling into poverty, once poor, it was either ineffective or higher levels of education were required in order to break out of poverty (see Naveed & Sutoris 2020 for details). This chapter has shed some light on what constrains the role of schooling in reducing social inequality. Social mobility for the poor appears to depend on processes that build on prior unequal access to opportunities and resources, with the chances of success strikingly limited. In estimating the average returns to each additional year of schooling for the entire population, the standard models of economic returns to schooling fail to take these inner workings of the social structure on mediating these returns. As Ray (2003, p. 8) notes: ‘there is no experience quite as compelling as the experience of your immediate family, and more broadly, those in your socio-economic and spatial neighbourhood’. By relying upon their own experiences and observations, they perform their cost–benefit analysis on a daily basis for educational decision-making, often recognising the risky nature of their educational investments compared with the low, but guaranteed, returns to engaging children in income-generating activities early on. The micro-level assessments of the value

Overriding social inequality? 141 of schooling for economic prosperity, social progress, equality and an overall better life suggest that the aspirations, which are shaped and restricted by the material realities of rural social life, need to be taken into account in educational policy debates. The family case studies discussed in this chapter demonstrate that the wider social, cultural and economic context, if not transformed, have the capacity to overpower the opportunity that schooling might have offered to the poor. Without creating a level-playing field, as a precondition, there is a little that schooling alone can achieve in reducing inequality and eradicating poverty in contexts such as rural Punjab.

Notes 1 The Habitus Listening Guide described four readings of interviews with father, mother, son and daughter (social structural; horizontal intergenerational; vertical gender and, mythic-ritual listenings). 2 One needs to recognise that religious schools also provide free Islamic education (including boarding and lodging) to the extreme poor children, across the country. 3 For example, Global Gender Gap Index produced by the World Economic Forum (2014) which is a composite measure of economic participation and opportunity, educational attainment, health and survival and political empowerment. 4 RECOUP investigated the social/economic outcomes of education for the poor in Pakistan, (see Colclough 2012). The first round of data was collected for the sub-project, ‘Youth, Gender and Citizenship’, (directed internationally by Madeleine Arnot) by a field research team based at the Mahbub-ul-Haq Human Development Centre, Islamabad which I headed. 5 Educational facilities in the village in 2010 included: two government primary schools for girls and a higher secondary school (teaching up to grade 12), a primary school for boys and a high school that taught only boys up to grade 10. Two private schools offered co-education. A  few small Madrasahs (seminaries) also offered religious education, which was common, particularly for girls (Naveed 2019b). 6 Given cultural norms, interviews with female members of the households were conducted by two female Research Assistants, Sidrah Saleem and Iqra Saleem. 7 The original purposive stratified sample recruited youth in these three categories of educational attainment. The three sets of parents in this second study also fitted these categories. 8 The interview covered various aspects of family history, educational aspirations, experiences and biographies. 9 Such as primary schooled son Nisar and son Imran who had two years of schooling. 10 The interviews were conducted and analysed in Punjabi. I  translated selected quotes from Punjabi to English. 11 In 1995, this would be approximately three to five times Bakht’s annual wage.

References Ahmed, S. (2004) Affective economies. Social Text, 79(22:2), pp. 117–139. Akhtar, A.S. (2008) The Overdeveloping State: The Politics of Common Sense in Pakistan, 1971–2007. PhD thesis, SOAS, London. Alavi, H. (1972) Kinship in West Punjab villages. Contributions to Indian Sociology, 6(2), pp. 1–27.

142  Arif Naveed Alavi, H. (1974) Rural bases of political power in South Asia. Journal of Contemporary Asia, 4(4), pp. 413–422. Appadurai, A. (2004) The capacity to aspire: Culture and the terms of recognition. In Rao, V., & Walton, M. (Eds.) Culture and Public Action. Stanford, CA: Stanford University Press, pp. 59–84. Arnot, M., & Naveed, A. (2014) Educational outcomes across the generational and gender divide: The rural family habitus of Pakistani families living in poverty. Gender and Education, 26(5), pp. 505–523. Bourdieu, P. (1977) Outline of a Theory of Practice (Nice, R., Trans.). Cambridge: Cambridge University Press. (Original work published in 1972). Chaudhry, A., & Vyborny, K. (2013) Patronage in rural Punjab: Evidence from a new household dataset. The Lahore Journal of Economics, 18, pp. 183–209. Cheema, A., & Naseer, M.F. (2013) Historical inequality and intergenerational educational mobility: The dynamics of change in rural Punjab. Lahore Journal of Economics, 18, pp. 211–231. Colclough, C. (Eds.) (2012) Education Outcomes and Poverty in South Asia: A Reassessment. London: Routledge. Darling, M. (1928) The Punjab Peasant in Prosperity and Debt. Oxford: Oxford University Press. Dreze, J.,  & Sen, A. (1995) India: Economic Development and Social Opportunity. Oxford: Clarendon Press. Elgar, Z. (1960) A Punjabi Village in Pakistan. London: Columbia University Press. Gazdar, H. (2007) Class, caste or race: Veils over social oppression in Pakistan. Economic and Political Weekly, 42(2), pp. 86–88. Government of Pakistan (2018) Pakistan Education Statistics 2016–17. Islamabad: Academy of Educational Planning and Management. Hart, C.S. (2012) Aspirations, Education and Social Justice  – Applying Sen and Bourdieu. London: Bloomsbury. Hilgers, M., & Mangez, E. (Eds.) (2015) Bourdieu’s Theory of Social Fields: Concepts and Applications. Abingdon: Routledge. Hussain, A., Salim, A.,  & Naveed, A. (2011) Connecting the Dots: Education and Religious Discrimination in Pakistan: A Study of Public Schools and Madrassas. US Commission on International Religious Freedoms. Available from: www.uscirf. gov/sites/default/files/resources/Pakistan-ConnectingTheDots-Email(3).pdf Jacoby, H., & Mansuri, G. (2011) Crossing Boundaries: Gender, Caste and Schooling in Rural Pakistan. World Bank Policy Research Working Paper No. 5710. Washington, DC: World Bank. Javid, H. (2012) Class, Power, and Patronage: The Landed Elite and Politics in Pakistani Punjab. PhD thesis, The London School of Economics and Political Science, London. Jeffery, P., Jeffery, R., & Jeffrey, C. (2005) The mother’s lap and the civilising mission: Madrasah education and rural Muslim girls in western Uttar Pradesh. In Hassan, Z., & Menon, R. (Eds.) The Diversity of Muslim Women’s Lives in India. New Delhi: Oxford University Press, pp. 108–148. Jeffrey, C., Jeffery, P., & Jeffery, R. (2008) Degrees Without Freedom? Education, Masculinities and Unemployment in North India. Stanford, CA: Stanford University Press. Kandiyoti, D. (1988) Bargaining with patriarchy. Gender and Society, 2(3), pp. 274–290.

Overriding social inequality? 143 Khan, S. (2012) The Un-official Performance of Official Business in Pakistan: The Interface with State Bureaucracy and Citizens. PhD thesis, University of Bath, Bath. Malik, A., & Mirza, R. (2015) Religion, land and politics: Shrines and literacy in Punjab, Pakistan. Pakistan Strategy Support Programme, Working Paper No. 030, May. Malik, R.,  & Rose, P. (2015) Financing Education in Pakistan: Opportunities for Action. [Online] Country case study for the Oslo Summit on Education for Development. Available from: https://reliefweb.int/sites/reliefweb.int/files/ resources/pakista.pdf [Accessed 5 August 2020]. McGregor, J.A., Copestake, J.G., & Wood, G.D. (2000) Inter-generational bargain: An introduction. Journal of International Development, 12(4), pp. 447–451. McNay, L. (1999) Gender, habitus, and the field: Pierre Bourdieu and the limits of reflexivity theory. Theory, Culture & Society, 16(1), pp. 95–117. Mohmand, S.K., & Gazdar, H. (2007) Social Structure in Rural Pakistan. Determinants and Drivers of Poverty Reduction and ADB’s Contribution in Rural Pakistan. Islamabad: The Asian Development Bank. Naveed, A. (2019a) A Habitus Listening Guide for Exploring Educational and Social Inequality. [Online] The Sociological Review Blogpost. Available from: www.thesociologicalreview.com/a-habitus-listening-guide-for-exploring-educa tional-and-social-inequality/ [Accessed 30 July 2019]. Naveed, A. (2019b) Reconceptualising the Role of Schooling in Intergenerational Social  Mobility: Patterns, Perspectives and Experiences from Rural Pakistan. PhD thesis, University of Cambridge. Naveed, A., & Arnot, M. (2019) Exploring educational and social inequality through the polyphonic voices of the poor: A  habitus listening guide for the analysis of family-schooling relations. Comparative Education, 55(2), pp. 175–196. Naveed, A., & Sutoris, P. (2020) Poverty and education in South Asia. In Sarangapani, P., & Pappu, R. (Eds.) Handbook of Education Systems in South Asia: Global Education Systems. Global Education Systems. Singapore: Springer. Ray, D. (2003) Aspirations, Poverty and Economic Change. New York: Department of Economics, New York University. Available from: www.econ.nyu.edu/user/ debraj/Papers/povasp01.pdf [Accessed 5 August 2020]. Shami, M. (2010) The Road to Development: Market Access and Varieties of Clientelism in Rural Punjab, Pakistan. PhD thesis, London School of Economics, London. Sointu, E. (2005) The rise of an ideal: Tracing changing discourses of wellbeing. Sociological Review, 53(2), pp. 255–274. Tamim, T., & Tariq, H. (2015) The intersection of caste, social exclusion and educational opportunity in rural Punjab. International Journal of Educational Development, 43, pp. 51–62. White, S.C. (2017) Relational wellbeing: Re-centring the politics of happiness, policy and the self. Policy & Politics, 45(2), pp. 121–136. Wood, G. (2003) Staying secure, staying poor: The “Faustian bargain”. World Development, 31(3), pp. 455–471. World Economic Forum (2014) The Global Gender Gap Report 2014. Geneva: World Economic Forum.

8 Confronting social inequality through fertility change in Punjab, Pakistan The role of girls’ schooling Feyza Bhatti and Roger Jeffery Less well understood are the mechanisms by which the educational attainments of women effect improvements to the health and nutrition of their children, to their fertility behaviour, and to their personal efficacy and autonomy within the household. (Colclough 2012, p. 3)

Education plays a central role in the poverty reduction targets of the Sustainable Development Goals with its transformational capacity in improving opportunities and outcomes particularly for the marginalised groups.1 Within this context, girls’ schooling has been promoted as a key policy tool to reduce fertility rates and improve mother and child health in Southern contexts as well as to reduce poverty and improve social and economic equality between and within nations. Despite a well-established relationship between women’s schooling and a reduction in fertility rates, the pathways through which this relationship operates remain unclear. This chapter aims to clarify (i) how women’s schooling influences fertility attitudes and behaviour in Punjab, Pakistan and (ii) the pathways through which female schooling might act as a means to reduce social inequalities within and beyond the existing gender and family systems. We address the questions of how, and how far, the expansion of female schooling has transformed gender and family relationships that bear on reproductive decisions, with particular attention to relationships with processes of marginalisation and inequality. In order to address these questions, it is important to investigate women’s fertility attitudes and behaviour in detail, capturing, if possible, the more complex and dynamic nature of the fertility attitudes and behaviour within the existing gender and family systems. By focusing on one country, Pakistan, and its contemporary demographic transitions, and one province  – the Punjab  – we aim to contribute new insights in the fields of gender and development, social change and demography. Insufficient attention has been paid to how changing socio-economic environments and the increasing schooling of children inflect the dynamics of intergenerational contracts and how they influence social inequality. Using the findings of our empirical research, we tease out how this is happening by analysing the interactions of fertility and reproductive decision-making with

Confronting social inequality through fertility in Pakistan 145 family and gender systems. We explore in-depth the role of female schooling in transforming gender roles and family relationships in rural and urban areas during these demographic processes. We begin by considering the prevailing knowledge of what affects maternal schooling and the fertility behaviour of women.

Linking maternal schooling to fertility behaviour and attitudes Mothers’ literacy or years of schooling completed have been included in almost every study exploring the determinants of fertility attitudes and behaviour since the 1970s. Current understandings are that a woman’s schooling (for more than a few years) is needed for the number of her children to decline, especially where patriarchal kinship structures and son preference are significant contextual influences. Intra-family relationships, particularly between genders, have also received attention since the 1990s (Agarwal 1997; Mason 2001; McDonald 2000). Family systems, and their intertwined relationships with gender, are beliefs, norms, practices and sanctions that shape kin relationships and the expectations held of men and women of different ages, including divisions of labour (Mason 2001). Family and gender systems influence fertility through their impact on demand for and supply of children, on the cost of fertility regulation, child survival, the use of postnatal fertility control and on the ideal or acceptable number of children. Women’s status is a key component of patriarchal family systems that have been widely discussed in the literature on South Asia. Women’s status (Basu 1992; Hakim 2000), autonomy (Acharya et al. 2010; Dyson & Moore 1983; Jeffery & Jeffery 1996; Jeffery & Jeffery 1994; Sathar 1996) and reproductive autonomy (Saleem & Pasha 2008) have been key variables to help explain fertility change in the region. Most of these studies indicate that women’s age at marriage, schooling and earning opportunities or income or access to assets (such as land) can empower women through increasing their relative bargaining power or status and their relative contributions to fertility decision-making. But qualitative studies have questioned the utility of the notion of autonomy (Mumtaz  & Salway 2009) – it is a moral concept and expressing, possessing or exercising individual autonomy is not always seen as desirable for those South Asian women who are socialised into a culture that emphasises the embeddedness of individuals in larger groups, rather than the rights of individuals (Jeffery  & Jeffery 1994; Kabeer 2011). The presumed linear link between female schooling and women’s autonomy or empowerment has also been widely criticised (e.g. Khurshid & Saba 2018) since schooling might even reinforce existing gender inequalities, particularly in South Asia where the content of schooling supports prevailing gendered values and norms (Islam  & Asadullah 2018; Jeffery  & Basu 1996). Similarly, women’s autonomy does not always lead to lower fertility, particularly in patriarchal settings where son preference is high and female labour force participation is low (Samari 2017). Following Kabeer (2018), we suggest that issues of autonomy and empowerment can be better conceptualised as aspects of women’s agency. As Kabeer

146  Feyza Bhatti and Roger Jeffery (2018, p.  7) argues, agency is ‘the ability to define one’s goals and act upon them’ and argues that agency connotes more than just ‘decision-making’. Young women in South Asia use covert, rather than overt, strategies to bargain for agency (Kandiyoti 1988; Kabeer 1999; Gram et al. 2018). Women’s schooling might improve women’s agency by providing greater knowledge of, and exposure to, the outside world: it can enable women to play greater roles in decision-making at home, greater opportunities for interacting with the outside world, greater space for emotional choice and closer bonds with their husbands and children and greater social and economic self-reliance (Jejeebhoy 1995). However, the extent of the impact of schooling on female agency and thus on their fertility behaviour depends on the culture and levels of gender stratification. In highly patriarchal South Asian societies, the impact of schooling on women’s agency is usually minimal and visible only when a moderate level of schooling is reached (Jeffery & Basu 1996; Sathar 1996). Recent studies, however, indicate first, that the negative relationship between female schooling and fertility behaviour is not as strong as it was before (Brahmanandam & Arokiasamy 2017; Guilmoto 2016) and second, that greater female agency does not always lead to lower fertility (Prata et al. 2017; Samari 2017). In sum, women’s schooling, work and other aspects of their status do not necessarily impact on their reproductive agency. Rather, female agency is generally shaped by traditional and cultural factors, including the number of living sons (Hakim et al. 2003; Jejeebhoy & Sathar 2001; Samari 2017). Despite recent studies indicating that having daughters also increases women’s agency in South Asia by increasing their say in household decisions and mobility (Heath & Tan 2018), their parents continue to value sons more highly, for economic and cultural reasons. In Pakistan, where filial (and to a great extent fraternal) piety is the norm and social security systems are non-existent, sons bring a wife with dowry, whilst daughters need a dowry on marriage (Winkvist & Akhtar 2000). Although daughters can contribute economically to their households before marriage or improve the social standing of their families through marriage, their expected returns to their parents’ old age are low, which leads to a strong son preference – particularly after fertility declines (Das Gupta & Bhat 1997; Guilmoto 2009). Mobility aspirations, which can be intra- and intergenerational (Zuanna 2007) or nuptial (Kasarda & Billy 1985; Sharma & Wotipka 2019), can therefore be strongly affected by parental fertility decisions as well as parents’ investments in children’s schooling. In Pakistan, son preference remains a major obstacle to accepting the norm of having a small family and to increasing contraceptive use (Hussain et al. 2000). Although lower among Punjabis, son preference patterns persist across the country for women with higher levels of schooling and in middle- and higherincome households (Sathar et  al. 2015). While some recent studies suggest an increase in son preference (Saeed 2012), others identify signs of declining son preference (Ahmed & Bould 2004) particularly for mothers with high levels of schooling and in higher-income families (Sathar et al. 2015). Son preference also influences the treatment and status of women in the household according to their ability to produce sons (Mumtaz et al. 2013; Winkvist & Akhtar 2000).

Confronting social inequality through fertility in Pakistan 147 In contrast to the extensive literature on son preference, far less attention has been paid to how the changing socio-economic environment, particularly increasing schooling of girls and labour market participation of women, inflect these dynamics of intergenerational contracts in a rapidly transiting patriarchal society. A few studies focus on children’s schooling opportunities as a stimulus for fertility transition through increasing the cost of children (Sathar et al. 2003; Zaidi et al. 2012). Some studies from India indicate that increased access to schooling creates mental models for parents in which they perceive schooling, particularly private schooling, as a pathway to upward social mobility through better marriage prospects and gainful employment for their children (Sharma & Wotipka 2019; Srivastava 2006). Those families who want to have socially and economically able children who are respected in their communities and have better lives than their parents (Sharma  & Wotipka 2019) are therefore less concerned with whether they are boys or girls (Ahmed  & Bould 2004) which, in turn, can reduce son preference and its upward pressure on total fertility rates. This can reduce gender inequalities within and beyond households, and in the long run, based on the level that intergenerational mobility aspirations are realised, can also contribute to declining socio-economic inequalities between households. Here we focus on the patterns of education and fertility in the Pakistani context. We start by describing the research strategy and the evidence we collected on the impact of girls’ schooling on fertility attitudes in relation to the desired number and sex of children and on female behaviour including contraceptive use broadly in the Punjab. Drawing on our qualitative data, we then explore the role of schooling in transforming family relationships, revealing a deeper set of changes in young wives’ lives, as a result of their education. In the final sections, we focus on young wives’ aspirations as parents for their own children despite the increasing costs of schooling.

Researching education and fertility in Punjab, Pakistan Pakistan is characterised by low levels of women’s literacy, educational attainment and labour force participation. According to World Bank data, in 2017, adult female literacy was 48%, but only 27% of women aged 25 and above completed upper secondary schooling. These data also showed that the female labour force participation rate was just 22%, and only one quarter of women with university degrees participated in the labour market. National evidence also shows that women’s labour force participation is greater where there are fewer children in the household, particularly fewer young children aged 0–5 (Azid et  al. 2001; Faridi et al. 2009; Naqvi et al. 2002). In Pakistan, major changes in fertility and contraceptive use since the 1990s have been driven by women with no schooling (Bashir & Guzzo 2018). While significant overall declines in fertility rates and increases in contraceptive prevalence rates have been observed among all women, the rate of these improvements is faster among women without schooling as compared to women with schooling. Women without schooling are predominantly from poor households

148  Feyza Bhatti and Roger Jeffery in Pakistan: according to the 2012–13 Pakistan Demographic and Health Survey (PDHS) data, the percentage of women who did not receive schooling was 92.4% for women from poorest households and 17.6% for women from richest households (Pakistan National Institute of Population Studies 2012). Our research explored the PDHS for the years 1990–91, 2006–07 and 2012–13. The PDHS collects nationally representative cross-sectional data from ever-married women aged 15–49 on women’s and households’ background characteristics, marriage, reproductive history – including knowledge and use of family planning, fertility preferences, antenatal and delivery care, pregnancy complications and maternal mortality, children’s health and other health issues. Here we focus on the data specifically for Punjabi women aged 25–34. Our sample characteristics are provided in Table 8.1. We analysed these statistical data to show the differences and changes in fertility attitudes (desires) and behaviour (contraceptive prevalence rates) for young Punjabi women with different educational attainment levels. We also drew on the findings of two qualitative studies in Sargodha, Punjab. The first study was conducted in January–April 2008 as a part of the Research Consortium on Educational Outcomes and Poverty (RECOUP) Health and Fertility study,2 while the second was conducted by the first author in September 2010 to June 2011 as part of the first author’s PhD.3 The aim of these qualitative studies was to provide insights into changes in gender and family systems that might have resulted from increased schooling of women. In order to maximise the chances of understanding contemporary processes, our selection of the district was based on an extreme case: that is, where the fertility transition was most advanced and female secondary school enrolments were high. Sargodha, the sixth most populous district in Punjab, has had the fastest declines in total fertility rates among the other central Table 8.1 Sample characteristics for Punjabi women aged 25–34, 1990/91, 2006/07 and 2012/13  

1990/91

 

Freq.*

Residence Urban Rural Educational attainment No education Incomplete primary Completed primary Incomplete secondary Completed secondary Higher Total

2006/07 %

Freq.

2012/13 %

Freq.

463 1,142

28.8 71.2

734 1,461

33.4 66.6

591 876

40.3 59.7

1,176 32 129 206 36 26 1,605

73.3 2.0 8.0 12.8 2.3 1.6 100

1,135 132 242 214 238 234 2,195

51.7 6.0 11.0 9.8 10.8 10.7 100

610 100 194 177 194 192 1,467

41.6 6.8 13.2 12.1 13.2 13.1 100.0

Sources: Pakistan National Institute of Population Studies 1992, 2008, 2013 Note: *=Frequency

%

Confronting social inequality through fertility in Pakistan 149 Punjab districts and high female literacy and school enrolment rates. The data were collected in one rural and one urban community, with selection based on the long-term availability of girls’ high schools to enhance the chances of attaining the desired number of young women with high schooling levels. Using a purposive sampling strategy, we selected maximum variation cases based on young women’s educational attainment and having at least one child under the age of six, because in this age group women are actively considering (and possibly acting on) their completed family size and composition. This enabled the acquisition of information about how education influences women’s reproductive agency and fertility outcomes through comparison of extreme cases. In the two qualitative studies, semi-structured interviews with young married women collected information about their schooling and family, marriage processes and their experiences of childbearing, including decisions on fertility regulation. The interviews, conducted in Punjabi or Urdu in the conjugal homes of the participants, lasted for about an hour. They were recorded and transcribed in Urdu and later translated into English and coded in Atlas.ti using a coding framework based on emergent themes and the observations of the researchers. In this chapter we use 59 interviews with young married women (see Table 8.2 for details). Both samples were mainly identified through the _kh_andān [family] registers of Lady Health Workers, who also helped in accessing the households for initial permissions to interview young women. Consent was taken both from the family and from the young women in order to prevent any negative consequences of accessing young women directly. Young women and members of her conjugal family were informed about the aims of the study, confidentiality and anonymity of the research and the management and use of data, before a date and time was set for the interview. Most study participants were from low- to middle-income households. Although the ideal was to have individual interviews, this was not always possible due to extended family structures and lack of private spaces, so we adapted to local conditions and documented who was present, alongside any evidence of the possible effects of co-presence on interview responses. These interviews provided Table 8.2  Sample characteristics for qualitative studies, by education levels and location Educational attainment

Rural 2008 RECOUP

No/incomplete primary Middle (completed grade 8) High (completed grade 9 or above) Total

Urban 2011 PhD study

2008 RECOUP

2011 PhD study

Total

7 3

5 3

7 1

4 4

23 11

8

4

9

4

25

18

12

17

12

59

Source: The 2008 sample, Bhatti and Jeffery (2012); the 2011 sample, Bhatti (2014)

150  Feyza Bhatti and Roger Jeffery insights into complex decision-making and negotiation processes regarding fertility-related decisions as well as better understandings of how schooling influences these decisions. All names in this chapter are pseudonyms.4

The impact of Punjabi girls’ schooling on their fertility Our investigation into the role of schooling in changing fertility attitudes and behaviour was highly relevant to the women we interviewed about their and their children’s education. Schooling clearly had a significant impact on young Punjabi women’s attitudes to fertility and to their behaviour in relation to pregnancy/ childbearing. The three different PDH surveys had indicated how young Punjabi women’s fertility attitudes (family size ideals) and behaviour (contraceptive prevalence) vary by their schooling levels. Since the 1990s, Punjab has experienced a 30% decline in the total fertility rate from 5.4 in 1990–91 to 3.8 by 2013 (Pakistan National Institute of Population Studies 2013).5 Desired family size and the behaviour of young women have also changed significantly since 1990–91 and have affected their view of the ideal number of children. Although the ideal family size among young Punjabi women remains around four children (Table  8.3), there has been a significant decline in nonnumeric responses to survey questions (for example, it was ‘up to God’) about the desired family size, irrespective of schooling levels, from more than half in 1990–91 to 4% by 2012–13 (Pakistan National Institute of Population Studies 1992, 2013). This change has affected all women, irrespective of education. But in most other respects, attitudes continue to vary by schooling levels. The increases in the proportion of women who wanted to have two children, for example, can be considered as declining family size desires, a change more visible among women who completed grades 12 and above. Table 8.3 Percentage distributions of desired family size of Punjabi women aged 25–34 by schooling grade completed, 1990/91, 2006/07 and 2012/13 Grade

1990/91 0–5

Numeric responses 0–1 2 3 4 5 and above N

42.2

2006/07

8–11

81.6 85.8

1.2 0.7 8.5 18.8 14.7 17.4 54.7 52.3 20.9 10.7 518 148

Non-numeric 57.8 responses N 709

12+

8–11

93.9

93.7

12+ 96.8

0–5 98.2

8–11 98.8

12+ 99.5

0.0 1.2 1.0 0.9 1.2 1.6 1.6 13.2 13.5 24.7 32.1 13.1 26.1 30.6 34.0 19.4 24.0 26.1 20.0 25.3 26.3 45.3 48.4 45.6 35.3 48.6 42.7 39.2 7.5 17.6 4.7 5.5 17.2 4.3 2.2 53 1397 384 218 867 368 186

18.4 14.2 33

0–5

2012/13

8

6.1 91

6.3 26

3.2 7

1.8 16

Sources: Pakistan National Institute of Population Studies 1992, 2008, 2013

1.2

0.5

4

1

Confronting social inequality through fertility in Pakistan 151 These findings are corroborated by our interviews. The number of children who were considered to constitute a ćhot·ī (small) family varied between two and four depending on the educational levels of the women we interviewed in both studies. For each quote in the rest of this chapter, the speaker is identified by pseudonym; R=rural, U=urban; the age of the woman; and the grade completed. Almost all the young women with higher schooling levels, irrespective of location, desired a ćhot·ī family of two or three children or ‘definitely one should not have more than four’ as Hafza suggested (U27-Grade16). However, the most commonly stated desired number of children was still four among the uneducated young women, irrespective of their rural or urban location. The following quotes give a sense of the range of answers of this kind: Now I ask for four as I wish to have a daughter as well, otherwise three children are ideally a good number. (Maryam, U25-No Schooling) Yes, I did think about it [number of children one should have], that’s why everything was on time [in English]. First I had a son and he died. After two years I had another son, now he is 13 Mā shā’ Allāh. Then I had my daughter who is nine . . . then I had my younger daughter. She is six now. Then I had this son [showing her two-year-old son on her lap]. Two sons and two daughters and we are doing waqfa (pause) for future. (Ghazala, R30-Grade 1) Everyone wants to have two daughters and two sons . . . I have three daughters. Now I pray to God that he gives me a son. [Only] then it will be over. (Adeela, U32-Grade 8) I think two kids are good but we have three and it is okay. (Sadia, R26-Grade 12) I want to have two children. No matter whether [I have] two daughters or one daughter and one son. (Meena, U33-Grade 14) However, schooling levels also affected the responses from young women about the desired sex composition of children. Indeed, young women who wanted to limit their family size to a ‘small’ one also wished to have a mukammal (complete) or pūrā (complete) family that would have kāfī (sufficient) children. A family was ‘complete’ only after achieving enough boys, and at least one daughter, which was becoming more difficult with the decreasing total number of children that the younger women wanted to have. The desired sex composition of children mainly varied according to the educational levels of the young women, the sex of the children the couple already had and cultural and familial pressures, particularly for sons. Among the young women interviewed, the desire for a second son was uncommon but having at least one daughter to complete

152  Feyza Bhatti and Roger Jeffery their family was universal. The right sex composition for a complete family was neither static nor the same for all young women. All the young women readjusted their family sizes upwards if they could not attain their desired sex composition; none of them readjusted their desires to a smaller number. Most uneducated young women, like Ghazala (R30-NoSchooling) or Abida (R28-NoSchooling), who already had two sons and two daughters, considered that this combination would make a family complete. In contrast, more educated women, such as Gulsum (R33-Grade 14), saw a complete family as one daughter and one son, like the family she had. Only one woman, Meena (U33-Grade14) who had a daughter, had no specific desires about the sex of her children and said she would be happy with two children of any sex. Those young women who did not have at least one son and one daughter amongst their initial two or three children usually readjusted their desires and tried again to complete their family. Safia (U28-Grade10) wanted one son and one daughter but tried again for a daughter after having two sons. When interviewed, she had recently given birth to a girl and now considered her family complete: When I got married, I wanted to have a small family. I had a son first and I asked god to give me a daughter and two were enough. Then God gave me another son. [I thought] now only one daughter would be bahut (plentiful). Thanks to God, he gave me a daughter as well. I have two sons and a daughter, and I think the family is pūrī. (Safia U28-Grade 10) Having an ‘inadequate’ family because of not having a daughter, even if the total desired number of children had been reached (which was not more than four children), generally meant ‘trying again’ once or twice, as in Safia’s case. However, when the inadequacy was due to not having a son, couples could not define a limit and would try until they had a son. For others, particularly uneducated or less educated women who had one son and at least one daughter like Malika (R27-Grade 9), and Tania (U25-Grade 5), one son was not considered to be enough to achieve a ‘complete’ family. While sons continue to be preferred to daughters in general, there are some initial signs that, as Vlassoff (1990, p. 19) predicted, having one son is becoming ‘sufficient to fulfil the necessary cultural obligations’ particularly among highly educated women. It is also evident that a family without a son and a daughter was ‘inadequate’ for almost all young women in Punjab.

Contraceptive prevalence rate Linked to the decision of how many children to have, and the desire for sons, was the contraceptive prevalence rate (CPR) which was also affected by the level of schooling achieved by these wives in the 25–34 age group. The CPR has major implications for the reduction of social inequality since, if fewer children are conceived, the children in smaller families potentially have more access to the family’s resources and there are also more opportunities for young wives to

Confronting social inequality through fertility in Pakistan 153 Table 8.4 Percentage of reported current use of contraception by married, non-pregnant Punjabi women aged 25–24 by schooling, 1990/91, 2006/07 and 2012/13  

 

Grades 0–5

Grades 8–11

Grades 12+

Urban

1990/91 2006/07 2012/13 1990/91 2006/07 2012/13 1990/91 2006/07 2012/13

21.4 42.5 44.0 6.0 33.1 39.9 9.3 35.1 41.0

41.4 56.5 45.3 39.0 50.4 32.2 40.5 49.5 39.7

46.3 42.2 40.7 37.0 40.0 42.4 45.5 41.8 42.2

Rural Punjab

Source: Pakistan National Institute of Population Studies (1992, 2008, 2013)

enter the labour market. The CPR rate among married women (aged 15–49) in Pakistan remained below 10% until the end of the 1980s, reaching only 12% in 1990–91 (Pakistan National Institute of Population Studies 2013). However, surveys conducted after 1990 indicated that between 1990–91 and 2001 (ibid.), CPR more than doubled to 28% and then gradually increased to 33% in 2006–07 and 35.4% in 2012–13 (ibid.). The Punjab has the highest CPR of all provinces with 40.7% (ibid.). Table  8.4 shows the current use of contraception by grades completed for young women in Punjab, indicating that the differences between women by level of schooling are closing. Women with up to the completion of primary schooling had very similar contraceptive use (or even higher levels in urban areas) to those of women who had spent more time in school. Significantly, the change in contraceptive behaviour was most prominent among uneducated young women in rural areas. While only 9.3% of young uneducated Punjabi women were using contraception in 1990–91, this increased to 35.1% in 2006–07 and 41.0% in 2012–13 (Table 8.4) at a time when the CPR among the highly educated young women was around 40–45% and might have declined over the same period. Our interviews with young wives in 2011–12 uncovered similar educational patterns in contraceptive use. There were almost no differences in use of contraception among young women by educational level or by rural/urban location. The question to answer, then, is why do those with less schooling increasingly resemble their more educated peers when it comes to the use of contraception? Our starting point is to consider the impacts of girls’ schooling on intrahousehold gender and family relationships and inequalities.

Female schooling and subtle transformations in intrahousehold inequalities Schooling of girls has, it seems, contributed to subtle but cumulatively significant transformations in micro-level gendered inequalities and family dynamics,

154  Feyza Bhatti and Roger Jeffery irrespective of the schooling levels achieved by their mothers. Girls in Punjab learn strong values that define how to ‘be a good woman’, such as being a homemaker, sustaining the honour of the family and being obedient to men and elders, even in decisions related to their fertility. Within the existing gender and family systems, socialising girls in their roles as good women continues, but families have adjusted to changing socio-economic environments such as increasing levels of girls’ schooling, rising employment opportunities for women along with rising perceptions of economic distress as aspirations grow faster than incomes. First, the increased importance given to the schooling of girls has partially changed expectations about women’s roles in their natal and affinal homes by altering the division of labour among women within the household. The expectations of contributions by daughters and daughters-in-law to regular household chores have declined compared with previous generations. Mothers of school-going children are increasingly expected to ensure their children are clean and well-dressed, arrive safely at school and undertake after-school tuition and complete their homework. Second, despite continuity in the desire of parents to maintain the honour of the family through restricting the mobility of girls, parental flexibility in girls’ mobility for schooling (and employment, to some extent) has been increasing. Third, the requirement of completing schooling before getting married has led to increases in the age at marriage for young women, although it has also created dilemmas for families about protecting the honour of the family. Schooling of young women has also affected to whom they were married as compatibility also included schooling. For young women, rishta (marriage alliance) was a decision taken by their families, even when the marriage was between first cousins, which is more common among less educated brides (Pakistan National Institute of Population Studies 2013). Finding a rishta from a respectable family was important, but having similar educational attainment levels, a good job or some similarity in lifestyles (economic, religious or social) were prerequisites for a good rishta. The likelihood that parents would seek the consent of their daughters also corresponded to the girls’ schooling levels, the family’s place of residence and the type of marriage. In the interviews undertaken in 2011–12, none of the uneducated young women from the rural area, and only two of the four uneducated women from urban area, were asked for their consent. Almost all educated young women from the urban area were asked for their consent, but in the rural area only a highly educated, affluent young woman had a say in spouse selection (Bhatti & Jeffery 2012; Jeffery et al. 2012). Such boundaries of gender and family systems however are not absolute, static or non-negotiable. Irrespective of their schooling levels, most of the young women also showed forms of agency, varying from direct submission to resistance to decisions, for example, the choice of timing of marriage and the marriage partner, which are major decisions in a woman’s reproductive life (Bhatti & Jeffery 2012; Ghimire & Axinn 2013). To sum up, our findings suggest that female schooling has a role in reducing household gender inequalities through bringing subtle transformations in gender and family systems that allow women to have higher mobility and a say in decisions

Confronting social inequality through fertility in Pakistan 155 that are central to their reproductive lives. Given these changes, although female schooling does not fully tackle the intra-household gender inequalities, it weakens the effect of gendered inequalities for women with schooling. The distribution of labour within households, constraints on women’s mobility or consent in marriage can enlarge the gap between women with and without education and create new forms of inequalities. Our interviews also suggest that young women’s reasons for preferring smaller families and their increasing use of family planning methods reflect how schooling is altering the costs and values of having and raising children. This in turn is significant for the patterns of social inequality. We now turn to analyse these changes.

Young women’s aspirations for their children’s schooling: increasing the costs and value of children The financial and opportunity costs of schooling (rather than the learning content or the levels of schooling) were having a perhaps unintended effect on desired family size and their prospects of social mobility. Most young women explained their reasons for wanting a ćhot·ī (small) family by referring to mehan·gāi (increasing costs of living). Irrespective of their own schooling levels, young couples felt compelled to plan their families according to the expenses incurred by children and the limits of their budgets: There is no fā’ida (benefit) of having many children, [particularly at a time] when a person cannot even fulfil their [children’s] needs of simple rot·ī (flat bread), leave aside their educational needs which is a matter for later . . . the more children one has, the more expenses he has to bear. (Sameena, U33-No Schooling) Now there is [the concept of] aććhī tarbiyat [good upbringing]. . . . We are the ones to educate them [children] and deal with their upbringing. Now it is not like give birth to a child and leave him. . . . Now even _garīb se _garīb [poorest of the poor] households educate their children well. (Afaf U24-Grade 10) We, the new generation, want smaller families and we’re getting by on that . . . So they [the children] can be raised and educated properly and our expenses can keep going up as well. (Ameena R25-Grade 10) In these young women’s accounts, mehan·gāi was associated with an increase in living expenses, childbirth costs and schooling which led them to perceive children as expensive. The increasing schooling costs, associated as they are with the expansion of private schools in rural and urban areas, were the most common reason reported by young women for mehan·gāi. These costs have been rising because of higher fees and charges for books and exams.

156  Feyza Bhatti and Roger Jeffery First, parents were being encouraged to educate all children irrespective of their gender: schooling has become a norm in a context where the Pakistani government has promoted mass schooling. In our study, parents reported feeling social pressure to send their children to school as ‘responsible parents’, particularly when education to secondary level was now easily available and accessible. Private tuition, which was perceived as necessary to gain good grades, despite adding to the costs, had become common: Aslam and Mansoor (2011) showed that of children aged 3–16, 11% receive private tuition in rural areas, but in urban areas like Lahore and Karachi it was over half. Tuition was more common among the children of uneducated women as they could not teach their children themselves. Private schooling was also preferred by the parents because it seemed to offer better teacher attention and greater language and computer skills than in sarkārī (government) schools. However, children were staying longer in education and often moved schools incurring readmission costs. In the case of private schools, pupils started school as young as three or four years old and stayed until at least high school. Parents now seem willing to pay high travel costs, if necessary, to send their children to good schools: I arranged tuition for him. From my _kh_arća [expenses for the household], the money that their father gives me, I pay for his tuition fees. We already have enough expenses and now I also have to pay this. . . . As much as we could afford, they should study. Otherwise they will say they [parents] did not study themselves and they did not make us study as well. . . . If their upbringing and schooling is good, then they will be able to become kućh [someone]. (Sameena, U33-No Schooling) One has lots of desires to educate their children, but for that you also need financial resources. . . . The times of āzādī [freedom] are over, one has to think a lot before taking a step  .  .  . I mean there [Beacon house private school] the fees for children are very high, and it [their schooling] is going to last long as well, I mean until their schooling is over. Then there will the problem of their tuition as well. (Gulsum, R33-Grade 14) The costs of schooling for daughters and sons were also equal: there were no gender differences in aspirations for children’s schooling or in the private schooling of children in relatively better off households. Girls in poor households, however, were the least likely to attend private schools and only in these households was there a difference in the types of schools that girls and boys attended.

Aspirations for intergenerational mobility: respectable jobs and good marriages The importance given to children’s schooling was explained by young women’s aspirations for their children’s futures. Almost all the women associated schooling

Confronting social inequality through fertility in Pakistan 157 with aććhī zindagī (a good life) and considered it as the main means to intergenerational mobility – to a better life than they had: I know only education can benefit them, otherwise they are just going to suffer as we are suffering today. (Aziza, U25-No Schooling) No matter how hard I have to work I would work for the sake of educating my children. And if my daughter is educated then she would not have to go through the same circumstances as mine and she would be able to find her own ways of life. And for my son also I wish him to get good education and he would be able to find a respectable job . . . I would educate my daughter and make her independent enough so that she could start her own business and then I will think of her marriage. (Kaneez, R22-No Schooling) My children will be lāik aur nāik (deserving and respected) and everyone will call them jī (sir) . . . they will have cars, motorcycles and will have their homes. (Aafia, R35-Grade 2) The number of years of schooling their children completed clearly had an effect on parental aspirations for their children’s future social status, raising their aspirations but not necessarily the choice of employment. Uneducated women, for example, in our study were less likely to mention the type of profession they desired for their children. Aafia, for example, wanted her sons to be in respectable jobs in which they would be called ‘jī’ in addition to earning money to buy a car and a house. Highly educated urban women also mostly wanted a well-respected profession for their sons and daughters. For example, Hafza (U27-Grade 12) and Meena (U33-Grade 14) wanted their daughters to become doctors, and Kameela (U26-Grade 16) wanted the same for her son. Safia (U28-Grade 10) wanted her daughter to have nām (title) like a doctor or an engineer. Schooling was considered to be a necessity for both sexes. However, social mobility for girls was still through marriage rather than through occupation. Schooling was needed if they wanted to marry educated men and live in good households. This also reflected the parents’ desires for upward social mobility for all their children, which for girls could only be achieved through a good marriage: If only they are educated, then they will manage it. They will reach their manzil (destination). Girls will grow and get married. If they are educated only then their rishta will happen in a good place. (Aafia, R35-Grade 2) If they study well, then they will have a good future. They will marry to a good place, they will have a good home. This is what one thinks for their daughters, that they have a good future. . . . Everything happens if they study well. (Afaf, U24-Grade 10)

158  Feyza Bhatti and Roger Jeffery With increasing female schooling and more employment opportunities for women, the relative economic value of daughters in their natal homes has also increased. Daughters were more likely to marry later, and if educated, they could also spend the time between schooling and marriage in paid employment. Compared to the previous generation, the mindset of their elders appears to have changed: it is now easier for women to get ijāzat [permission] from the family to work. We want this [more sons] so that they can earn [for the family] but nowadays even daughters get education and can do jobs. . . . Times are changing and you have to follow them. Our time has passed. Now he [husband] is the only one earning and we are six people to be fed. . . . I want to educate my daughters and God should give them jobs. At least my daughters will have some earnings and they can arrange themselves. (Khalida, U32-Grade 5) Now it [having a job] is not a problem. They [women] work. My brother’s daughter has started working, in our house my sister-in-law works. This was not the case before. . . . This nasl (generation) has very high education [levels]. . . . They used to think it [women’s employment] wrong before that she is going to be outside . . . but now they give ijāzat [permission], the soć [thinking] of the elders has changed a lot. (Adeela, U32-Grade 8) Among these young women, however, working was a preference rather than compulsory. Only young women with high schooling reported their willingness to work. Iqra was working from her rural home as a beautician while waiting for job openings in local public schools; Gulsum wanted to restart schooling when her children left school, as her sisters did and Iffat became a Lady Health Worker after her marriage. In the urban area, Hafza was waiting for her children to grow up to continue her schooling and start working again, and Kameela was planning to apply for jobs when her 40-day-old son was at least three months old. None of these women wanted their schooling to be wasted. They had permission to work and wanted to help their husbands financially, but in jobs that would not affect their responsibilities as mothers. While these changes were generally the result of economic forces, increasing female schooling and increasing job opportunities for women, the expansion of the private education sector and government jobs all contributed towards closing the gap in the values of sons and daughters by increasing the relative financial value of daughters.

Conclusion Our data show the significant points at which schooling affects the personal and intimate lives of young women – their aspirations for a family, their choices in

Confronting social inequality through fertility in Pakistan 159 terms of the number of children, the sex of the child, decisions about contraception. Our findings suggest that women from both urban and rural backgrounds, and with different levels of schooling themselves, were actively involved in decisions about their fertility and wanted to limit the number of children they had irrespective of their educational backgrounds. While women’s schooling leads to subtle transformations in gender and family systems, what pressurised all women to limit their fertility, irrespective of their own schooling, was their aspirations for their children’s schooling, rather than the schooling they had themselves experienced. We show that, whereas in the past there was a strong inverse relationship between a woman’s schooling and her desired fertility, this relationship is breaking down in contemporary Punjab, partly with respect to desired family size but almost entirely with respect to contraceptive use, even when controlling for residence and number of living children. We suggest that the key intervening factor is families’ perceptions of increasing costs of living, particularly those related to schooling and their aspirations for upward social mobility. Our research suggests changes in fertility attitudes and behaviour of young Punjabi women and indicates the possible pathways by which women’s schooling is contributing to these changes. Only slight differences remain between young women with high and low levels of schooling in terms of the desired number of children and their sex composition. While this might suggest a decreasing role for women’s schooling on fertility, the qualitative interviews suggest that this would be wrong. Within young Punjabi households, despite conflicting pressures of various kinds (including intergenerational ones), there is common ground that education is absolutely necessary, that the best possible education should be provided for children of both sexes and couples agree that the costs are so large that small families are necessary – and all this varies little if at all with respect to education or marginality. The pathways through which schooling influenced these decisions were rather more complex and cannot be considered without also analysing the changes occurring in the wider social, economic and familial contexts. With the expansion of facilities in the education and health sectors, in both rural and urban areas, even young women from the more marginalised families had high aspirations for a better life for themselves and their children. This increased the costs of living for the couples who wanted to invest in the schooling of sons and daughters alike. Given the economic pressures that are felt as increasing costs of childrearing, women think that small families are required. The schooling of children has become more important than mothers’ schooling in influencing couples’ fertility decisions as the aspirations of parents for children’s upward social mobility emerge. The relationship between fertility and social mobility is twofold. Having many children limits the mobility of their parents, particularly mothers, by bringing extra constraints on them and diverting their time and efforts from work to childcare and childrearing. Furthermore, intergenerational mobility of children can also be hindered when families are large and the financial, time and effort resources of the parents are diluted among the children (Dribe et  al. 2012).

160  Feyza Bhatti and Roger Jeffery Our findings suggest that the latter explains the contemporary fertility declines in Pakistan, where families increasingly tend to limit their family size in order to invest in schooling of their children, both boys and girls. Nonetheless, a cautionary note is in order. Once education is seen as something that must be consumed across all parental educational backgrounds, it has different effects. On the one hand it tends to be levelling: with two to three or at most four children, parents can think of investing in quality in a wider range of households. On the other hand, if parents buy into this goal of educating all their children to the best of their ability, they are chasing a positional good which can never be available for all, in terms of upward social mobility. What happens when the spell is broken? [Believing that] education is invariably accompanied by social mobility is to ignore the important constraints that current economic and social conditions and practices continue to have upon the lives of those for whom schooling is meant to be a liberating force. (Froerer 2012, p. 355)

Notes 1 Roger Jeffery would like to acknowledge the contributions of Patricia Jeffery in helping to make this chapter possible. From 1982 to 2010 we worked together on a number of projects that included research and publications on education and fertility in South Asia. This chapter has benefitted from that work in a general way, but does not draw directly from it, except where use is made of research material from already published joint works, when citations have been made to the original source. 2 The Research Consortium on Educational Outcomes and Poverty (RECOUP) led by Christopher Colclough was designed to investigate the social and economic outcomes of education for the poor households in Africa and South Asia. The Health and Fertility study, which was one of the six qualitative studies, aimed to explore the relationships between schooling and health and fertility-related behaviour among women of poor households. Its specific focus was on how female schooling was linked with reproductive citizenship and women’s agency in poor communities. 3 Bhatti (2014) provides more information on this study. 4 For more details of the research design, see Jeffery et al. (2012). 5 During the same period, women’s access to schooling in Punjab also rose. Based on Pakistan: Demographic and Health Survey (Pakistan National Institute of Population Studies 1992, 2013), the proportion of women with no schooling decreased from 63.7% in 1990/91 to 51.1% in 2013, and the proportion of women who completed a level higher than secondary schooling increased from 7.7% in 1990/91 to 20.4% in 2013.

References Acharya, D.R.,  & Bell, J., Simkhada, P., van Teijlingen, E.,  & Regmi, P. (2010) Women’s autonomy in household decision-making: A demographic study in Nepal. Reproductive Health, 7(15). Agarwal, B. (1997) ‘Bargaining’ and gender relations: Within and beyond the household. Feminist Economics, 3(1), pp. 1–51.

Confronting social inequality through fertility in Pakistan 161 Ahmed, S.S., & Bould, S. (2004) One able daughter is worth 10 illiterate sons: Reframing the patriarchal family. Journal of Marriage and Family, 66(5), pp. 1332–1341. Aslam, M., & Mansoor, S. (2011) The private tuition industry in Pakistan: An alarming trend. In Annual Status of Education Report. Lahore: South Asia Forum for Education Development. Azid, T., Aslam, M., & Chaudhary, M.O. (2001) Poverty, female labour force participation, and cottage industry: A case study of cloth embroidery in rural Multan. The Pakistan Development Review, 40(4), pp. 1105–1118. Bashir, S., & Guzzo, K.B. (2018) Couple-level Agreement on Future Fertility Intentions and Contraceptive Use. [Online] Population Association of America 2018 Annual Meeting, Denver, USA. Available from: https://paa.confex.com/paa/2018/ mediafile/ExtendedAbstract/Paper19977/Saima%20Bashir_PAA%202018.pdf [Accessed 9 June 2020]. Basu, A.M. (1992) Culture, the Status of Women, and Demographic Behaviour: Illustrated with the Case of India. Oxford: Clarendon Press. Bhatti, F. (2014) Punjabi Families in Transition: An Intergenerational Study of Fertility and Family Change. PhD thesis, University of Edinburgh, Edinburgh. Bhatti, F., & Jeffery, R. (2012) Girls’ schooling and transition to marriage and motherhood: Exploring the pathways to young women’s reproductive agency in Pakistan. Comparative Education, 48(2), pp. 149–166. Brahmanandam, N., & Arokiasamy, P. (2017) Fertility transition in Andhra Pradesh: Role of diffusion in use of contraception among illiterate women. Social Science Spectrum, 2(4), pp. 239–247. Colclough, C. (2012) Investigating the outcomes of education: Questions, paradigms and methods. In Colclough, C. (Ed.) Education Outcomes and Poverty. Oxford: Routledge. Das Gupta, M., & Mari Bhat, P.N. (1997) Fertility decline and increased manifestation of sex bias in India. Population Studies, 51(3), pp. 307–315. Dribe, M, Van Bavel, J.,  & Campbell, C. (2012) Social mobility and demographic behaviour: Long term perspectives. Demographic Research, 26, pp. 173–190. Dyson, T., & Moore, M. (1983) On kinship structure, female autonomy, and demographic behavior in India. Population and Development Review, 9(1), pp. 35–60. Faridi, M.Z., Chaudhry, I.S., & Anwar, M. (2009) The socio-economic and demographic determinants of women work participation in Pakistan: Evidence from Bahawalpur District. South Asian Studies, 24(2), pp. 353–369. Froerer, P. (2012) Learning, livelihoods, and social mobility: Valuing girls’ education in central India. Anthropology & Education Quarterly, 43(4), pp. 344–357. Ghimire, D.J., & Axinn, W.G. (2013) Marital processes, arranged marriage, and contraception to limit fertility. Demography, 50(5), pp. 1663–1686. Gram, L., Skordis-Worrall, J., Mannell, J., Manandhar, D.S., Saville, N., & Morrison, J. (2018) Revisiting the patriarchal bargain: The intergenerational power dynamics of household money management in rural Nepal. World Development, 112, pp. 193–204. Guilmoto, C.Z. (2009) The sex ratio transition in Asia. Population and Development Review, 35(3), pp. 519–549. Guilmoto, C.Z. (2016) The past and future of fertility change in India. In Guilmoto, C.Z., & Jones, G.W. (Eds.) Contemporary Demographic Transformations in China, India and Indonesia. Switzerland: Springer, pp. 113–132. Hakim, A. (2000) Are status of women and contraceptive prevalence correlated in Pakistan? Pakistan Development Review, 39(4), pp. 1057–1073.

162  Feyza Bhatti and Roger Jeffery Hakim, A., Salway, S., & Mumtaz, Z. (2003) Women’s autonomy and uptake of contraception in Pakistan. Asia-Pacific Population Journal, 18(1), pp. 63–82. Heath, R., & Tan, X. (2018) Worth fighting for: Daughters improve their mothers’ autonomy in South Asia. Journal of Development Economics, 135, pp. 255–271. Hussain, R., Fikree, F.F.,  & Berendes, H.W. (2000) The role of son preference in reproductive behaviour in Pakistan. Bulletin of the World Health Organization, 78(3), pp. 379–388. Islam, K.M.M.,  & Asadullah, M.N. (2018) Gender stereotypes and education: A comparative content analysis of Malaysian, Indonesian, Pakistani and Bangladeshi school textbooks. PloS One, 13(1), e0190807. Jeffery, P., & Jeffery, R. (1994) Killing my heart’s desire: Education and female autonomy in North India. In Kumar, N. (Ed.) Women as Subjects: South Asian Histories. Charlottesville: University Press of Virginia. Jeffery, P., & Jeffery, R. (1996) What’s the benefit of being educated? Girls’ schooling, women’s autonomy and fertility outcomes in Bijnor. In Jeffery, R., & Basu, A.M. (Eds.) Girls’ Schooling, Women’s Autonomy and Fertility Change in South Asia. London: Sage. Jeffery, R., & Basu, A.M. (1996) Schooling as contraception? In Jeffery, R., & Basu, A.M. (Eds.) Girls’ Schooling, Women’s Autonomy and Fertility Change in South Asia. London: Sage. Jeffery, R., Bhatti, F., Noronha, C., & Jeffery, P. (2012) Girls’ schooling and women’s autonomy in South Asia: Revisiting old debates with new data from India and Pakistan. In Colclough, C. (Ed.) Education Outcomes and Poverty. London and New York: Routledge. Jejeebhoy, S.J. (1995) Women’s Education, Autonomy, and Reproductive Behaviour: Experience from Developing Countries. New York: Oxford University Press. Jejeebhoy, S.J.,  & Sathar, Z.A. (2001) Women’s autonomy in India and Pakistan: The influence of religion and region. Population and Development Review, 27(4), pp. 687–712. Kabeer, N. (1999) Resources, agency, achievements: Reflections on the measurement of women’s empowerment. Development and Change, 30(3), pp. 435–464. Kabeer, N. (2011) Between affiliation and autonomy: Navigating pathways of women’s empowerment and gender justice in rural Bangladesh. Development and Change, 42(2), pp. 499–528. Kabeer, N. (2018) Unit-5 Resources, Agency, and Achievements: Reflections on the Measurement of Women’s Empowerment-Part 1. [Online] New Delhi: Indira Gandhi National Open University. Available from: www.egyankosh.ac.in/bitstream/123456789/42463/1/Unit-5.pdf [Accessed 1 April 2020]. Kandiyoti, D. (1988) Bargaining with patriarchy. Gender and Society, 2(3), pp. 274–290. Kasarda, J.D., & Billy, J.O.G. (1985) Social Mobility and fertility. Annual Review of Sociology, 11(1), pp. 305–328. Khurshid, A.,  & Saba, A. (2018) Contested womanhood: Women’s education and (re)production of gendered norms in rural Pakistani Muslim communities. Discourse: Studies in the Cultural Politics of Education, 39(4). pp. 550–563. Mason, K.O. (2001) Gender and family systems in the fertility transition. Population and Development Review, 27(Supplement), pp. 160–176. McDonald, P. (2000) Gender equity in theories of fertility transition. Population and Development Review, 26(3), pp. 427–439.

Confronting social inequality through fertility in Pakistan 163 Mumtaz, Z.,  & Salway, S. (2009) Understanding gendered influences on women’s reproductive health in Pakistan: Moving beyond the autonomy paradigm.  Social Science and Medicine, 68(7), pp. 1349–1356. Mumtaz, Z., Shahid, U., & Levay, A. (2013) Understanding the impact of gendered roles on the experiences of infertility amongst men and women in Punjab. Reproductive Health, 10(3), pp. 1–10. Naqvi, Z.F., Shahnaz, L., & Arif, G.M. (2002) How do women decide to work in Pakistan? [with comments]. The Pakistan Development Review, 41(4), pp. 495–513. Pakistan National Institute of Population Studies (1992) Pakistan: Demographic and Health Survey, 1990/91. [Online] Islamabad: National Institute of Population Studies. Available from: https://dhsprogram.com/pubs/pdf/FR29/FR29.pdf [Accessed 9 June 2020]. Pakistan National Institute of Population Studies (2008) Pakistan Demographic and Health Survey 2006–07. [Online] Islamabad: National Institute of Population Studies. Available from: https://dhsprogram.com/pubs/pdf/FR200/FR200.pdf [Accessed 9 June 2020]. Pakistan National Institute of Population Studies (2013) Pakistan Demographic and Health Survey 2012–13. [Online] Islamabad: National Institute of Population Studies. Available from: https://dhsprogram.com/pubs/pdf/FR290/FR290.pdf [Accessed 9 June 2020]. Prata, N., Fraser, A., Huchko, M.J., Gipson, J.D., Withers, M., Lewis, S., Ciaraldi, E.J.,  & Upadhyay, U.D. (2017) Women’s empowerment and family planning: A review of the literature. Journal of Biosocial Science, 49(6), pp. 713–743. Saeed, S. (2012) Modelling Son Preference in Pakistan. PhD thesis, University of Texas, Arlington. Saleem, A., & Pasha, G.R. (2008) Women’s reproductive autonomy and barriers to contraceptive use in Pakistan. The European Journal of Contraception & Reproductive Health Care, 13(1), pp. 83–89. Samari, G. (2017) Women’s agency and fertility: Recent evidence from Egypt. Population Research and Policy Review, 36(4), pp. 561–582. Sathar, Z.A. (1996) Women’s schooling and autonomy as factors in fertility change in Pakistan: Some empirical evidence. In Jeffery, R.,  & Basu, A.M. (Eds.) Girls’ Schooling, Women’s Autonomy and Fertility Change in South Asia. London: Sage. Sathar, Z.A., Lloyd, C.B., Mete, C., & ul Haque, M. (2003) Schooling opportunities for girls as a stimulus for fertility change in rural Pakistan. Economic Development and Cultural Change, 51(3), pp. 677–698. Sathar, Z.A., Rashida, G., Hussain, S.,  & Hassan, A.  (2015) Evidence of Son Preference and Resulting Demographic and Health Outcomes in Pakistan. [Online] Population Council, Islamabad. Available from: www.popcouncil.org/uploads/ pdfs/2015PGY_SexSelectionPakistan.pdf [Accessed 9 June 2020]. Sharma, G., & Wotipka, C.M. (2019) Mothers’ gendered aspirations for their children: A case study of Forbesganj, Bihar. Compare: A Journal of Comparative and International Education, 49(3), pp. 358–374. Srivastava, P. (2006) Private schooling and mental models about girls’ schooling in India. Compare, 36(4), pp. 497–514. Vlassoff, C. (1990) The value of sons in an Indian village: How widows see it. Population Studies, 44(1), pp. 5–20.

164  Feyza Bhatti and Roger Jeffery Winkvist, A, & Akhtar, H.Z. (2000) God should give daughters to rich families only: Attitudes towards childbearing among low-income women in Punjab, Pakistan. Social Science and Medicine, 51(1), pp. 73–81. Zaidi, B., Sathar, Z., ul Haque, M.,  & Zafer, F. (2012) The Power of Girls’ Schooling for Young Women’s Empowerment and Reproductive Health. [Online] Population Council, Islamabad. Available from: https://knowledgecommons.popcouncil. org/cgi/viewcontent.cgi?article=1009&context=departments_sbsr-pgy [Accessed 9 June 2020]. Zuanna, G.D. (2007) Social mobility and fertility. Demographic Research, 17(15), pp. 441–464.

9 Teenage pregnancy and social inequality An impediment to achieving schooling for all in Uganda Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo Achieving schooling for all has been a major goal for many African countries. With multiple reforms, many countries in Africa have progressed in expanding access to education across the different levels, especially basic education. However, in many sub-Saharan African (SSA) countries, gender disparities remain a concern, especially at secondary and higher education levels. In SSA, over the period 2010–15, only 36% of countries had achieved gender parity at primary level, 26% and 9% at lower and upper secondary level respectively, and none had reached such parity in tertiary education (UNESCO 2018). Evidence indicates that expanding enrolment in secondary level education does not guarantee completion since only 86 and 78 of female students for every 100 male students completed lower and upper secondary respectively (UNESCO 2018). In Uganda, the Gender Parity Index in completion was 0.87 and 0.64 (in 2015) for lower and upper secondary education respectively (UNESCO 2017). These educational patterns have implications for the social and economic status of girls and young women. Indeed, evidence shows that gender inequalities in education restrict young women’s opportunities for employment in decent work, thus limiting their earnings, agency and access to productive resources as well as curtailing their participation in society and their future well-being (Wodon et al. 2017). While girls’ low educational accomplishment is partly attributed to teenage pregnancy, evidence shows that teenage pregnancy, in turn, is a marker of low education. Both educational inequality and pregnancy are influenced by the interaction of social inequalities within girls’ social lives (United Nations Fund for Population Activities 2013; Wodon et al. 2017). Consequently, education has been identified as a key strategy for addressing persistent teenage pregnancies in Africa. In this chapter, we draw on Bronfenbrenner’s (1979) ecological theory to explain the complex and multiple factors at play in the different ecological echelons of teenagers’ lives. This theory enables us to recognise the interaction of various factors (social and economic inequalities) across the different ecological levels which influence adolescent girls’ developmental outcomes, such as teenage childbirth and educational attainment. According to Bronfenbrenner (1979, 1994) an individual (in this case an adolescent girl) is positioned within a set of

166  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo interwoven systems that includes five layers – namely the microsystem, mesosystem, exosystem, macrosystem and chronosystems. The adolescent girl’s microsystem is her family/home; the mesosystem consists of the interaction between family, school and peer groups; her exosystem is the community. Whilst the macrosystem refers to the national and societal level,1 the chronosystem covers the changes that have taken place such as socio-economic transformations, which influence adolescent sexual behaviours and schooling. Bronfenbrenner’s ecological model, therefore, is useful in explaining the experiences of adolescent girls in education and provides an understanding that teenage pregnancy cannot be treated in isolation but rather is influenced by multiple factors operating in a girl’s social ecology. Studies on teenage pregnancy and girls’ education conducted in SSA among adolescents and their relations (caregivers) underscore the interaction of social and economic disadvantage across these different ecological levels with the incidence of teenage childbearing and education inequality (e.g. Jewkes et al. 2009; Chohan & Langa 2011; Ahorlu et al. 2015; Krugu et al. 2017). They reveal some of the sites in the region in which such girls face inherent patriarchal tendencies that privilege boys over girls and value marriage and motherhood as the only option for girls (Walker 2012). Such patriarchal tendencies create inequalities across the different ecological systems that increase girls’ risks of early childbirth and limit their educational opportunities. Our contribution here is to highlight the importance of bringing teenage pregnancy into the debates about education’s role in reducing social and economic inequality. In this chapter, we explore from different angles the educational challenges associated with teenage pregnancy which increase and aggravate the social and economic disadvantage girls’ face. The first section of the chapter reflects on the plethora of Ugandan legislative reforms, policies and prevention strategies (including the provision of sexual and reproductive health education) that aim to protect girls from early childbearing. The subsequent sections focus on trends in teenage pregnancy in SSA, particularly in Uganda. They consider evidence of the factors and patriarchal norms which shape high rates of incidence along with the implications for girls’ education. Finally, we discuss implications for policy-makers aiming to achieve improved educational attainment for girls.

Our research study Teenage pregnancy affects schooling experiences in many different ways, but schooling itself also has a role in both facilitating or preventing such early childbirth. The qualitative research2 we conducted in 2013 and 2014 in 12 districts3 in Uganda explored this interface between early childbirth with education and the role of education in reducing these patterns. We conducted individual interviews with 29 married teen mothers and 20 married young men as well as 16 focus group discussions (FGDs)4 with 130 teen mothers (65 married and 65 unmarried). Individual interviews were also held with in-school and out-of-school unmarried adolescent girls (58) and boys (59) aged between 14 and 19 years and 64 FGDs5 (204 girls and 182 boys). We also constructed 37 case studies of young

Teenage pregnancy and social inequality in Uganda 167 women who, as outliers, were examples of ‘positive deviance’ – in other words, they had completed schooling against the odds. We complemented these data by interviewing their parents (61 mothers, 43 fathers) and holding discussions with 72 mothers, 63 fathers, 25 teachers (12 females, 13 males) and 95 community members (48 females, 47 males). All participants in the study were purposively selected  – they were diverse in terms of religion, culture, language, education level and socio-economic status. The research was conducted in conformity with national and international ethical standards built upon principles of respect for the rights and needs of children and doing no harm. Key ethical considerations included voluntary participation, informed consent, confidentiality and anonymity. In our interviews and discussions, we explored a wide range of issues with different participants. For young mothers and unmarried adolescents, the issues covered were family background; their ideals of masculinity and femininity; education and vocational training; economic well-being and livelihood strategies; social connectedness and access to sexual and reproductive health information. Other themes investigated included the reproductive health services; gender-based violence and access to justice in the community; marriage, the laws, policies and programmes related to marriage; childbearing and education and the intersection of early marriage/pregnancy with girls’ education, among others. The issues we covered with families and community members were, for example, changes in marriage practices; teenage pregnancy and girls’ education; major political, economic, environmental, technological changes; forms and types of marriage, age at marriage; incentives, sanctions, laws, policies, programmes and services with respect to marriage and education; the intersection between marriage, pregnancy and education and perspectives on ideals of masculinity and femininity. In the following section, we start by outlining the legislative and policy commitments the Ugandan government has developed to reduce teenage pregnancy as well as to address social and education inequality.

Legislative and policy commitments to address teenage pregnancy and education inequality The Ugandan macrosystem has a robust policy and legal environment for promoting schooling for all as a key strategy for empowering girls and women. The 1992 White Paper on education that initiated Uganda’s education policy reform process stressed education for all as a human right and called for provision of free universal education (Government of Uganda 1992). This commitment has guided subsequent legislation and policies to ensure equal opportunities in education which are anchored in, for example: the constitutional provisions promoting gender equality (Articles 21 and 33); affirmative action (Article 33 (5); the right to education (Articles 30 34); the legal age of marriage or engagement in sexual acts at 18 years (Article 31(1) articulated in the 1995 Constitution operationalised in the Education Act 13 (Articles 4, p. 2 and 10 (3c)) (Republic of Uganda 1995, 2008). Also, Vision 2040 and the National Development Plan I  and II emphasise the importance of keeping girls in school up to completion and the

168  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo need to address institutional, gender and cultural barriers to girls’ participation in education. Teenage pregnancy and early marriage are some of the key concerns that are noted as compromising girls’ educational opportunities. Table 9.1 Table 9.1  The National Legal and Policy Framework Legislation/Policy/Plan

Provisions

Children’s Statute (1996)

• Right to education, protection against discrimination and sexual exploitation.

Teachers’ Code of Conduct (1996)

• Protection against sexual exploitation (abuse and harassment).

The Penal Code (amendment) Act (CAP 120 2007) – Section 129

• Protection of girls against early sex by prohibiting sexual acts for persons below 18 years (defilement) and penalise coerced sexual intercourse (life imprisonment; or a death sentence).

The Marriage Act (1904)

• Sets the age of consent at 21 years and written consent for marriage of minors (Section 17).

The Customary Marriage Act (1973)

• Sets the age of consent for females at 16 years (Article 11a) and males at least 18 years (11b); requires the consent of the father/mother/ guardian(s) in the marriage of minors (Article 11d, Schedule 11).

The National Population Policy (2008)

• Identifies teenage motherhood as a major contributor to persistent high fertility and maternal mortality ratio.

Gender in Education Sector Policy (Ministry of Education, Science, Technology and Sports 2016)

• Enhancing access and participation in education for both boys and girls. • Commitment to facilitating re-entry of girls who drop out as a result of teenage pregnancy. • Promotes sexuality education.

National Adolescent Health Policy Guidelines and Service Standards 2012 (Ministry of Health, Uganda 2012)

• Address adolescent health concerns, promote provision of better health and information services (adolescent friendly services) and the right to health and education for young people. • Promote positive health behaviours amongst adolescents including relations based on equity and mutual respect between genders. • Promote protection against harmful traditional practices and all forms of abuse including sexual abuse, exploitation, trafficking and violence.

Teenage pregnancy and social inequality in Uganda

169

Legislation/Policy/Plan

Provisions

National Strategy for Girls’ Education in Uganda (2015–19) (Ministry of Education and Sports 2013)

• Commitment to eliminate sexual and gender-based violence, early sex, teenage pregnancy in schools, retention of pregnant girls/mothers. • Supports local governments to work with communities to enact byelaws to increase enrolment, reduce absenteeism and school dropout.

Guidelines for the Prevention and Management of Teenage/ Unintended Pregnancy and HIV/AIDS in School Settings (Ministry of Education, Science, Technology and Sports 2015a)

• Guide prevention and management of teenage/unintended pregnancy and HIV/AIDS in school settings.

National Sexuality Education Framework (2018) (Ministry of Education and Sports 2018)

• Guides provision of sexuality information to enable young people to make healthy choices about their Sexual and Reproductive Health (SRH), promotes prevention and protection against sexual abuse, early sexual debut, teenage pregnancies.

National Strategy to End Child Marriage and Teenage Pregnancy (2015)

• Framework to end child marriage, teenage pregnancy and other forms of violence against children.

summarises Uganda’s legal and policy provisions for addressing educational and social inequalities that have a bearing on teenage pregnancy. Uganda operates a duo-legal system that recognises cultural and traditional laws with a mandate that, in case of any conflict, the constitutional provisions prevail (Article 2(2)6 (Republic of Uganda 1995). One of the major contradictions in the law is around the legal age at which people can get married. This is set at 18 years in the constitution while the Customary Marriage Act (1973) sets a lower age of 16 for girls (Article 11a) and also allows the marriage of minors if parental consent is given (Article 32) (Republic of Uganda 1973). In reality, adherence to constitutional provisions remains a challenge as the mandate for constitutional provisions to override inconsistent customary laws is often ignored at community level. Mounting evidence from both the literature (Ochan et al. 2013; Atuyambe et  al. 2015; Ministry of Education, Science, Technology and Sports 2015b) and our research indicate ineffective and inconsistent enforcement of the policies. For instance, the implementation of the policy on re-entry into school for teenage mothers appears to be optional. Schools only allow girls who become pregnant in Primary 7 and Senior 4 to return and sit their national examinations, while students who become pregnant in other years are expelled from school (Ministry of Education, Science, Technology and Sports 2015b). Implementation of some of these macro-level policy initiatives, particularly those related to provision of Sexual and Reproductive Health (SRH)

170  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo information, faces resistance from cultural and religious institutions. For instance, the national framework for the provision of sexuality education in Uganda, developed in 2018, was halted after being challenged by some parents and religious and cultural leaders. They argued that the framework was against the moral values of Ugandan society (Ninsiima et al. 2019; UNESCO 2019). Experience from South Africa illustrates that a successful reduction in teenage pregnancy requires a strong social policy agenda that targets traditional norms and practices. These interventions must work with young people to increase the awareness of their rights and the risks of sexual intercourse (Jewkes et al. 2009). Programmes for reducing teenage pregnancy have focussed, for example, on: sensitisation of the value of girls’ education; mentoring; deployment of female teachers in rural schools; counselling and guidance (provision of SRH information); the distribution of reading materials on avoiding abuse and encouragement to communities to set up customised bye-laws and ordinances to support girls’ education (compulsory education) (Population Secretariat 2015). However, there has been minimal progress in implementation of these programmes, which are often short-term and have limited coverage (Ministry of Education and Sports 2019). Since the implementation of the National Strategy for Girls Education (NSGE), only 17 districts have enacted ordinances for the re-entry of teenage mothers (Ministry of Education and Sports 2019). Another concern is the existence of multiple initiatives addressing similar issues with a lack of coordination within and between sectors. This leads to duplication of efforts and implementation challenges. Further, the initiatives are not anchored in the ecological approach which stresses the need to address issues of teenage pregnancy across the different levels of adolescent girls’ ecology (United Nations Fund for Population Activities 2013). Consequently, despite this robust macro-level legal and policy framework and the interventions described earlier, teenage pregnancy rates have remained static and high in Uganda (see Table 9.2). By exploring the ways in which girls’ lives are shaped by the mesosystem and exosystem of their communities, in the

Table 9.2  Trends in teenage pregnancy rates (%) in Uganda between 1995 and 2016 Age

1995

2000

2006

2011/12

2016/17

15 16 17 18 19 Total (15–19)

7.7 22.1 43.3 64.7 70.8 42.9

3.3 12.9 23.2 54.0 61.2 31.4

1.9 8.5 25.5 41.0 58.6 24.9

1.6 8.5 20.8 37.4 57.6 23.8

3.1 9.4 22.1 40.2 53.9 24.8

Source: Ministry of Finance and Economic Planning, Statistics Department & Macro International Inc. (1996); Uganda Bureau of Statistics and ORC Macro (2001); Uganda Bureau of Statistics and Macro International Inc. (2007); Uganda Bureau of Statistics & ICF International (2012, 2018)

Teenage pregnancy and social inequality in Uganda 171 following discussion we show the various social and economic inequalities and cultural norms that continue to encourage or influence teen-childbearing and education inequality.

Teenage pregnancy: pressures, risks and powerlessness Whilst the prevalence of teenage pregnancy is a global concern – with an estimated 12.8 million births per annum among teenage girls aged 15–19 years – constituting 44 births per 1,000 teenage girls each year – it is noteworthy that adolescent birth rates are eight times higher in low-income countries than in high-income countries (World Health Organisation 2018a). The SSA region experiences the highest prevalence of teenage fertility, and teen-childbirth constitutes more than half of all the births (United Nations Fund for Population Activities 2013). In 2018, evidence indicates that 50% of countries in SSA have an adolescent birth rate of over 100 births per 1,000 women aged 15–19 years; 15 countries in SSA have teenage birth rates ranging from 132 to 2297 births per 1,000 teenagers (World Health Organisation 2018b). Uganda is among the top 10 SSA countries that have a high risk of teenage motherhood and where progress in reducing this has been stagnant. Data from the 2018 World Health Organisation record an average of 141 births per 1,000 girls aged 15–19 years over the period 2007–16 (World Health Organisation 2018b). Table 9.2 shows the proportion of girls (aged 15–19) who had given birth or were pregnant between 1995 and 2016. Table 9.2 also indicates that, although there was a considerable decline in the teenage pregnancy rate from 43% in 1995 to 25% in 2016–17, it remains high.8 Despite the various legal and policy pronouncements outlined in Table  9.1 earlier, teenage pregnancy is still influenced by multiple factors related to social and economic inequalities which characterise the exosystem that shapes young people’s daily lives. In our research in the different regions of Uganda, the study participants associated the rising levels of adolescent pregnancy outside of marriage either with girls themselves or with the social and economic inequalities which characterised their lives. The study participants in the exosystem of the community, for example, blamed teenage pregnancy on the girls, suggesting they were ill-disciplined and did not listen to their parents’ counsel. Moral decay and exposure to videos and discos were noted as responsible for girl’s early engagement in sex. Most of our study participants stated that pregnancy outside of marriage was a growing phenomenon among young girls with a concern that ‘children don’t care at all’. Such loosening of morals among children in Ugandan communities was partly associated with children’s rights, as the adults we interviewed observed; ‘child rights have made it worse’ and that ‘children today do not want to listen; they act with impunity’. Similar findings were found by another study in the Eastern region where adolescents themselves associated teenage pregnancy with moral decay, indiscipline and greed for material things (Sekiwunga & Whyte 2009). However, one could argue that these individual experiences and behaviours are shaped by

172  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo the situations that girls find themselves in. As Odimegwu and Mkwananzi (2016) argue, individual behaviour is a construct of the contextual environment in which individuals live. Whereas teenage pregnancy can be found across all socio-economic groups, evidence shows that girls from poor households are the most vulnerable to teenage pregnancy (Uganda Bureau of Statistics & ICF 2018). At the mesosystem and microsystem levels, natal families, especially those with low economic status, can put considerable pressure on girls, attribute low value to girls’ education and maintain traditional attitudes around early childbirth and motherhood which propel early childbirth and education inequality. In the context of poverty, parents are unable to provide adequately for their children’s well-being. Those girls whose parents had died, leaving the home with limited resources to cater for essential needs and care, were reported to be particularly vulnerable to early childbearing. Poverty creates inadequate living arrangements where children live in crowded houses, potentially exposing children to parents’ marital liaisons. Our research found that girls from poor households were prone to engage in transactional sex with wealthy boys and men in order to meet their basic necessities. These basic necessities include meeting the cost of their education such as school fees and buying materials. An adolescent girl we interviewed reported that ‘when a girl demands for assistance and the parents do not provide [the girl] gets a boyfriend who has money to care for her [one who is] able to pay their school fees’ (adolescent girl, Eastern Uganda). Household poverty coupled with the low value attached to girls’ education impelled some parents to withdraw girls from school: Since most of the girls lack school fees, you end up dropping from school and once you are at home this is where you are likely to get a boyfriend and the end result is early sex, pregnancy and marriage. (Married girl – Central Uganda) Other studies conducted in Uganda and elsewhere in SSA attest to the fact that girls from poor households are more at risk of getting pregnant and being married off, instead of continuing with their education, compared to those in relatively better households (e.g. Lubaale 2013). According to United Nations Fund for Population Activities (2015) and the Uganda Bureau of Statistics and ICF (2018), there are high adolescent birth rates among girls from low-income families compared with those in wealthy families. In other contexts, such as Nigeria, Kenya and Lesotho, teenage births are attributed to large family size and the marital status of parents (Jelili et al. 2013; Oyefara 2011; Ugoji 2011). These studies argue that teenagers of single parents were more likely to become pregnant at an early age than those living with both of their parents. Although there is evidence of shifts in the chronosystem, since attitudes around education for girls are changing, there are lingering negative social norms with parents preferring to invest in their sons’ education than their daughters. This is especially the case for higher education. In our research, we found a widely held

Teenage pregnancy and social inequality in Uganda 173 view that marriage and motherhood is the singular ideal for girls. Some parents expressed worries about their daughter’s failure to get a husband and that they prefer girls to be married as soon as an opportunity arises. Such attitudes and beliefs encourage early sexual relationships and increase vulnerability to teenage pregnancy. Another aspect of parental support relates to the micro-level provision of guidance of young people, especially for adolescent girls. Parents observed that they had failed in their responsibility of guiding and counselling their children on the dangers of early marriage and sex. The girls in our research agreed that most parents failed to provide adequate guidance, especially in relation to information about the dangers of early sex and helping them know how to prevent early pregnancy. This lack of guidance was also associated, in their view, with parents’ low levels of education. They felt parents were themselves ignorant and unable to guide them. The girls in our study reported the lack of information and the skills to resist boys’ and men’s sexual invitations and handle peer pressure. Consequently, amongst all the teen mothers we talked to, their pregnancies were unintended and some ended in forced marriages. One of the girls narrated ‘I got pregnant by accident when I was aged 17. I felt so bad!’ (Teenage mother, Eastern Uganda). Another girl stated ‘I was 15 and I also felt terrible, nobody gave me advice on sexuality because by then I was still young’ (Teenage mother, Eastern Uganda). Some adult participants in the communities we visited raised a concern over lack of parental monitoring of children’s, especially girls’, mobility and that this exposed girls to risks of early pregnancy. There were cases of parents sending their daughters to the trading centres to buy or sell pancakes and tomatoes, and the girls ended up staying out late and meeting boys and men who enticed them with money. Another study conducted in Uganda by Sekiwunga and Whyte (2009) also found inadequate parental guidance and lack of control over children’s mobility as predictors of the persistent high teen childbirths. The interactions between family, schools and peer groups which characterise the mesosystem level are also implicated in the pressures which adolescent girls are under and their lack of power. Schools as sites for new forms of interaction bolster discriminatory gender norms that increase girls’ vulnerability to pregnancy. In our research, we were told that girls experience more challenges compared to boys, especially those related to sexual harassment and enticement from boys and teachers within school and as girls travel to and from school. The girls expressed their powerlessness to report and resist teachers’ harassment with a lack of support from the school administration, including senior female teachers9 who one girl reported mostly as ‘not ready to help’ (adolescent girl, Western Uganda). Harassment of schoolgirls by teachers was reported in all study sites. One of the district officials affirmed the fact that the role of senior female teachers was diminishing, leaving a few older women who – as mothers themselves – were supporting girls in school. Some studies stress the significant role of female teachers in acting as role models, providing guidance and counselling to girls on issues relating to academics and biological development (see Ministry of Education and

174  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo Sports 2012a; Watson et al. 2018). However, the number of female teachers has remained low across the different levels of schooling, with 42.7% and 24.7% of teachers being women in primary and secondary schools respectively. These low proportions of women teachers are especially prominent in rural areas (Ministry of Education and Sports 2017). The risk of sexual violence is exacerbated by the long distances many girls have to travel to and from school. An adolescent mother in Eastern Uganda recounted how walking long distances to school was one of her nightmares because boys chased after her. In addition, peer groups as spaces of interaction for adolescents were noted to encourage early sexual engagements leading to teen births. A focus group of schoolgirls argued that girls got pregnant because of peer groups where ‘other girls tell them that having a boyfriend is cool’ (adolescent girls, Central Uganda) leading to unwanted pregnancy. Ochan et  al. (2013) attribute high levels of teenage pregnancy in Uganda to coerced first sexual intercourse, which they find to be a common experience for many young girls. One way that schools, working at the mesosystem level, can be expected to reduce girls’ risks and strengthen their resilience is by providing new knowledge, information and life skills for a better future. However, girls (and boys) in our research pointed to the limited provision of sexuality and life skills education especially in primary level education. Sexuality education was reported to be provided only in secondary schools, therefore, excluding those girls who had dropped out of primary education. In another study, conducted in Uganda, teenagers reported that the sexuality education offered was inadequate; it was largely prescriptive and feminised (talks are organised by senior women teachers and exclusively for girls), divorced from personal experiences and contradictory (Muhanguzi  & Ninsiima 2011; Haas  & Hutter 2019). Ninsiima et  al. (2019) also point to institutional weaknesses in sexuality education provision. They argue that because schools are primarily concerned with achieving good grades in examinable subjects, non-examinable subjects like sexuality education can be overlooked. Haas and Hutter (2019) note that sexuality education in Ugandan schools is influenced by cultural and religious beliefs around moral upbringing and focuses more on an abstinence-only message. Within this environment, teachers find it difficult to provide comprehensive sexuality education including information that may be regarded as culturally and religious inappropriate for young people such as contraception, safe sex practices, sexual orientation, among others (Haas & Hutter 2019). Similar perspectives of limited sexuality education in schools have been reported in other African contexts in countries such as Nigeria (Aham-Chiabuotu & Aja 2017) and Senegal (Chau et al. 2016).

Patriarchal social norms and practices Another major factor influencing adolescent girls is patriarchal norms and beliefs. These are to be found at all ecological levels. The cultural values expressed by communities and families were found to be deeply shaped by traditional patriarchal social norms. These norms place a high value on women’s fertility and

Teenage pregnancy and social inequality in Uganda 175 creates strong socio-cultural expectations of young girls. At the exosystem level, for example, we found beliefs associated with a girl’s sexual maturation, such as a girl is ‘ripe for the picking’ as soon as their breasts begin to bud; and ‘eito telyenga’ (what is not ready does not ripen). According to some men and boys in our research, premarital pregnancy provides proof of the girl’s fertility as the boy will ‘be sure that the girl can give birth’ and raises the father’s social status as a grandfather. The married adolescent girl recounted the pressure she felt: These days a girl no longer cares about virginity. If she declines men’s sexual advances, they say she is a fool (omudofu), which is socially stigmatising. (Married girl, Eastern Uganda) Young people, especially boys, believed that engaging in sex before marriage would give them experience for their future marital encounters, and that men prefer girls who have had sexual experience. These perceptions encourage early sexual relationships and risk girls getting pregnant. Our research found that with the gradual deterioration of sanctions10 against premarital sex and loss of virginity, engaging in sex at an early age has become a common phenomenon. These experiences combined with women’s low social status make young women vulnerable to acts of sexual violence, such as rape and defilement, by male members in the community. All of the girls we interviewed reported these acts of sexual violence were very common. One of the girls narrated her experience: I got pregnant while on vacation after I was raped. My father didn’t believe my story, but my mother did. . . . The young man accepted to marry me but I refused; he now does not accept the child because I refused to marry him. The community says that I am a ‘left over’ – they wanted to force me into marriage when I was pregnant. . . . My father says he made a loss in me because I refused to get married to the man. (Unmarried teen mother, Eastern Uganda) Lack of support from the family especially male members of the family – the fathers – was noted to have affected community support mechanisms. The structures such as police and local council courts for handling sexual violence cases were reported by girls to be extremely weak. Seeking justice for sexual violence in communities is also compromised by parents seeking compensation and negotiating with the boy’s parents for them to look after the girl and her baby or pay bride wealth.11 In addition, it was reported that, in many cases of rape and defilement, the police solicit for money, release the offenders and drop the cases. One of the key informants noted that about 70% of defilement cases never move beyond the police report stage. Overall, macro-level, patriarchal norms have affected sexual and reproductive health services for adolescent girls to a degree that they remain limited in the country and in the region (Ochan et  al. 2013; Sethuraman et  al. 2018). In our research, study participants, especially SRH services providers, expressed

176  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo difficulties in the provision of adolescent-friendly SRH services in a context where all sexual relations involving under 18s  – consensual or not  – are classified as ‘defilement’ and are against the law. This continues to restrict adolescent girls’ access to information and services to protect themselves against unwanted pregnancies. A non-governmental organisation working in one of the districts visited explained that their policy was to give information only to under 18s and would not issue them with contraception. Some parents and community members blamed those macro elements such as the reproductive health promotion organisations for promoting early sex noting that ‘the promotion of condoms has promoted sexual activities’ among young people. The guidelines from the Uganda Ministry of Health on adolescent SRH services for health workers are ambiguous on the provision of contraceptives to girls under 18 years. Thus, health workers are nervous about giving contraceptives to adolescents in such an unclear policy context). Several studies in SSA report a similar situation of a lack of SRH services for adolescent girls, who are therefore left to seek care only when they are pregnant (Ochan et al. 2013; Sethuraman et al. 2018). Even then, pregnant adolescents are reluctant to seek services due to the fear of being scolded by health workers, stigmatisation by community members, delays and the long distances to the health care facilities (Sethuraman et al. 2018). Our research further found that changes in the socio-economic environment, characterised by a rise in the cash economy, amidst income inequalities also contributed to teenage pregnancy in Uganda. In communities with sugarcane plantations, rice farms and fishing, men (and boys) sought out labour activities which earned them money that was then reportedly used for transactional sex with young girls with consequences of teenage pregnancies and early marriage. There was reported proliferation of trading centres in districts with video and disco halls, accompanied by new forms of communication and social behaviours including drug abuse and alcoholism. These changes exposed young girls to early sex and, as a result, increased the rates of teenage pregnancies in Ugandan communities. While teenage pregnancy is known to be a marker of limited education and economic opportunities, evidence also suggests that teenage pregnancies curtail women’s educational prospects and economic potentials. In the next section we highlight the impact of teenage pregnancy on female education attainment in Uganda.

Education cut short by teenage pregnancy Whilst the various ecological levels associated with poverty and tradition shape the likelihood of high teenage pregnancy rates, teenage pregnancy itself potentially has a major role to play in shaping the differential educational achievement of girls and boys in a country like Uganda. The ecological system that surrounds and positions young women is dynamic and interactive. Evidence from studies conducted in SSA countries such as Chad, Guinea, Mali, Niger, South Sudan (Lopez Calix et al. 2018; Wodon et al. 2018; Sethuraman et al. 2018), for

Teenage pregnancy and social inequality in Uganda 177 example, indicates that many girls continue to drop out of school due to teenage pregnancies and end up getting married. In Uganda, the second main reason for girls leaving school prematurely is pregnancy, the first being a lack of school fees (Ministry of Education and Sports 2012b; Ochan et al. 2013; Ministry of Education, Science, Technology and Sports 2015b; Sebudde et al. 2017). In our research, despite the re-entry policy, teachers and school administrators clearly stated that once a girl becomes pregnant, she drops out of school and the chances of her ever returning are limited. In the communities we visited, it was common for parents to force their pregnant daughters to marry the boy or man responsible for the pregnancy. Forced marriage is an automatic impediment to further schooling, as girls drop out to care for their babies and families. The sudden ceasing of education for the adolescent girls is rooted in the cultural and gender norms held by many communities in Uganda which give low value to girls’ education and perceive negatively premarital pregnancy. Teenage pregnancy carries stigma and shame for both the girls and their parents in the community and at school. A girl who becomes pregnant and has a child, whilst still in her father’s home, is referred to as ‘a curse because she fits into no specific stage in life’, ‘a disgrace’, ‘useless’, ‘wasted’, ‘worthless’, is ‘not loved’, becomes a ‘laughing stock’ and a ‘burden to her parents’. Girls can be despised and isolated and ‘may even commit suicide’. The teenage mothers, especially the unmarried girls, we interviewed reported that they had to endure trauma, lack of self-confidence, feeling out of place and being a burden to the parents. In 2015, the Ministry of Education, Science, Technology and Sports (2015b) found strong negative perceptions about teenage mothers who were thought to be bad examples and failures who are not expected to return to school but rather get married. Unmarried teenage mothers in our study were also looked at as bad examples/role models to the rest of the girls in their communities. Fellow students were negative and nicknamed the girls while they were pregnant or have ever given birth as ‘scrap’ and ‘second hand’. According to some study participants, teenage pregnancy is associated with a lack of morals and values and thus no school could accept the girls back. In addition, such girls can lack basic necessities and school requirements as their ‘parents are already annoyed and will not want to help’. Early childbearing was reported to be one of the main factors propelling girls into early marriage as it is often seen as the best option. While there is appreciation of returning to school for girls who become pregnant, girls and parents in our research suggested that such girls should be taken to other schools outside their natal communities or build their own schools. Our findings further show that, at the mesosystem, teenage pregnancy also concerned boys. Boys who are linked to the pregnancy are also negatively perceived in the community- ‘treated as evils who only destroy and put shame on families’. These young fathers expressed their vulnerability when they have to face the law and said that the pregnancy was not intended, they were ‘just having fun’. They are forced to marry the girls when they are not prepared and thus some drop out of school while others go into hiding and relocate to other

178  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo communities. Most of the married boys we interviewed had been forced to marry girls they had made pregnant. Early exit from school means that girls leave without acquiring the skills required for gainful employment in already competitive labour markets. Consequently, the girls remain trapped in a cycle of poverty. In all the districts we visited, girls who had dropped out of school faced severe challenges in finding well-paid and productive employment. For instance, in Eastern Uganda, adolescent boys and girls who had dropped out of school were reportedly working in unskilled employment, for example, in rice and sugarcane farms. They often faced hazardous conditions, poor remuneration, threats of sexual and genderbased violence and exploitation. Studies conducted in Uganda and in other African countries show that the social and economic disadvantages associated with dropping out of school for girls have lifetime negative consequences for them, their communities and the nation at large. These negative consequences include: high reproduction; low earnings; poor standards of living; lower psychological well-being; limited agency in their adulthood and, the propagation of disadvantage and gender discrimination over time (Wodon et al. 2016; Wodon et al. 2017; Wodon et al. 2018; Sebudde et al. 2017; Sethuraman et al. 2018). Teenage childbearing thus leads to a generational cycle of missed opportunities for education, as these girls are unable to care and support the education of their children.

The role of education in ending teenage pregnancy: a mesosystem challenge It is well established that female education has multiplier effects for the individual girl, her community and overall economic growth (Sebudde et al. 2017). Keeping girls in school is regarded a key strategy in the efforts to reduce teenage pregnancy. Evidence from low- and middle-income countries including Uganda shows that girls’ education, particularly secondary education and above, remarkably reduces teenage childbearing (and child marriage) (Uganda Bureau of Statistics  & ICF 2018; Wodon et  al. 2017, 2018). Similarly, evidence shows that for each additional year of secondary schooling an adolescent girl receives, the risk of giving birth before her eighteenth birthday is reduced by 5.8 percentage points (Wodon et al. 2018). This important role of education in preventing pregnancy has also been reported in other studies in SSA (Ahorlu et al. 2015; United Nations Fund for Population Activities 2015; Sethuraman et al. 2018). As mesosystems are sites where children spend a significant portion of their time into adolescence, schools serve as intermediate institutions for the change of social norms and practices that create vulnerabilities in the life cycle of girls (Muhanguzi et al. 2017). In our research, the young girls and boys testified that schools served as sources of new information and learning where they gained knowledge (albeit limited) about SRH issues which they had not got from their senga (paternal aunt) or parents. Provision of sexuality education in schools has been found to contribute to a reduction in teen childbearing. A number of studies

Teenage pregnancy and social inequality in Uganda 179 conducted in the region attest to this (Chau et al. 2016; Aham-Chiabuotu & Aja 2017; Haas & Hutter 2019; UNESCO 2019). Our research further shows that schools are also seen as establishing processes whereby children – both boys and girls – increasingly consider aspiring to a better and higher standard of life. Adolescent girls in our research consistently talked about role models including former students who had graduated and moved on to successful endeavours. Evidence from SSA shows that adolescents, especially girls who see a future for themselves, are less likely to engage in early sexual relations that expose them to early pregnancy (Biddlecom et al. 2007). Indeed many girls said they postponed becoming involved in sexual liaisons in order to protect their goals and aspirations in life such as desiring to be ‘a responsible person in the country’, ‘join University’, ‘be rich’, ‘gain respect of others’, ‘be a doctor’, ‘be a nurse’, ‘be self-reliant’, ‘educate their children’, ‘earn lots of money’, ‘be confident’, ‘be a member of parliament’ and ‘an information technologist’. Schools as mesosystem spaces of interaction among young people and with adults (teachers) are also important for building young people’s social capital. In our research, we found that those girls who interacted with other girls formed bonds and consequently were building up social capital that went beyond the household and neighbourhood, moving into the ecosystem that reshapes young women’s opportunities. Their psycho-social well-being was positively affected by strong relationships, for example, with senior women teachers who provided girls with adult support structures. These support structures could either complement or palliate deficiencies of actual parental support – especially on sexual and reproductive matters. Our research found that greater social value was placed on providing education for girls since it was a means to expand thinking capacity and hence being able to resist the temptation of early sexual relationships. Evidence from Uganda and elsewhere shows the need to invest more in education beyond primary schooling. Wodon et al. (2018), in their study that included some countries in SSA, argue that investing in secondary education for girls is valuable as it increases their aspirations for further development of their skills. Girls’ completion of secondary school enables them to secure the foundational skills needed to succeed in the labour market and live rewarding lives (Wodon et al. 2018). The persistent high rates of teenage pregnancy challenge the attainment of education for all in Uganda and most SSA countries. As we have shown, these high incidences of teenage pregnancy are caused by multiple factors largely related to social and economic inequality operating at different levels of the adolescent girls’ ecology – namely individual, family, school and community. These factors range from poverty, lack of parental guidance and monitoring, sexual violence in schools and limited provision of SRH information and services. Consequently, girls do not attain the necessary skills for gainful employment and decent work. Low levels of education for girls lead to high fertility rates, poor standards of living and a lack of agency in their adulthood. While there is a strong macro-level policy and programme commitment to address teenage pregnancy in Uganda, the expansion of education opportunities

180  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo and their poor implementation has not, it seems, effectively addressed the challenges of teenage pregnancy. There is only limited improvement in educational attainment indicators especially for young women and girls. Chronosystem strategies to address teenage sexual and reproductive health and rights, such as the provision of sexuality information, are reported to have been met with resistance, especially among the religious and cultural leaders and some parents. Initiatives to address teenage pregnancy should target the multiple social and economic factors that exist within the girl’s ecology. Education is a key strategy for enhancing a girl’s life skills and aspirations, for delaying engagement in early sex and encouraging social values that envision a different future for the girl beyond childbearing. This however requires provision of education beyond the adolescent girl, to include other people in her social ecology including parents, boys, men and the community at large. There is a clear need to further research the intersection between social and education inequalities in relation to teenage pregnancy if attainment of education for all is to be achieved. In-depth research into existing interventions at the different ecological levels affecting girls’ lives will help identify the most effective strategies with which to address teenage pregnancy.

Notes 1 This comprised the laws, the general economic status of the country, investment in human capital, political and humanitarian crises. 2 The research was supported by the Overseas Development Institute with financial support from the Department for International Development (C0101004 Flagship 4) and the United Nations Children’s Fund (UNICEF) (43170881). 3 Moroto, Kapchorwa, Amuru, Arua, Bundibugyo, Bugiri, Mayuge, Kiboga, Rukungiri, Masaka, Sembabule and Kampala representing Northern, Central, Eastern and Western regions. 4 8 for unmarried and 8 for married teen mothers. 5 34 FGDs for girls and 30 FGDs for boys. 6 If any other law or any custom is inconsistent with any of the provisions of this Constitution, the Constitution shall prevail and that other law or custom shall, to the extent of the inconsistency, be void (Republic of Uganda 1995, p. 32). 7 Tanzania and Central African Republic respectively. 8 Uganda Bureau of Statistics and ICF 2012, 2018. 9 These are female teachers charged with helping guide, advise and support girls. 10 Sanctions range from chasing the girl away from home to deliver at an aunt’s home, forcing the girl into marriage, cutting off her ears, setting her ablaze, beating and even killing her by throwing her over the cliff on a waterfall, ritual cleansing, among others as varied from one culture to another. 11 Bride wealth is a gift in the form of either money or property or other form of wealth paid by a boy (groom or his family) to the family of the girl (bride) he will be married to symbolising (among other things) the transfer of a girl’s productive and reproductive capacity from the father’s clan to that of her husband.

References Aham-Chiabuotu, C.B., & Aja, G.N. (2017) There is no moral they can teach us’: Adolescents’ perspectives on school-based sexuality education in a semiurban,

Teenage pregnancy and social inequality in Uganda 181 southwestern district in Nigeria. American Journal of Sexuality Education, 12(3), pp. 315–336. Ahorlu, C.K., Pfeiffer, C., & Obrist, B. (2015) Socio-cultural and economic factors influencing adolescents’ resilience against the threat of teenage pregnancy: A crosssectional survey in Accra, Ghana. Reproductive Health, 12(117), pp. 1–11. Atuyambe, L.M., Kibira, S.P.S., Bukenya, J., Muhumuza, C., Apolot, R.R.,  & Mulogo, E. (2015) Understanding sexual and reproductive health needs of adolescents: Evidence from a formative evaluation in Wakiso District, Uganda. Reproductive Health, 12(35), pp. 1–10. Biddlecom, A., Hessburg, L., Singh, S., Bankole, A., & Darabi, L. (2007) Protecting the Next Generation in Sub-Saharan Africa: Learning from Adolescents to Prevent HIV and Unintended Pregnancy. [Online] New York: Guttmacher Institute. Available from: www.guttmacher.org/report/protecting-next-generation-sub-saharanafrica [Accessed 18 June 2020]. Bronfenbrenner, U. (1979) The Ecology of Human Development: Experiments by Nature and Design. Cambridge, MA and London: Harvard University Press. Bronfenbrenner, U. (1994) Ecological models of human development. In International Encyclopaedia of Education 3, 2nd ed. Oxford: Elsevier, pp. 37–43. Chau, K., Traoré Seck, A., Chandra-Mouli, V.,  & Svanemyr, J. (2016) Scaling up sexuality education in Senegal: Integrating family life education into the national curriculum. Sex Education, 16(5), pp. 503–519. Chohan, Z.,  & Langa, M. (2011) Teenage mothers talk about their experience of teenage motherhood. Agenda, 25(3), pp. 87–95. Government of Uganda (1992) Government White Paper on the Education. Kampala: Education for National Integration and Development. Haas, B., & Hutter, I. (2019) Teachers’ conflicting cultural schemas of teaching comprehensive school-based sexuality education in Kampala, Uganda. Culture, Health and Sexuality, 21(2), pp. 233–247. Jelili, M.O, Akindele, O.A.,  & Akintayo, O. (2013) Teenage pregnancy and home environment factors in Ogbomoso, Nigeria. Research on Humanities and Social Sciences, 3(18), pp. 69–77. Jewkes, R., Morrell, R.,  & Christofides, N. (2009) Empowering teenagers to prevent pregnancy: Lessons from South Africa. Culture, Health and Sexuality, 11(7), pp. 675–688. Krugu, J.K., Mevissen, F., Münkel, M., & Ruiter, R. (2017) Beyond love: A qualitative analysis of factors associated with teenage pregnancy among young women with pregnancy experience in Bolgatanga, Ghana. Culture, Health and Sexuality, 19(3), pp. 293–307. Lopez Calix, J.R., Lemiere, C., & Moller, L.C. (2018) Disrupting the Gender Divide in Mali, Chad, Niger and Guinea. [Online] Available from: http://documents. worldbank.org/curated/en/605471541607872022/pdf/Disrupting-the-Gen der-Divide-in-Mali-Chad-Niger-and-Guinea.pdf [Accessed 22 June 2020]. Lubaale, Y.M.A. (2013) Child marriages in Uganda after enacting the death penalty for defilement. Psychology and Behavioural Sciences, 2(3), pp. 138–147. Ministry of Education and Sports (MoES). (2012a) A Report on a Study of Female Primary Teacher Training Recruitment and Deployment in Uganda. [Online] The Republic of Uganda. Available from: www.education.go.ug/wp-content/ uploads/2019/07/gender_FemaleTeacher-Study.pdf [Accessed 18 June 2020]. Ministry of Education and Sports (MoES). (2012b) Report on Dropout Study in Universal Secondary Education (USE). [Online] The Republic of Uganda. Available

182  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo from: www.education.go.ug/wp-content/uploads/2019/08/Dropout-Report-inUSE-schools-2012.pdf [Accessed 18 June 2020]. Ministry of Education and Sports (MoES). (2013) National Strategy for girls’ Education in Uganda (2015–2019). [Online] Republic of Uganda. Available from: https://planipolis.iiep.unesco.org/en/2013/national-strategy-girls-educationuganda-2015-2019-6144 [Accessed 18 June 2020]. Ministry of Education, Science, Technology and Sports (MoESTS) (2016) Gender in Education Sector Policy. [Online] Kampala: Republic of Uganda. Available from: https://planipolis.iiep.unesco.org/en/2016/gender-education-sector-policy2016-6453 [Accessed 18 June 2020]. Ministry of Education and Sports (MoES) (2017) Education and Sports Sector Facts Sheet 2002–2016. [Online] Republic of Uganda. Available from: www.education. go.ug/wp-content/uploads/2019/07/FACT-SHEET-2016.pdf [Accessed 18 June 2020]. Ministry of Education and Sports (MoES) (2018) National Sexuality Education Framework. [Online] Republic of Uganda. Available from: https://s3-eu-west-1. amazonaws.com/s3.sourceafrica.net/documents/119376/UNFPA-68-090518. pdf [Accessed 18 June 2020]. Ministry of Education and Sports (MoES) (2019) Review of Ministry of Education and Sports (MoES) National Strategy for Girls’ Education (NSGE) in Uganda. [Online] Republic of Uganda. Available from: www.ungei.org/resources/files/ Resource_Uganda_NSGE.pdf [Accessed 18 June 2020]. Ministry of Education, Science, Technology and Sports (MoESTS) (2015a) Guidelines for Prevention, Management of Teenage/unintended Pregnancy and HIV/ AIDS in School Settings. Kampala: MoESTS. Ministry of Education, Science, Technology and Sports (MoESTS) (2015b) A Study on Linkage between Pregnancy and School Dropout in Uganda. Kampala: MoESTS. Ministry of Finance and Economic Planning, Statistics Department & Macro International Inc. (1996) Uganda Demographic and Health Survey, 1995. [Online] Republic of Uganda and Macro International Inc. Available from: www.dhsprogram.com/pubs/pdf/FR69/FR69.pdf [Accessed 18 June 2020]. Ministry of Health, Uganda (2012) National Adolescent Health Policy Guidelines and Service Standards. Kampala: Ministry of Health, Uganda. Available from: http://library.health.go.ug/publications/adolescent-health/adolescent-healthpolicy-guidelines-and-service-standards Muhanguzi, F.K., Kyomuhendo, G.B.,  & Watson, C. (2017) Social institutions as mediating sites for changing gender norms: Nurturing girl’s resilience to child marriage in Uganda. Agenda, 31(2), pp. 109–119. Muhanguzi, F.K., & Ninsiima, A. (2011) Embracing teen sexuality: Teenagers’ assessment of sexuality education in Uganda. Agenda, 25(3), pp. 54–63. Ninsiima, A.B., Coene, G., Michielsen, K., Najjuka, S., Kemigisha, E., Ruzaaza, G.N., Nyakato, V.N., & Leye, E. (2019) Institutional and contextual obstacles to sexuality education policy implementation in Uganda. Sex Education, 20(1), pp. 1–16. Ochan, W., Nalugwa, C., & Apuuri, F.A. (2013) Too young for motherhood: Profile, consequences and drivers of teenage pregnancy in Uganda. In The State of Uganda Population 2013. [Online] The Republic of Uganda  & United Nations Population Fund. Available from: http://npcsec.go.ug/wp-content/uploads/2013/06/ State-of-Uganda-Population-Report-2013.pdf [Accessed 18 June 2020].

Teenage pregnancy and social inequality in Uganda 183 Odimegwu, C.,  & Mkwananzi, S. (2016) Factors associated with teen pregnancy in sub-Saharan Africa: A  multi-country cross-sectional study. African Journal of Reproductive Health, 20(3), pp. 94–107. Oyefara, J.L. (2011) Determinants of adolescent fertility in contemporary Yoruba society: A multivariate analysis. Gender and Behaviour, 9(2), pp. 3979–4004. Population Secretariat (2015) The State of Uganda’s Population 2015. [Online] Kampala: Republic of Uganda. Available from: http://popsec.org/wp-content/uploads/2015/12/THE-2015-STATE-OF-UGANDA-POPULATIONREPORT.jpg Republic of Uganda (1973) Customary Marriage (Registration) Act CAP 248. Kampala: Republic of Uganda. Republic of Uganda (1995) The Uganda Constitution of the Republic of Uganda. Kampala: Republic of Uganda. Republic of Uganda (2008) The education (Pre-Primary and Post-Primary) Act 13. Acts Supplement No. 8. Uganda Gazette, 44(CI), pp. 1–84. Sebudde, R.K., Wodon, Q.T., & Mawejje, J. (2017) Accelerating Uganda’s Development: Ending Child Marriage, Educating Girls. [Online] Available from: http:// documents.worldbank.org/curated/en/553381512398131516/Uganda-economic-update-tenth-edition-accelerating-Ugandas-development-ending-childmarriage-educating-girls [Accessed 22 June 2020]. Sekiwunga, R., & Whyte, S.R. (2009) Poor parenting: Teenagers’ views on adolescent pregnancies in eastern Uganda. African Journal of Reproductive Health, 13(4), pp. 113–127. Sethuraman, K., Tara, K., & Sommerfelt, A.E. (2018) Adolescent Pregnancy and Its Impact on the Prevalence of Stunting: Programatic Considerations for Food for Peace Programs that Aim to Reduce Stunting. [Online] Washington, DC: FHI 360/ FANTA. Available from: www.fantaproject.org/sites/default/files/resources/ FANTA-Adolescent-Pregnancy-Brief-Sep2018_0.pdf [Accessed 18 June 2020]. Uganda Bureau of Statistics  & ICF International (2012) Uganda Demographic and Health Survey 2011. [Online] The Republic of Uganda & ICF International Inc. Available from: https://dhsprogram.com/pubs/pdf/FR264/FR264.pdf [Accessed 22 June 2020]. Uganda Bureau of Statistics  & ICF International (2018) Uganda Demographic and Health Survey 2016. [Online] The Republic of Uganda  & ICF. Available from: https://dhsprogram.com/pubs/pdf/FR333/FR333.pdf [Accessed 22 June 2020]. Uganda Bureau of Statistics and Macro International Inc. (2007) Uganda Demographic and Health Survey 2006. [Online] Calverton, MD: UBOS and Macro International Inc. Available from: https://www.dhsprogram.com/pubs/pdf/ FR194/FR194.pdf [Accessed 22 June 2020]. Uganda Bureau of Statistics and ORC Macro (2001) Uganda Demographic and Health Survey 2000–2001. [Online] Republic of Uganda  & Macro International Inc. Available from: https://dhsprogram.com/publications/publication-fr128dhs-final-reports.cfm?cssearch=126341_1 [Accessed 22 June 2020]. Ugoji, F. (2011) Parental marital status and peer influence as corelates of teenage pregnancy among female teens in south-south Nigeria. Gender and Behaviour, 9(2), pp. 4125–4138. UNESCO (2017) Global Education Monitoring Report 2017/8. [Online] Available from: https://unesdoc.unesco.org/ark:/48223/pf0000259338 [Accessed 22 June 2020].

184  Florence Kyoheirwe Muhanguzi and Grace Bantebya Kyomuhendo UNESCO (2018) Global Education Monitoring Report Gender Review: Meeting Our Commitments to Gender Equity in Education. [Online] Available from: https:// en.unesco.org/gem-report/2018_gender_review [Accessed 22 June 2020]. UNESCO (2019) Policy Paper 39. Facing the Facts: The Case for Comprehensive Sexuality Education. [Online] Available from: https://en.unesco.org/gem-report/ node/2791 [Accessed 22 June 2020]. United Nations Fund for Population Activities (2013) The State of the World Population 2013. Motherhood in Childhood: Facing the Challenges of Adolescent Pregnancy. [Online] Available from: www.unfpa.org/sites/default/files/pub-pdf/ EN-SWOP2013.pdf [Accessed 22 June 2020]. United Nations Funds for Population Activities (2015) Girlhood, Not Motherhood: Preventing Adolescent Pregnancy. [Online] Available from: www.unfpa.org/sites/ default/files/pub-pdf/Girlhood_not_motherhood_final_web.pdf [Accessed 22 June 2020]. Walker, J.A. (2012) Early marriage in Africa: Trends, harmful effects and interventions. African Journal of Reproductive Health, 16(2), pp. 231–240. Watson, C., Kyomuhendo, G.B., & Muhanguzi, F.K. (2018) The paradox of change and continuity in social norms and practices affecting adolescent girls’ capabilities and transitions to adulthood in rural Uganda. In Harper, C., Jones, N., Ghimire, A., Marcus, R., & Kyomuhendo, G.B. (Eds.) Empowering Adolescent Girls in Developing Countries: Gender Justice and Norm Change. London: Routledge, pp. 83–101. Wodon, Q., Male, C., Onagoruwa, A., Savadogo, A., & Yedan, A. (2017) The Cost of Not Investing in Girls: Child Marriage, Early Childbearing, Low Educational Attainment for Girls, and Their Impacts in Uganda. [Online] World Bank. Available from: http://pubdocs.worldbank.org/en/297781512451885312/The-Cost-ofNot-Investing-in-Girls-Child-Marriage-Early-Childbearing-Low-EducationalAttainment-for-Girls-and-Their-Impacts-in-Uganda.pdf [Accessed 22 June 2020]. Wodon, Q., Montenegro, C., Nguyen, H., & Onagoruwa, A. (2018) Missed Opportunities: The High Cost of Not Educating Girls. [Online] World Bank. Available from: https://openknowledge.worldbank.org/bitstream/handle/10986/29956/High CostOfNotEducatingGirls.pdf?sequence=6&isAllowed=y [Accessed 22 June 2020]. Wodon, Q., Nguyen, M.C.,  & Tsimpo, C. (2016) Child marriage, education, and agency in Uganda. Feminist Economics, 22(1), pp. 54–79. World Health Organisation (2018a) World Health Statistics 2018: Monitoring Health for SDGs. [Online] Available from: https://apps.who.int/iris/bitstream/han dle/10665/272596/9789241565585-eng.pdf?ua=1 [Accessed 22 June 2020]. World Health Organisation (2018b) Facts Sheet: Adolescent Pregnancy. [Online] Available from: www.who.int/news-room/fact-sheets/detail/adolescent-pregnancy [Accessed 22 June 2020].

10 Complementary basic education Parental and learner experiences and choices in Ghana’s northern regions Leslie Casely-Hayford with Adom Baisie Ghartey and Justice Agyei-Quartey1 According to the World Bank (2015) report on Out of School Youth in Africa, the number of out-of-school youth on the continent is estimated to be 89 million. This gives an indication of the challenges that governments face in the coming decade, given extreme inequality, poverty, migration and conflict. The effect of the high incidence of out-of-school youth (OOSC) will not only be felt by youth themselves, but it has a tremendous impact on intergenerational inequality in society as a whole (ibid.). A number of key supply and demand factors represent key challenges for governments attempting to address the OOSC problem in sub-Saharan Africa (SSA). These are: •

•

Supply factors: communities having limited access to primary schools within a 3–5 km radius, poor quality of primary schooling due to lack of trained teachers, poor supervision and time on task, etc.; Demand factors: poverty and direct and indirect costs of primary schooling, socio-cultural practices including child fosterage and early marriage, growing incidence of teenage pregnancy, unsustainable livelihood patterns among households, high demand for child labour at home and on the farm and special problems of children being trafficked or engaged in seasonal migration.

Further, evidence from SSA reveals that there are major barriers and bottlenecks preventing access to, and participation in, basic education (UNESCO 2017; Associates for Change 2017; Farrell & Hartwell 2008). Information concerning these deterrents within the Ghanaian context can be found in Casely-Hayford and Ghartey (2008) and the Out of School Ghana Study (CBE Management Unit 2016; UNICEF 2012).2 Re-entry into and completion of primary education remains a major challenge, even though there has been a reduction in the number of those children who are affected at primary and lower secondary school levels. Ghana is still faced with a significant number of 5–14-year-old children out of school (CBE Management Unit 2016; UNICEF 2012; Casely-Hayford & Hartwell 2010). Non-enrolment and poor school attendance in Ghanaian

186  Leslie Casely-Hayford et al. primary schools are attributed to the direct and indirect costs of primary education including in some cases school ‘fees’, exam fees and Parent Teacher Association (PTA) dues. The introduction of the capitation grant has reduced the parental burden in terms of the payment of school fees and levies, yet the capitation grant is still associated with irregular disbursement and grant inadequacy. There are also other direct and indirect costs incurred by parents such as the cost of transportation, exercise books, pencils, food and sanitary materials for girls. These indirect costs increase as children progress to higher levels of education (UNICEF 2012; Casely-Hayford & Hartwell 2010). Poor quality education across SSA (including Ghana) represents one of the greatest barriers to the participation and retention of primary-age children and is reinforced by the inefficient educational resource allocation across the most deprived regions. Casely-Hayford (2019) found that parents were experiencing poor quality of teaching and learning outcomes at primary schools, especially in the rural areas of the country, hence preventing them from sending more children to school. The UNICEF (2012) study reported that falling quality standards were attributed to high rates of teacher absenteeism and lack of effective supervision on the part of authorities, particularly in these areas. Findings from the Ghana Living Standards Survey (GSS 2016) and data from the Complementary Basic Education Management Unit3 also highlight a lack of school infrastructure/school buildings and the unwillingness of trained teachers to serve in remote areas (CBE Management Unit 2016). In this context, this chapter explores the choices made by parents to send their children to one of the Ghanaian government’s flagship programmes which addresses the out-of-school challenge and provides alternative educational pathway for children in remote rural areas – Complementary Basic Education (CBE). This was an initiative by the civil society sector in 1995 through an NGO called School for Life. Ghana’s CBE programme aims to provide education for outof-school children between eight and 14  years of age from the most poverty endemic areas of the country, with the ultimate goal of (re)integrating them into mainstream primary schools. The selection of out-of-school children or potential learners for the CBE programme is dependent on them not already attending any formal school, having either dropped out or never attended. The government-led CBE programme is still in its initial years of implementation through government but was implemented through NGO support from 1995 to 2018 before the Ministry of Education took over the programme in 14 districts across the country in 2018. There is no compulsion on the part of parents to ensure their OOSC wards attend CBE classes, but it is encouraged; as of 2019–20, this CBE option is available only in 17 districts across Ghana. The programme is often located in rural and remote communities which have a person able to read and write in the local language of the community and who can act as a volunteer CBE facilitator (Ministry of Education, 2013). Approximately 90% of children who complete the nine-month CBE afternoon classes in literacy and numeracy in their mother tongue have been able to integrate into the upper primary system from class 4 to class 6 (IMC 2018; Kantur 2018).

Complementary basic education in Ghana 187 Between 2013 and 2016, there was evidence that some parents and children were opting to support their children in CBE classes rather than formal primary education, first, because of the educational quality, relevance and flexibility that the programme offered particularly to households in the poorest quintiles in Ghana’s three northern regions (CBE Management Unit 2015; Ghana Statistical Service (GSS) 2016). Second, there were also indications from predominantly migrant and ethnic minority language areas that the CBE model was more responsive to these communities than primary schools. Further, CBE was often seen as the only viable option for families who need their children to support their daily livelihood (Associates for Change, 2013). These varied contexts have influenced the need to pursue a qualitative inquiry into the motivational factors behind parental choice in relation to complementary basic education where extreme poverty persists. Hence the key research questions in our study focused on the extent to which parents and out-of-school/drop-out children living in rural areas preferred to participate in CBE classes compared to formal primary education and the key factors which influenced their educational choices. This allowed us to address the perceived interaction between the poverty and quality of education factors shaping children’s attendance of CBE, rather than formal education. We addressed the following questions: (1) What are the factors which deter parents from choosing formal primary education and favouring CBE? (2) Are these factors affecting girls and boys differently? (3) What are the beliefs and values which influenced educational choices and to what extent were they affected by gender, religion and socio-cultural traditions? Our chapter starts by providing an overview of the core demographic characteristics of two northern Ghana research districts, their communities and their family, communal social structures, livelihood patterns, religion, access to primary school and other key educational factors. The second section describes the different views shared by the various respondents concerning the options and circumstances in which community stakeholders, including parents, preferred to support their children’s participation in CBE. Finally, we explore how their educational choices were being made within the family setting and the circumstances in which these choices were being made in relation to the child’s sex and position within the family. In the following section we outline our study in two northern regions: the Upper East (UER) and the Northern (NR) Region of Ghana.4

The research study In-depth qualitative fieldwork was carried out by the research team5 which focussed on three districts in each of three northern regions.6 A purposeful and stratified sampling framework was used to select the districts, using a range of context-related factors such as: high incidence of child labour; relatively high

188  Leslie Casely-Hayford et al. access per capita to formal schooling; high incidence of out-of-school children and limited access to formal schools. Five communities across the three districts were identified using other background characteristics such as the number of CBE cycles that the district had experienced and implemented, the transition rates of children entering the formal education system from CBE programme cycles 3 and 4, the availability of primary schools within the district and the access of children to these schools within a 3–5 km radius of the community using cycle 3 data.7 In this chapter, we draw on data from two sampled districts: the Gushegu District in the Northern Region and the Pusiga District in the Upper East Region. The five communities in Gushegu are labelled for the sake of anonymity, GC1, GC2 GC3, GC4 and GC5; the Pusiga communities are listed as PC1, PC2, PC3, PC4 and PC5. Figure 10.1 provides an overview of the actual incidence of OOSC across these two districts alongside other districts in these two regions. It demonstrates that, in 2016, the Gushegu District had the highest incidence of OOSC children (11,130) whilst Pusiga had the lowest incidence (2,782). Figure 10.1 also indicates that 72 (65%) out of 111 active CBE communities in the Gushegu District had no access to formal schools within a 3–5 km radius. At least 19 (36%) out of 53 active communities without primary schools in communities in East Mamprusi (with 7,214 OOSC) and in the Mion District (with 7,139 OOSC), at least 30 (39%) out of the 77 target CBE communities were without local primary schools. In contrast, with only 2,782 OOSC in the Pusiga

120

12000

100

10000

80

8000

60

6000

40

4000

20

2000

0

Nanumba North

East Mamprusi

Total CBE communities

Gushegu

Kpandai

Jirapa

Nanumba South

Mion

No. of communities without a formal primary school

Pusiga

0

Out-of-school

Figure 10.1  Incidences of OOSC and access to primary schools Source: Casely-Hayford et al. (2018) Note: The left-hand axis refers to total CBE communities and number of communities without a formal primary school. The right-hand axis refers to the number of out-of-school children.

Complementary basic education in Ghana 189 District, there were only four (11.7%) out of 34 active CBE communities which had no formal schools within that radius. In-depth interviews and focus group interviews (gender disaggregated) were held with 150 parents across the two districts8 with over 300 CBE learners9 and 200 non-CBE learners from the 2017–18 CBE (cycle 5) programme to capture the factors affecting parental and learner choice. These were followed by focus group discussions with school management committees (SMCs)/local committees (LCs), chiefs and elders, integrated cycle 4 learners in the primary schools and children in upper primary who had never participated in the CBE programme. Subsequently, in-depth interviews were held with facilitators and head teachers.10 The chapter draws on slices of our data starting with the demographic and contextual features of UER and NR households and families participating in CBE.11

Contextual factors affecting the use of complementary basic education at family and community levels The contextual factors of age, gender, religion and ethnicity play a vital role in these households’ and communities’ ability to engage in formal schooling and complementary basic education. From a demographic perspective, it is important to note that the two selected study districts had a high proportion of youth (16–35), comprising between 40 and 50% of the population (GPHC 2010). The average household size was also very large often varying between 10 and 12 children under the age of 15. For example, in the Pusiga District (UER), the research team found that the average Muslim household size had approximately 18 people with at least six children per wife and with three wives per husband, compared to Christian households of six people where men rarely practice polygamy. Children between five and nine years of also constituted the largest proportion of the household – with 17% in the Gushegu District and 14% in the Pusiga District. Not least because of migration, several households did not have any active adult population between 30 and 50  years of age. The characteristics of some households revealed that an elderly and ageing adult population was in charge of providing for the families’ basic needs in the absence of an adult workforce. For instance, in the Pusiga District, some households were headed by grandparents since their parents had migrated south to work for indefinite periods of time, rarely remitting funds to sustain the family. The gender power structure and socio-cultural patterns within the family define the women’s ability to support their children in school and out of school. Our research found that the high proportion of married and widowed women who had had no education had a negative impact on the value given to education for children, particularly for girls – a point we return to later. There were also different ethnic/language groups and variations in the religious make-up of the population although, in both regions, the predominant religion was Islam. Islam was the major religion in the Pusiga District (close to the Burkina Faso Border), with Christians found in all five selected communities. Findings from our interviews also suggest that the religious background of

190  Leslie Casely-Hayford et al. a family/community affected the relationship between men and women. Some men justified their ability to support only a few children from each of their wives in order to ensure equity in the household – therefore not all of each women’s children were enrolled in formal schooling. In some families, each wife was given an opportunity to educate only one child in the formal primary education system so as to make it equitable to all wives in the household. The Gushegu District with the highest proportions of OOSC in Ghana was predominantly Islamic and traditionalist12 – with these populations representing some 68.1% and 22.2% respectively, with Christians making up an estimated 8% (GSS 2010). There were ongoing ethnic tensions between two main ethnic groups  – the Konkombas and the Dagombas. Historically, the Konkombas in this district were considered the minority ethnic group with less access to education over the last two decades, particularly because of the ethnic conflict which erupted in the early 1990s. The other ethnic groups (Guan, Akan, Ewe, Maimbe, etc.) constituted only 3.1% of the population.

Livelihood patterns Farming patterns also had a significant impact on children’s participation and retention in formal primary and secondary schooling in the zones with endemic poverty. Subsistence crop farming and animal rearing were the main sources of livelihood in each of the study districts in Gushegu and Pusiga Districts, with some variations. Families found farming a challenge because of the continuous decline in crop yields over the years, erratic rainfall patterns and the high cost of labour. For these reasons, some families (particularly men) had resorted to migrating south to offer their services as farm labour, commonly known as ‘by day’ (our field survey in 2017). The findings from our fieldwork in the Upper East region reveal livelihood practices and subsistence farming such as millet, maize and soya crops along with dry season vegetable farming. There was clearly a correlation between these farming patterns and school attendance, as some children dropped out of school because they could not get food to eat. Another livelihood factor which affected OOSC was that children in these districts were engaged in several income-generating activities for the family, including trading across borders and in the nearby towns. Also, since the weaving of fabrics for traditional smocks is a major source of livelihood for women in some communities in the Pusiga District, the care of their children was left in the hands of ‘grandparents’ or elderly caregivers who were not able to cultivate the land to produce enough food to feed the family.

Experiences of primary and complementary basic education We turn now to the different perceptions which parents and children in Pusiga and Gushegu Districts had regarding their experience of CBE in comparison with basic primary education. We first outline the educational experiences of

Complementary basic education in Ghana 191 their communities, school management committees, parents, CBE learners and other stakeholders, before focusing on their experiences of CBE and primary school systems in terms of their quality, methodology, peer cultures and learning outcomes.

Community experiences of education Access to education in several communities has been a serious challenge over the last 20 years and is well documented (Associates for Change 2003, 2011). For instance, in 2015, approximately 20% of communities in the 25 districts where the CBE programme was operating in years 1 and 213 had no access to formal primary schools (CBE 2015). According to the 2010 Ghana Population and Housing Census (GPHC), less than 6% of the total population in the Gushegu District had attended school in the past or were still in school. The Ghana census also revealed that over 71% of the population there had never been to school, with girls less likely to have attended school in their district in the past. It also found that, of the district population aged 11 years and older, 71.9% were illiterate in English and a Ghanaian language, 20.3% were literate only in English and 7.5% were literate only in a Ghanaian language. Two out of the five communities in the Pusiga District did not have a local formal primary school. In PC3, one of the two communities without formal primary schools, pupils had to walk about 3 km to the nearby community to participate in primary education. In some communities, adults displayed a growing sense of urgency to provide their children with access to primary education. In three of the five communities visited in the Pusiga District (UER), community members had built their own schools and were paying their own community-volunteer teachers to ensure that lower primary education (KG to P2 or P3) was provided to very young children. These three out of the five communities studied in Pusiga District (e.g. PC1, PC2 and PC3) designed their own structures to serve as classrooms for their children. Their efforts were mainly led by their chiefs and CBE facilitators; in the case of PC2 community, the chief made great efforts to mobilise the population to pay a small stipend for the volunteer teachers (approximately GHC 40.00 every month). Another community (PC4) had a formal government primary school, but the parents did not trust that the school could provide their wards with quality education; so they resorted to sending their children to low-cost private schools in the district capital. Members in several other communities in the Northern and Upper East regions showed concern for their children’s academic performance and the poor quality of education found in formal public schools. The chief and community opinion leaders in focus group discussions in Pusiga District had this to say: ‘When you choose a head teacher who is not functional, work cannot go on’. ‘The GES directors are killing the schools by posting unserious head teachers to manage the schools’ (PC4 Pusiga District). This concern was echoed in several communities across the Northern and Upper East regions. Parents and other key stakeholders including chiefs and elders over years of observation had noted that, although primary school pupils started with the

192  Leslie Casely-Hayford et al. English language from the very onset, they struggled to acquire competency in speaking, reading and writing. In contrast, they pointed to how the use of the local language as the language of instruction in CBE classes had enabled the children to acquire proficiency in speaking, reading and writing (in that language) which, in turn, helped them to learn the ‘English language faster than their class mates when they transitioned into the primary school’. CBE learners’ reported that the usage of the local language in CBE facilitated their ease in learning to read and built their confidence when in primary school (after transitioning); this was also attributed to the CBE teaching methods, curricular content and the CBE facilitators’ commitment (a point we return to later). In the Gushegu District, all respondents believed that the use of the local language, coupled with the availability of teaching/learning materials for each learner made learning to read and write in nine months at the CBE centre easier and more effective compared with formal schooling. In contrast, they thought that the fact that primary schools used the English language with little or no access to teaching and learning materials, and the fact that the teachers were from outside the community and not regularly at the primary school made them less effective. The phonic and syllabic methods, the learner-centred approach, communitybased facilitators and the context-specific learning materials used in the CBE class were thought to enable the children not just to read and write in the local language but to apply their knowledge to other spheres of life. The smaller CBE class size (a maximum of 25 in a class as opposed to over 50 children per class in the primary school), having a full set of primers which children were given and which they could take home, and the elimination of corporal punishment in the CBE class made it possible for the facilitator to provide individual attention to the learners in a caring and empathetic manner. This, parents believed, often led to better performance among CBE learners in both literacy and numeracy. Some went further suggesting that those children who had entered CBE were afterwards more easily able to learn the English language at primary school, which helped them perform across all their subject areas. Consequently, some communities made sure that their children first entered CBE classes before moving to primary schools in order that they had a firm grounding, particularly in reading. Comments in the SMC group in the Gushegu District (NR) included the following: CBE learners grasp a concept faster because it is taught in Dagbani compared to the primary school . . . Those who pass through CBE understand Dagbani better and tend to do better in JHS and even in SHS than in formal school. (SMC participants in GC5 Gushegu District)

Those in CBE are more intelligent than those who go to primary directly . . . CBE graduates even help the primary school students to do their homework since they can read and write. (SMC focal group, GC5 Gushegu District)

Complementary basic education in Ghana 193 Nevertheless, all the stakeholders acknowledged that the two systems complemented each other. For instance, in the Upper East, the key stakeholders interviewed (including parents) repeatedly said they preferred ‘both systems’ since they needed CBE to prepare the children for the formal education system. Similarly, children in this region said that they preferred CBE to learn to read and at the same time they ‘liked the formal primary system since it allows them to go higher’. For instance, in one Pusiga community (PC1), a parent said he preferred both the CBE and formal school: ‘I will not say one approach is better than the other. I  will go for both approaches to education since the CBE will help the children to catch up in the formal primary school’. These community views are reflected amongst CBE learners. For example, in that community one such learner said ‘I like both the CBE and the primary school. I will be able to learn English at the primary school and learn my mother tongue in the CBE class’. While another CBE learner in the PC2 community in the Pusiga District stated, ‘I will choose the CBE because when I finish the CBE, I will be able to go to the primary school. The facilitators told us that when we finish the CBE we can join the primary school’.

Pedagogic and disciplinary approaches The stories in CBE are great . . . it makes learning fun. (Child from GC2 Gushegu District)

What emerged from these discussions were that there was a range of concerns about primary schooling that included: the language of instruction, literacy acquisition, the flexibility/relevancy of the school system, the methodological approaches to teaching and learning (especially literacy), access to teaching and learning materials (TLMs) as well as the degree of teacher commitment. In primary schools, the syllabus/content and the methodological approaches are generic in nature and not focused on the specific environment, culture and traditions of the community. In contrast, in the CBE class, the content is contextrelevant often bringing concrete examples from the community and is focused on the holistic development of the child in relation to language acquisition, culture and traditions. The CBE primers are developed to respond to the socio-cultural environment of the learners, thus facilitating learning, raising the child’s selfconfidence and supporting their language identity. Interviews with CBE learners revealed the importance of songs and play in the classroom whilst parents focused on the moral values, self-discipline and respect for elders created by the CBE programme. They often reported that CBE children were well behaved and the other children in the primary school copied this good behaviour. Similar perspectives were shared in a number of communities, where parents indicated that ‘(T)he content and methodology in CBE is what made some of us send our children to the CBE programme. It is a good programme, it is improving the mental capacity of the children’. Gushegu District parents reported that the

194  Leslie Casely-Hayford et al. facilitators at the CBE class were thought to use a pedagogic methodology that addressed the needs of learners, which was not the case in primary schools, which led to CBE learners being ‘more serious with their books’. When the CBE learners return from class, they loved to read their books. This, they said, is because the facilitators have made learning interesting and fun for the children. ‘The methodology of teaching in CBE classes makes the children learn faster especially since they learn the sounds through play, i.e. writing in the air, on the ground and before the black board’. This method, parents claimed helped their children to learn better compared with the formal school method. One parent in this district, expressed her appreciation of the teaching methods used in the CBE class which she commented: [T]he method of teaching is good. The last time I visited them in class I was happy to see how practical it is, and all the children were very enthusiastic about it. It is a pre-school kind of thing for especially children who have never been to school. (Parent, GC3 Gushegu District) Even parents who have never participated in the CBE in the Pusiga District indicated that, from what they had observed, they preferred the CBE approach and pedagogy. Most communities seemed to be struggling with the poor attendance of primary school teachers and sometimes compared this with the higher commitment levels and regular attendance of CBE facilitators. Most parents we interviewed rated highly the commitment levels and teaching practice of CBE facilitators who were from the community noting that the latter visited parent’s/children’s homes to find out why children were absent from class: [CBE] facilitators’ attention is always on children which enables them to do the right thing and what is expected of them. Officials of CBE come from far and near to visit the children, which the children always come to talk about it at home, and they are always happy. This tells we the parents that they have the children at heart. (Female parent of CBE learner in the focus group, PC2 Pusiga District) I have observed that, the CBE facilitator makes the class very interesting for the children. So when the children come home, you see that they take their books to read. I also went to see what goes on at the class and I realised that, whenever the teacher asks a question the children are able to answer. (Female parent of CBE learner in the focus group, PC2 Pusiga District) CBE learners in this district resonated with these views: I will choose the CBE because they teach better than in the primary school. At the CBE class, there is something on the board for us to learn when the

Complementary basic education in Ghana 195 facilitator is not yet in class. But with the primary school, when the teacher is not in class, they play around. (Female CBE learner transitioning to primary school, PC3 Pusiga District) They teach vowels and consonants. The facilitator always visits the children whenever they are not in class to find out what the problem is. The CBE distributes free books and pencils to the learners, and there is no payment of fees in the CBE. The CBE teaching is very effective and it enabled us to understand things better. (Male CBE learner, PC1 Pusiga District) All the groups we interviewed in these two northern districts acknowledged that a key difference between the two systems (CBE and formal education) related to the disciplinary approach used by facilitators/teachers in the classroom. Major differences in disciplinary approaches had been observed by parents and children themselves and by local committees at schools and in CBE classes; parents were keenly aware of such differences because of the feedback from their children. A key contrast was the manner in which discipline was executed in the primary school with a heavy reliance on the use of corporal punishment. Some of the CBE learners stated that they did not like the formal primary school because their friends in the primary schools have told them ‘their teachers do not teach them well and are fond of caning them’. This was in sharp contrast to the child-friendly and alternative disciplinary approaches used by CBE facilitators to ensure class management. Several children explained that one of the key reasons for their dropping out of the primary school was verbal and physical abuse by the teachers when they were late or could not answer a question. Also, in terms of disciplinary practice, in CBE classes children could wear what they had available, but in the primary school the children had to wear prescribed uniforms. They appeared neater in their uniforms and knew about personal hygiene better than those who go to CBE. Finally, in the Northern Region, most communities were also alert to the poor attendance of primary school teachers and sometimes those we interviewed compared this with the higher commitment levels and regular attendance of CBE facilitators. Parents argued that they preferred CBE teaching because the facilitators ‘teach with patience’ which was not the case, they felt, in the primary school. They attributed this difference to the larger class size in the primary school, thus leaving teachers in large classes with no time to attend to each pupil. Also, there is always one teacher in CBE classes while in the primary school there are more teachers and more classes. Various views not just on the effects of class size but also on restrictions that affected the quality and effectiveness of teaching were noted in the interviews with chiefs, elders and SMC and local committee members across communities in the Gushegu District: In spite of having a primary school, some parents want CBE because they believe that those who go through CBE to formal school have the upper

196  Leslie Casely-Hayford et al. hand over the others, especially in mathematics and Dagbani. . . . CBE class sizes are manageable compared to large class sizes in formal schools. (GC4 Gushegu district) Primary school has well-trained teachers as compared to the CBE class. However, the primary school performance has not been good due to attitudinal and resource constraints faced by the teachers while the opposite is true in CBE where the children receive books for free. (GC3 Gushegu district)

Linking decision-making and schooling choices Decisions about whether to send a child to CBE were clearly based on the families’ financial ‘strength’, role of the child in farming and at home along with the livelihood conditions of the family; some families could not afford to send children to primary school and they had been left at home/out of school for long periods in order to care for the animals or work on the farm. One parent stated that ‘if all the children are enrolled, their farming activities will be affected and the family cannot survive’. Where the children are many, the parents, particularly the fathers, decide to enrol either one or two children in formal school depending on how many they can predict they can support with the available farm produce and the demand for using the other children on the farm to support the family. There were also other circumstances where young girls had given birth in the cities and returned home, and their parents had refused to help these teenage mothers educate their children because they were not happy that they had become pregnant, placing an additional burden to the extended family. Several instances were found in the Upper East where young teenage mothers were using CBE as the only alternative to enrolling their children in some form of education. In contrast, a number of positive reasons were given by parents, community elders and women’s groups to explain why families opted to send their children to CBE classes compared to formal primary school. The majority view shared by the parents, learners and other stakeholders’ groups focused on the fact that (i) CBE was not just ‘high quality’ and ‘fee free’ – but the parents were also not required to pay for textbooks or exercise books, and (ii) the flexible nature of the programme regarding the timing of the classes meant that their children could also work on the farm and attend CBE in the afternoon. These aspects had both general and specific financial and educational benefits as this parent in the Gushegu District observed: Sustaining the CBE programme in the community will help parents who cannot afford formal school to enrol children in CBE. (Parent, GC2 Gushegu District) Most parents we interviewed were happy that they did not have any financial pressure associated with buying any education-related items for the CBE programme, including uniforms or shoes, PTA dues or examination fees. There was

Complementary basic education in Ghana 197 no required uniform, and children were supplied with free teaching and learning materials, including textbooks, pens and notebooks. Children also remarked that the fee-free nature and the fact that all the necessary items were provided for them was one of the most important factors in helping their parents opt for and prefer CBE. A focus group discussion in one of the Gushegu communities (GC1) with some girls in primary school who had never attended CBE again revealed how important it was to have free textbooks for all: Anytime we close from school, some of us always go to stand by the window to see what they are doing in the CBE class. And we always observed that in the CBE class, the facilitator has time for every learner as compared to the primary school. In our school, anytime we have reading the teacher always groups us to share one book, but we realize the CBE class is not like that because every child has his or her own text book to read. And because of that if they integrate into the primary school, they perform even better than some of us. Our current class prefect is a CBE integrate. (Notes from a girls’ focus group, GC1 Gushegu District) An uncle in the Gushegu District summed up the benefits: I opted for CBE because of free fees, books, etc. The facilitator teaches well. We don’t buy anything. There is maturity in the child and it is a flexible system – no uniforms, and it is in the afternoon so you can get the child to support you in the morning. The child is performing well, and I get him to support me in the morning. The child will be literate. The facilitator is regular and punctual. The facilitator takes good care of the children. My niece is behaving maturely after starting CBE. Everything is free. (Notes from interview with uncle of CBE learner GC3, Gushegu District) In addition, the absence of a school in the community or long distance to the nearest school was relevant to choosing between CBE and primary school. For communities without schools, CBE becomes the obvious choice for the families to opt for until their child was old enough to walk the long distance to a primary school and manage the associated risks (particularly for girls). Other parents noted that most of the children’s minds were not tuned to school, but the CBE class made it flexible for them to adapt to the system and finally adjust to the formal school system. The children are given a certificate after completing CBE which motivates the learner to ‘learn hard’. A parent noted that: I know school is about acquiring certificates so if at an early age you have a certificate, it is good. That is one thing I like about CBE. They also take good care of the learners. They provide them with all the learning materials. The facilitator is also good and up to the task – he is regular and punctual. CBE teachers and learners learn about our culture which I like very much. (Parent, PC2 Pusiga District)

198  Leslie Casely-Hayford et al.

Decision-making within families Interviews from the communities and across regions revealed that parents, both men and women, were making choices in relation to CBE and primary school education for their children depending on the structure of the family, economic circumstances and gendered culture. For instance, in some communities in Pusiga District, where the level of OOSC is the lowest in the study, mostly mothers – widows and female-headed households – were the ones making the decision of who went to school or CBE. While in the strongly patriarchal culture in the Gushegu District (with the highest level of OOSC) where out-migration was much less compared to the Upper East Region and where the fathers were the visible and active heads of households, the choice to send a child to CBE or primary school was primarily made by them (sometimes in consultation with their wives. In a polygamous family in which the number of children in a family could be up to 12, the father has to select which child from each wife would go to primary school, as well as a few other children who would likely join CBE. Those who joined the CBE are mostly the children who can help the family on the farm. Views shared by the chief and elders in the Gushegu District revealed the following reasons: Decisions as to sending children to school are made in the community by the fathers. Fathers usually exempt their first male children to assist them on the farm . . . depending also on the number of sisters and number of female children, the father reserves them for helping his sister, which affect the child’s chances of going to school. (Noted in focus group discussions with chief and elders, GC3, Gushegu District) Interviews with CBE facilitators and head teachers also point to the fact that family size has an influence on the decision of whether a child will go to primary school or the CBE. The facilitators were of the view that most of the children who attended the CBE class were from large polygamous families and that their parents could not cater for all their educational needs. For instance, in one of the Gushegu communities (GC1) in the Northern Region, the head teacher believed that parents in the community were doing well economically because the community members were engaged in various trading activities, so they were capable of sending all their children to the formal school. But the interest of families, especially the fathers, is that their children take care of their animals or take part in their trading activities. Because of this, the participation of children, particularly boys, in formal basic education was very poor. Interviews with teachers at the school confirmed this. In contrast, interviews explaining who decides and under what circumstances revealed a range of decision-makers in the Pusiga District in the Upper East Region. In some cases, the children themselves or other primary caregivers

Complementary basic education in Ghana 199 responsible for the child – including relatives (aunts and uncles) and foster parents – made the choice of sending the children to primary school or CBE. The findings suggest the fathers, mothers and aunts who are often responsible for educating the girls in the family and young people themselves are involved in the decision-making: It is their father who decided to send some of the children to the CBE class, and his decision is based on finance. (Mother of CBE child in PC1 Pusiga District) My auntie decided I should join the CBE. She says it will help me to acquire knowledge and go to primary school. (Girl interviewed in the PC1 Pusiga District) My mother decided to send me to the CBE class. But it was because I wanted to go to school. Because she could not afford the primary school, I told her I will join the CBE. (CBE boy interviewed in PC1 Pusiga PC1 District) The children, especially those who have never been to primary school, in some instances, decided for themselves to join the CBE, with the consent of their parent. I decided to join the CBE by myself because I know in the CBE class I will get books to learn. (Boy in PC1 Pusiga District) I decided to join the CBE so I told my father that I was interested and he allowed me to join. (CBE girl in PC3 Pusiga District) Based on tradition, the father is the authority of the household and main decision-maker regarding the daughters’ education, although he might consult with his wives since she has responsibility in the household for them. Interviews with SMCs suggest that there is a preference towards sending the son to CBE and/or primary schooling since, in the long term, parents believed that their daughter ‘might lose interest when she is given away for marriage’. The long-term vision for girls in the family is that they will eventually be married ‘off’, and in the case of Northern girls, this often happens between the ages of 13 and 16 years – which is the same time as CBE enrolment. Girls are still seen to be the ‘property’ of the man who marries her, and the family may still not believe that she will directly benefit them in the long run since she will be under the authority of the man and his family. An observation that was made is that, if a woman is the head of household, then she prefers to send the girl to the CBE class so that she can continue to assist

200  Leslie Casely-Hayford et al. her in the household. The parents explained that this is because ‘it is much more costly to educate the girl child compared to the boy child. The girls usually have more needs to be taken care of compared to the boys’. Parents also believed that it was risky to invest in girls’ education because of the uncertainty in the outcome. They were of the view that the girl ‘can easily get pregnant and her education will be stopped’. In contrast, they believed the boy child should be given the opportunity for education since the ‘girls will marry and leave the house, but the boys will remain in the house’ and one day inherit (male parental interview, PC4, Pusiga District). Another key tension is between the older (12–16) and younger children (8–11) who vie for family resources in education. The older children are often given the preference to enrol in school, and this affects the younger children’s opportunity to enter formal school. As one CBE child reported: ‘my mother lost her husband so she decided that the eldest son needs to farm to support the family but could not enrol all of us in school’ (CBE child interviewed in Gushegu District). In these cases, CBE offers a very important alternative route for the younger children in the family. Choices to enrol in the CBE programme were sometimes influenced by the CBE facilitator who went around the community to discuss with families the need to send their OOSC to CBE. Another influence of the parents and OOSC in the community was their peers who had attended CBE since they could see the benefits of participating in CBE. This visible benefit included the CBE child’s ability to read and the increased determination of the child to study at night and then enrol in the formal system after the nine-month programme. Some children also stated that their parents told them they don’t have money to take them to the primary school so that in the near future when ‘they get money’, then they will join the formal primary school. Although these children were faced with uncertainty as to whether they will ever join the primary school after they completed the CBE cycle, they were ready to be integrated into formal school. One mother in the Gushegu District spoke, like others, of her commitment to get her son into CBE and onto primary school: Knowing that the child will transition into formal school, I have started saving to buy the school uniform and the basic needs when he integrates into the formal school. (Mother’s view of CBE, GC2 Gushegu District) However, it was not easy to sustain children in formal education for some families: I decided that he should go for the CBE class because he dropped out from P3 and didn’t want to go back to the primary school. My son was sacked from school because of school fees, which I could not afford to pay in time, and by the time I had the money, he refused to go back to school. (Mother of a CBE learner, GC3 Gushegu District)

Complementary basic education in Ghana 201

Reflections and implications High rates of social inequality and extreme poverty continue to characterise northern Ghanaian communities and families continuing to balance the need to maintain the food security at the family level with sending only some of their children to school (Casely-Hayford et  al. 2009). Our study found in both the Gushegu District in the NR with a high incidence of OOSC children and in Pusiga District in the UER with its relatively low incidence of OOSC, that communities found it very difficult to send all the children to school and sustain them in primary school. The high levels of food insecurity and their reliance on children to work and support them at the family farm and in the home, made CBE a much more attractive option for parents compared to formal primary schooling. This was coupled with the fact that parents could not afford any extra direct and indirect educational costs in providing for the educational needs of their children. This meant that CBE was the parental preference for marginalised OOSC since it was fully free, required no uniform, and books were provided. Also, the pedagogic environment was seen by parents and children as more conducive to learning and more fun in helping them learn to read and acquire the basic knowledge to equip them for primary school. The adaptability of CBE to support the families within a complex web of gender inequality within the family, polygamous relations and patriarchy ensured that girls and boys who were not selected for primary school or the last to attend in the family could negotiate their way to CBE since it was in the afternoon; often after they completed their major chores on the farm and in the household. The study therefore showed that in areas with a high and low incidence of OOSC, CBE was more adaptable and responsive to the needs of marginalised children within the family and communities experiencing extreme poverty; this was an important finding particularly for girls in these areas because of some critical negative socio-cultural patterns and beliefs. Complementary Basic Education is clearly seen as an important educational pathway for children in the sampled communities in the two regions of northern Ghana. Parents and learners themselves presented evidence that the programme was of high quality in teaching children to read, write and numerate, mainly because of the emphasis on phonic and syllabic approaches to literacy along with the high commitment levels of the facilitator. Most of the stakeholders acknowledged that the methodology adopted in the CBE programme builds on the children’s known environment, uses participatory mother tongue approaches to teaching and learning, provides uninterrupted/individual access to primers/textbooks and promotes child-centred learning. These approaches were effective in helping children between 8 and 14 years of age skip three years of primary schooling in formal systems, thereby reducing the cost of educating children. Our previous research has suggested that CBE helped the children to increase their reading and learning outcomes in the CBE classes which in turn, was improving their outcomes in the formal school system (Casely-Hayford & Ghartey 2008).

202  Leslie Casely-Hayford et al. The absence of primary schools in some communities was a key factor contributing to the inability of families to enrol their children in school. Though government has a policy of children not travelling more than 3  km to access primary school, some districts were still experiencing a large number of communities without schools within a 3- to 5-km radius. Our research suggests that CBE should continue to be financed by government as a key priority in such communities and be used as a basis for creating community schools in order to ensure all children access to basic education. Intensive advocacy for the continuation of the CBE programme in these districts and communities along with the establishment of wing or annex schools to facilitate the transition of the learners into the formal school system needs to be pursued by parents, community leaders and NGOs in northern Ghana. Our interviews with parents also suggest that parents and some communities are losing faith in the public primary school system as they believe it is not providing the children with good quality teaching. Improving the quality of teaching and learning outcomes in formal schools was a high priority for them. Most communities we visited complained about poor teacher attendance and shortage of trained teachers in their schools, particularly in the Northern and Upper East regions. We found that others, particularly single mothers and widows, were keenly aware of the long-term costs of primary education and the financial responsibilities this would demand. CBE provided them with an option which was shorter-term and less risky in terms of their ability to cope with education-related payments. The decision to enrol a child in an educational pathway such as CBE was also based on the desire to balance the family’s immediate farming and food security needs with educational options which did not interfere with their survival as a family. Parents realised that their livelihood from farming was critical to their survival and that their children also needed basic education which was flexible and relevant in order to escape the poverty cycle. They also recognised that CBE provided them with an approach which would assist children’s learning of and ‘respect’ for their language, culture and traditions, which are particularly important in CBE. They noticed certain attitudinal and behaviour changes among their children who attended CBE such as respect for elders and more willingness to assist them on their farms. All the communities in both regions spoke of the need to continue the CBE programme because of the endemic poverty levels, the persistence of OOSC and the continued household constraints to send all their children to school – particularly those between 8 and 15 years of age. When we asked them about their ability to sustain the CBE programme once the DFID and USAID completed its financing we found that communities were willing to support the small stipend or “soap money” provided to their community-based CBE facilitators, whilst indicating that they would still need to have the training, teaching learning materials, books and some minimal supervision provided by the Ghana Education Service or NGO partners to sustain the programme. Some communities in the Northern and Upper East regions had started their own community schools since trained

Complementary basic education in Ghana 203 teachers continued to refuse postings to these areas but these were targeted at younger children aged 5–8 years and often supported by volunteer CBE facilitators acting as the teacher. This showed the resilience and determination of communities in Ghana’s extreme poverty zones to find answers to the ongoing lack of basic education in their area. CBE was able to adapt to the farming and socio-cultural dynamics restricting child educational participation by ensuring that no barriers which would normally deter parents (cost of uniform, gender imbalances, timing of the school and length of the school day) persisted. Our data suggests that universal primary education in such poverty zones in Ghana will only be realised when education is more flexible, relevant and adapted to addressing the current realities of extreme poverty; and when families are not dependant on children for their subsistence farming, food security and livelihoods; the demand for child work on family farms will be reduced and possibly eliminated if and when new innovations and farming techniques are introduced. Most households were reliant on subsistence farming and rain-fed agriculture for survival along with traditional farming patterns which required high levels of labour. This was particularly the case for the Upper East and Northern regions where outmoded farming practices are preventing children from attaining basic education. Our study concluded that, despite significant efforts by government to ensure fee-free education in Ghana over the last decade (e.g. free school uniforms, no user text-book fees, school feeding and social safety net cash transfers), there are three major deterrents for all children being enrolled in primary school. The deterrents were long-standing documented problems across the northern regions of Ghana and included: poverty/food insecurity of the households; large family size and negative socio-cultural practices particularly in relation to girls’ education (e.g. fosterage and early marriage). Despite government interventions, flexible complementary basic education provision for these most remote areas remains a high priority to address the poverty and socio-cultural constraints to education. These findings were confirmed to be the demographic make-up of families which exhibited internal barriers such as polygamy, large family size and competition for resources because of inequitable treatment by husband in not educating more than one child per wife. Our research concludes that persistent negative socio-cultural practices which are preventing the progress of girls’ education should be addressed. Strengthening government policies on the care and protection of the poor and of vulnerable children along with educating the population about outmoded socio-cultural practices (e.g. early and forced marriage) can be addressed through working with social welfare, NGO and human rights agencies in Ghana.

Notes 1 The authors would like to thank all the families, communities and children who shared their views/insights with the research team into the challenges and advances of reaching their educational pathways in the context of often harsh

204  Leslie Casely-Hayford et al. socio-economic and strong cultural tradition. Special thanks also goes to Professor Madeleine Arnot who assisted the authors in finalising this chapter for publication. 2 This study was a meta-analysis of literature and data on OOSC children barriers and bottlenecks using the UNICEF framework for analysis and conducted by Harvard University and Associates for Change (UNICEF 2012). 3 The Complementary Basic Education Management Unit is under the Basic Education Division of the Ghana Education Service (GES) and is responsible for the implementation of CBE across the country. 4 For more information on the research study see Associates for Change (2011) (www.associatesforchange.org). 5 Each of the three authors led the month-long research fieldwork across one of the three research sites/districts in each region. 6 Communities were selected on the following basis: 70% of communities with a formal school in the community and 30% of communities where the formal school is beyond 3–5 km from the community. 7 This is the cycle in which these data were last collected. 8 In Gushegu (Northern Region) 22 (44%) fathers and 28 (56%) mothers were interviewed, while in the Pusiga District (Upper East Region) interviews were conducted with 60 women to 40 men reflecting the significant role women were playing in relation to the education of their children. 9 The CBE learners, the integrated CBE learners and the formal school pupils who were interviewed but never participated in the CBE programme were between 8 and 15 years of age. 10 All interviews were conducted using gender-sensitive approaches, often with men and women separated during focus group interviews in order to facilitate participation based on cultural norms. 11 Households in the Northern Region often include more than one family unit. 12 African traditionalist religions are those which are specific to ethnic groups such as the Dagomba and the Konkomba. 13 Districts refer to the 25 districts across the Northern, Upper East and Upper West regions.

References Associates for Change (2003) Reaching Underserved Populations with Basic Education in Deprived Areas of Ghana: Emerging Good Practices. [Online] Atlanta: CARE. Available from: www.associatesforchange.org. Associates for Change (2011) Inclusive Education in Ghana: A Look at Policy and Practice in Northern Ghana. [Online] Available from: www.associatesforchange.org. Associates for Change (2013) The Quality of Education in Northern Ghana: Assessing Learning Efficiency and Effectiveness Across Ghana’s Three Northern Regions. VSO/ Comic Relief Commissioned Study. [Online] Available from: www.associatesforchange.org Associates for Change (2017) Stories of Most Signification Change in Child Protection across Ghana’s Northern Regions. Accra, Ghana: UNICEF. Casely-Hayford, L. (2019) Annual Education Budget Analysis to Parliament of Ghana. Accra, Ghana: Associates for Change. Casely-Hayford, L., & Ghartey, A. (2008) The Leap to Literacy and Life Change in Northern Ghana: The Impact Assessment of School for Life. [Online] Available from: www.associatesforchange.org.

Complementary basic education in Ghana 205 Casely-Hayford, L., Ghartey, A., & Agyei-Quartey, J. (2018) Parental and Learner Choice in Basic Education Across Ghana’s Three Northern Regions. Accra: Crown Agents/Associates for Change. Casely-Hayford, L., & Hartwell, A. (2010) Reaching underserved populations with complementary education: Lessons from Ghana’s state and non-state sector. Journal for Development in Practice, June, pp. 527–539. Casely-Hayford, L., Salifu, E., & Arnot, M. (2009) The Education Outcomes Gap in Ghana: Young People’s Insights into Schooling, Social Dislocation and Poverty. Paper presented at the UK International Education Conference, University of Oxford, September 11–13. CBE Management Unit (2015) End of Cycle Report on the Complementary Basic Education Programme. Accra, Ghana: Crown Agents/Associates for Change. CBE Management Unit (2016) Updated Analysis of Out of School Children in Ghana with District Level Projections. Accra, Ghana: Crown Agents/Associates for Change. Farrell, J.P.,  & Hartwell, A. (2008) Planning for Successful Alternative Schooling: A Possible Route to Education for All. Paris: UNESCO IIEP. Ghana Statistical Service (GSS) (2010) Ghana Population and Housing Census (GPHC). Accra, Ghana: Ghana Statistical Service. Ghana Statistical Service (GSS) (2016) Ghana Living Standards Survey. Accra, Ghana: Ghana Statistical Service. IMC Worldwide (2018) Cycle 4-Tracker Study of Complementary Basic Education Students. Commissioned Study. London: DFID. Kantur Consulting (2018) Final Evaluation of Ghana CBE Programme. Commissioned Study. London: DFID. Ministry of Education (2013) Complementary Basic Education Policy Framework. Accra, Ghana: Ministry of Education. UNICEF (2012) Global School Initiative on Out of School Children: Ghana Country Study. [Online] Available from: www.associatesforchange.org. UNESCO (2017) Accountability in Education: Meeting our Commitments. Global Education Monitoring Report 2017. Paris: UNESCO. World Bank (2015) Out of School Youth in Sub Saharan Africa: A Policy Perspective. Washington, DC: World Bank.

11 Addressing dilemmas of difference Teachers’ strategies to include children with disabilities in rural primary schools in India Nidhi Singal Undertaking educational reform with a focus on addressing the historical exclusion of children with disabilities has become imperative in recent years. The Sustainable Development Goal 4 (United Nations 2015) committed to ensuring ‘inclusive and equitable quality education and promote lifelong learning opportunities for all’, including those with disabilities. This is a marked shift given the invisibility of disability in the Millennium Development Goals (United Nations 2011). Whilst education of children with disabilities has only recently become centre stage in mainstream development, the Universal Declaration of Human Rights (United Nations 1948) had in fact already set the agenda for the right to education for all children, including those with disabilities. Subsequently, the United Nations Convention on the Rights of Persons with Disabilities (United Nations 2006) elaborated on this commitment in Article 24, noting that ‘state parties shall ensure inclusive education system at all level and provide free, inclusive, quality primary education’. UNESCO Institute for Statistics’ (2018) data show that learners with disabilities continue to be one of the population groups most likely to be excluded from formal schooling. Based on analysis of data from 49 countries, the report confirmed that: [P]ersons with disabilities are nearly always worse off than persons without disabilities: on average, the former are less likely to ever attend school, they are more likely to be out of school, they are less likely to complete primary or secondary education, they have fewer years of schooling, and they are less likely to possess basic literacy skills. (p. 3) Various estimates suggest that around a third of primary school-age children with disabilities are out of school globally (Mizunoya et  al. 2016). The continued exclusion of persons with disabilities from schooling has shown to have a profound negative impact on their future life opportunities, leading to increased inequities in access to employment, limited social networks and reduced access to health care and an inability to move out of poverty (Mitra et al. 2012; Hoogeveen 2005).

Addressing dilemmas of difference 207 Along with colleagues I have argued that the growing focus on educating children with disabilities internationally can be linked to three prominent themes (Singal et al. 2017): (1) the need to uphold human rights in education, which has a lasting impact on building cohesive societies; (2) an economic rationale which recognises that educating persons with disabilities has both significant individual gains and increased national productivity and, (3) a view that including children with disabilities in mainstream schools has the potential to bring about quality reforms in the larger education system. The last theme is particularly significant given the current ‘global learning crisis’ (UNESCO 2013) and the innumerable efforts to remedy this situation. In this chapter, my aim is to exemplify the tensions, challenges and opportunities accorded in the inclusion of children with disabilities in mainstream classrooms, particularly given the growing focus on inclusive education. I specifically focus on India (a policy-rich context) with the aim of understanding how teachers in mainstream government primary schools, who are central to making inclusion a reality, understand the changes in the country that have gained momentum over the last decade. The Indian government has taken important steps in promoting education for children with disabilities, particularly at the primary level. National education policies have at various times focused on issues of access for children with disabilities to both mainstream and special schools (Singal 2006). Significant impetus to these efforts was provided through the passing of the Right of Children to Free and Compulsory Education Act (Ministry of Human Resource Development 2009) and the Rights of the Persons with Disabilities Act (Ministry of Law and Justice 2016), both of which uphold the right to education for children/ young people with disabilities. In parallel, various national programmes such as the District Primary Education Programme and Sarva Shiksha Abhiyan (Ministry of Human Resource Development 2012) have included a focus on children with disabilities. A measure of success of these policy efforts is reflected in more than a fourfold increase in the number of children with disabilities accessing school, from 566,921 in 2002–03 to 2.35 million in 2012–13 (National University of Educational Planning and Administration 2014). Notably, of these, 1.64  million children were enrolled in primary and 0.70  million in upper primary classes. However, this also means that children with disabilities constituted only around 1 and less than 1% of the total number of children in primary and upper primary classes respectively. Furthermore, significant disparities in relation to gender and types of impairments remained (Singal et al. 2017). Despite such increases in access to schooling, research on the experiences of children with disabilities in Indian schools presents a negative picture, highlighting their continued exclusion from teaching and learning processes (Sawhney 2015) and negative social experiences (Das  & Kattumuri 2011). Reflected in these research findings is the continued neglect of process-based issues (such as teacher training, adaptation of curriculum and accessible teaching and learning materials) in programmes such as the Sarva Shiksha Abhiyan, which largely has

208  Nidhi Singal focused on addressing issues of access, such as identification, assessment and provision of aids and appliances (see Singal 2015). It is vital to acknowledge that concerns raised in relation to the poor quality of education received by children with disabilities are not very different from the growing body of research noting broader issues with the poor quality of teaching and learning received by most children, particularly those belonging to lower socio-economic groups (Batra 2015). However, scholars do not necessarily draw on these connections and neither does the government regard disability and inclusive education as an opportunity through which to bring about systemic reform of the overall education system (Singal 2019a). In this chapter, I examine critically how the educational interests of children with disabilities are enacted in practice in mainstream schools in India. In doing so, I draw on the dilemmas of difference framework (Minow 1990) as an analytical tool to highlight how it can be a powerful way of interrogating the policy–practice praxis in relation to the education of children with disabilities.

Dilemmas of difference Martha Minow, a US legal theorist, first referred to dilemma of difference in her book Making all the Difference (Minow 1990), which is based on contemporary American law. Here she argued that the dilemma of difference is the choice that we assume must be made between treating all people as indistinguishable – as if there are no differences – and acknowledging difference in order to accommodate special treatment of implicitly ‘special’ needs. Hence the dilemma of difference is raised in the following questions: When does treating people differently emphasize their differences and stigmatize or hinder them on that basis? And when does treating people the same become insensitive to their difference and likely to stigmatize or hinder them on that basis? (ibid, p. 20) The dilemma of difference offers two broad options: to recognise or not to recognise difference – however both options are associated with negative risks. Recognising difference can lead to different provision (such as special education which might be stigmatising and devaluing) but not recognising difference can lead to not providing adequately for individuality. Thus, tensions occur between the values of inclusion and individuality, as Braham Norwich (2002, p.  498) points out: The dilemmatic position does not look forward to some utopian future when rival values will be finally reconciled. These tensions are seen as inherent in our humanity and social existence. It looks rather to a future where tensions can be minimised while trying to fulfil multiple values. It is, therefore, a position which accepts that choice at whatever level can involve some loss and

Addressing dilemmas of difference 209 pain. It is a view which assumes the continued struggle with ambivalence in society and in personal life. It is a position that underpins human rights but accepts that different rights can come into conflict. Differences are addressed throughout education in relation to language, age, gender, ethnicity, caste and social class (leading to some expressions of this dilemma of difference in these areas). However, when we deal with issues of disability, these dilemmas are presented in a more accentuated form. In this chapter I draw extensively on the three main dilemmas which Norwich (2013) suggests are found specifically in relation to core questions within education, namely: (1) The identification dilemma: the issue here is whether to identify individual children as having a disability. This dilemma reflects more general tensions between recognising individuality while maintaining respect for the person. (2) The curriculum dilemma: what should children learn? Should there be a common curriculum and, if so, how can a common curriculum take account of different social and individual interests and needs? (3) The location dilemma: where should children learn and with whom? Should children learn in generic schools or in specialised schools and classes which focus on specific content (e.g. on music or on teaching methods which, for example, are designed for those with visual impairments). The dilemma also reflects general tensions between how to use scarce resources or how to make systems viable for economies of scale, a concern in a country like India. Norwich acknowledges that there is no easy, simple or single correct solution to these dilemmas. Each of the available alternatives carries some negative consequences and hence resolving dilemmas requires fulfilling some values without jeopardising other important values. There is no ideal resolution in these scenarios. Rather, the resolution reached depends on the balance and relative weight given to different values. Thus, there can be a range of different resolutions which can change with time. What is called for is tolerance to these different resolutions, whilst acknowledging that there can be no final answer. However, as Minow (1990) acknowledged, it is essential to become ‘immersed in the dilemma in order to gain a better understanding of it in order to look for a more productive struggle’ (p. 384). This deeper analysis helps create a better understanding of the decisions taken and can support positive desirable improvements in schools and broader systems. Minow (1990) argues that recognising dilemmas and their pervasiveness within education is itself a step forward. A dilemmic approach is equally applicable to the contemporary educational reform efforts in India. Here we have a society committed to equality and pluralism, despite its history of social hierarchy and discrimination towards those who are not part of the majority groups, for example, those from lower castes and those with disabilities. In such a context, the efforts to redress discrimination and bring about reforms in education raise important questions. In the following section I  first describe the focus of our

210  Nidhi Singal empirical study in Indian primary schools, analysing in particular how the teaching of children with disabilities was understood and acted upon by their teachers. Even though a dilemmatic perspective did not define the research design of our study, it emerged as a powerful framework in helping make better sense of teacher narratives during the analysis phase.

Research focus and overview The primary data discussed in this chapter draws on an empirical study which aimed to understand how efforts towards inclusive education were being implemented in rural mainstream school settings in the state of Karnataka, in the south-western region of India. It is a state which is performing well on various educational indicators. For example, the 2011 census noted its literacy rate as 75%, in comparison to the national average of 74%. Primary school enrolment for Karnataka was reported at 102.6 % in 2011, while the national average was 93% (Government of Karnataka n.d.). Even though the size of the private school sector is increasing in India, research has consistently shown that the poorest and the marginalised are more likely to access the government system (Srivastava et al. 2013), which held true for the three case study schools. In collaboration with a team of researchers,1 we conducted semi-structured interviews and observations in rural mainstream government schools in the district of Chamrajnagar. All three schools were rural with primarily first-generation learners and served children from low-income families. Each school had a concrete building with blackboards painted on the class wall (two of the three had desks and chairs for the children and their teacher). Each class had approximately 30 children who, in most cases, had textbooks and some notebooks – but there were very few additional resources, such as display charts. Interestingly, there was a well-equipped resource centre at the district level, filled with an impressive range of multi-sensory teaching and learning materials, but none of the teachers seemed to have accessed it, primarily because of their inability to travel the distance required (20 minutes by a motor vehicle). Over a period of four months, interviews were conducted with 12 primary class teachers and three head teachers, across the three primary schools and with six other para-professionals who were employed at the district level and were either designated as Inclusive Educational Resource Teachers (IERTs) or Home-Based Volunteers (HBVs). Additionally, 16  hours of classroom-based observational data were also gathered in the three schools. Finally, various informal visits were undertaken with the HBVs, when visiting homes of children with disabilities. This chapter draws specifically on data collected from the primary school teachers. The broader project included interviews with parents, which have been published separately (Singal 2016). Our interviews with teachers explored various themes, such as the respondent’s understanding of the needs and strengths of a child with disabilities; how these needs were identified; the kind of teaching practices used and so on. With

Addressing dilemmas of difference 211 school head teachers, the focus of the interviews was on how school-based support was extended to teachers and parents of children with disabilities and the role of IERTs employed by the state at the district level. Interviews were conducted in Kannada, the local language, which participants felt most comfortable with. Translated and transcribed transcripts of the interviews were uploaded to QSR-NVivo and a three-step coding approach developed by Neuman (2013) was followed. The data were first coded for conceptual categories, followed by identifying relationships between categories, before moving on to developing core themes across these different categories and stakeholder data. After this, a cross-case thematic analysis was undertaken to identify common and unique perspectives. Based on this analysis, a decision was made to employ the dilemmatic framework to capture more carefully the many tensions and challenges evident in stakeholder discourse.

Exploring the three dilemmas The decision to use the dilemma of difference framework was made after the analysis had been completed and provides only one reading of the data. However, using the three dilemmas listed by Norwich (2013) provides a significant opportunity to move beyond a tendency to seek resolutions towards acknowledging the complexity of efforts to achieve inclusive education. In the following sections, I discuss the three dilemmas as evident in teachers’ narratives and the efforts they were sometimes making to resolve these.

The identification dilemma Identification can be hugely problematic as, on one hand, identifying a group of children as different (‘othering’) can result in unnecessary labelling and stigmatisation. However, not identifying children’s needs can also result in these needs not being met. Given the limited opportunities for psychological testing in rural parts of India, the teachers we interviewed did not explicitly use the language of identification, and their notions of differentiation were vague, somewhat simplistic and fraught with tensions. In most instances, there was a tendency among teachers to subsume difference into a universal category of children, that is, by saying that ‘all children are the same’ whilst noting that ‘all children are different in one way or the other’. For example, a teacher suggesting the notion of sameness stated: [T]here are some things they [children with disabilities] can do, they are some things they can’t do . . . it is like the others [non-disabled children]. . . . Lalit is unable to run but participates in throwing the ring. They do what is convenient for them. (T, F, 17 years)2

212  Nidhi Singal Another teacher further elaborated: They (children with disabilities) can come easily to the school, and there isn’t any problem. There is no problem from teacher side, or student side, they are just below average students. (T, M, 11 years) While, on the one hand, teachers noted ‘sameness’, as the interviews progressed, they discussed at length how these children were very different from their peers. Teachers commonly stated that children with disabilities were different in terms of ‘not (being) able to learn things easily’, ‘are unable to write a lot’, ‘unable to learn things as much as others, can’t follow what is taught to others’. The notion of ‘inability to learn’ was further differentiated, according to type: If the boy has problem in his leg, it’s okay for them. They do not cause problem, intellectually they are well and clever. But we have some students who have a mental disorder, such as Subhash, he hasn’t learnt anything. He is unable to learn. (T, F, S2) Another teacher voiced similar thoughts when elaborating on the differences between children with and without disabilities: They are unable to learn like other students. Some students learn easily but not those with disabilities. I  give them the same materials and help them learn. They can’t improve drastically but they learn . . . for other children we have targets but not for them. We can’t have targets as they are unable to learn as easily as others. Other children can learn A-Z; these children can learn at least half of it, not the whole . . . I try to teach them half of the syllabus. If others teach 100%, I try to teach them 50%, if not then at least 25%. (T, F, 17 years) These commonly articulated perceptions about children with disabilities being unable to learn like others had real consequences in terms of social equality. They can, and as we see later, affect how teachers responded to the child’s learning needs. These perceptions informed the kinds of changes that teachers perceived as useful and/or were willing to make in the curriculum and the types of teaching strategies they were ready to adopt (or reject) in order to include children with disabilities.

The curriculum dilemma The struggle to interpret the characteristics of children with and without disabilities was compounded by the tensions experienced by teachers in these rural schools in relation to what exactly children with disabilities should learn and how

Addressing dilemmas of difference 213 they should be taught. Here normative values about what is in their best interests, what is ‘useful’ or ‘good’ were highlighted. Like other studies conducted in India (e.g. Sarangapani 2003; Vijaysimha 2013), teachers in our research viewed the prescribed official curriculum as fixed and something that could not be changed. Thus, while the teachers interviewed believed that the school curriculum was inherently of relevance to all children, they tended to argue that, because children with disabilities take longer to learn, they cannot be expected to cover it entirely. This resulted in teachers deciding to overlook parts of the syllabus for children with disabilities. As one teacher noted, ‘I only teach them 20% of the syllabus and I do not think it is useful to spend lot of time explaining just one thing’ (T, M, 12 years). Another theme which emerged was a tendency to lower expectations for children with disabilities. Most of the teachers rationalised that providing the same learning and curricular experiences for all children was inappropriate because ‘they cannot be expected to perform at the same level as others in the class’ (T, F, 15 years). Also, when teachers were requested to provide examples of lesson plans or any other kind of evaluation sheets that they were using particularly with children with disabilities, they had nothing to share. They also noted that they had no assessment strategies to monitor children’s progress. While the teachers we interviewed felt that there was an implicit expectation of the government that all children should have access to the same curriculum, they did not think this was reasonable given that some children were not equipped to learn at the same pace. Thus, to resolve this particular tension, teachers covered a reduced curriculum for children with disabilities, which meant that quite often they would be sitting in class, but the teacher would not expect them to follow the lesson. Also, since teachers perceived children with disabilities as slow in grasping concepts and as having limited learning ability, the common teaching approach adopted was to give them extra time to complete tasks: ‘I give him extra time, when children do four sums in the same time, I  let him do only one’ (T, F, 10 years). Additionally, during interviews, teachers reported that they provided individualised attention, which in practice meant that they made the child with disabilities sit next to the teacher’s desk. In other cases, they noted giving children with disabilities extra attention in class: ‘I will always make her sit with me, I give extra time and I teach her’; another noted: ‘Umesh’s case is different and I give him more importance and attention during class’, with the aim of including the child in the teaching. Thus, in such a scenario to resolve or address these tensions, teachers reframed and in effect lowered their expectations of children with disabilities. Other attempts at resolving this dilemma were evident when teachers highlighted making small modifications, for instance, in relation to seating arrangements in the classroom. Seating a child with disability closer to the front of the class was the most commonly described teaching practice, the assumption being that this would keep the child more focused on the lesson. Ironically, even in the head teacher interviews, this was the most identified strategy of support.

214  Nidhi Singal More commonly, across the sample, teachers noted that they needed extra support to meet diverse needs of students in their classrooms. Here the role of the IERTs was frequently raised in interviews. Teachers stated that it was important that they had the support from ‘experts’. As one teacher noted: The IERT we have attached to this school takes care of the children with disabilities. IERTs are getting equipment for these children like hearing aid, crutches and callipers . . . He visits the school once a fortnight or once in a month. He has already identified children who have disabilities and he also talks to their parents about equipment. (T, M, 16 years) Another teacher describing the role of the IERT stated: We maintain a file on each child with disabilities, and he looks through these files. He also talks to the child. Sometimes he also visits the child in each class. (T, M, 18 years) While the government policy positions the IERTs as supporting teachers to help learners, a teacher provided contrary evidence, when she observed: ‘He has never told us how to teach the children in the classroom’. Despite not receiving any support from the IERTs in terms of how to adapt their teaching practices, teachers relied on them to provide direct assistance to the child with disabilities. Even though teachers acknowledged that IERTs only made occasional visits to the school. For example, a teacher noted that nearly a month back a new child with hearing impairment had joined her class, but she was not able to do anything with him, as she was waiting for the IERT to visit the school and help the child. But, since these para-professionals were positioned as experts with the requisite know-how for educating children with disabilities, teachers largely relinquished themselves of the responsibility of educating children with disabilities. While the IERTs acknowledged that they had an important role of supporting teachers, they admitted that there were too few of them for the magnitude of the task at hand: There are approximately 286 government schools in the block and we [IERTs] are only three of us. We cannot visit all schools, but do so only 15–16 schools, once a month and have half a day visit in each school. (IERT-A) It was clear that all the teachers we interviewed felt that the curriculum offered to children with disabilities in special schools was more appropriate for their development and needs. For example: In special school, they will be with teachers who know how to teach them. Also, they will learn things that are more important for them . . . like how

Addressing dilemmas of difference 215 to talk to others, read braille, move around in society. Their parents cannot teach them these things, they are busy earning a living. (T, F, S2) Another teacher noted that the special school curriculum with its focus on daily living skills is more appropriate for children with disabilities. They can learn how to take care of their basic needs, like hygiene . . . in special schools they also teach vocational skills, like candle making, card making and so on . . . all that is so useful for these children. (T, F, S1)

The location dilemma The question of where and with whom children with disabilities could and should learn was possibly the most frequently discussed by all those interviewed. It is also the space where real tensions arose in what teachers felt they should be doing and the resources on offer to achieve desired objectives. While teachers were largely supportive of including children with disabilities, their discomfort in teaching them alongside other children was clearly apparent. Teachers noted tensions in relation to how giving extra time to one child raises concerns about the quality of learning experience of other children in the class: It’s very difficult to teach a disabled child and it will take more time. When we give extra time and extra care to such children then it creates problem for other children. (T, F,10 years) If we give them more attention, they would work but if we do that the rest of the class suffers. The lesson would never be complete if we keep concentrating on them . . . We have to manage these children and the others in the class. It is difficult. (T, F, 9 years) Another voiced concerns around safety of other children in the class: Those with disabilities don’t get what they need in mainstream classrooms – too noisy, pace is too fast. It is not suited for them and their behaviour is distracting and harmful to others. (T, M, 15 years) Given these concerns, it is not surprising that teachers provided alternative suggestions for where they thought children with disabilities could be taught better. A clear consensus emerged that special schools, which employed trained teachers

216  Nidhi Singal and equipped children with disabilities in basic skills, had an important role to play: In special school there will be a teacher especially for them and they would be taken care of. They will maintain some discipline for children with disabilities and also prepare them for things that they will need later. (T, M, 10 years) Like language  .  .  . in special school they can be taught in their language (using sign language) which will help them to understand easily. (T, F, 10 years) Being in a school with others like them will help them communicate better with each other. (T, F, 15 years) Whilst clearly struggling in practice, to varying degrees teachers argued that special schools might be a better place for educating children with disabilities. They were, however, acutely mindful of the stigma attached to special schools, and this was expressed in different ways. Teachers did not present a homogenised narrative of special schools as being better, rather they highlighted sophisticated dilemmatic perspectives in relation to location of education. This is captured powerfully in the following quotes: They come to mainstream school and they feel that I am also student of this school like others. If we send them to special school, he might have a feeling that I have some problem. But they come to mainstream school, they won’t get that feeling . . . if we have them in our class then they don’t feel they are disabled. They will mingle with all children. They will play with other children and feel that they also have to win praise from the teachers. Because when they come to school, they are able to have these feelings. They have enthusiasm to study. If there is only a group of disabled children then it’s not fair, he should mingle with others as well. They can happily play with other students. (T, F, 17 years) We have to improve them. It’s the responsibility of the teacher. At any cost without any inferiority complex we have to teach them. They have to mingle with other children. (T, M, 15 years) Another teacher reflected that, given the lack of any special schools in the vicinity, it was only natural that children with disabilities attend the mainstream local school: There is no special school for children with disabilities in this community. They come to mainstream school and they feel that I am also student of this

Addressing dilemmas of difference 217 school like other children. . . . Government also has this feeling what is why they gave money to provide ramps, special toilets for them in the school. They gave Rs. 50,000 for the special toilet. Like this so many benefits are being given by the government, it is better that they come to school here. (T, F, 10 years) The benefits of including children with disabilities into the mainstream were primarily seen in relation to socialisation opportunities. As one of the teachers summed up, ‘the (state) government has instructed us to allow them to mingle with other students, and then only they will improve’ [emphasis added]. The construction of mainstream school as a space where children learn to play and be social with each other meant that the goal of learning, specifically for children with disabilities, was relegated to the margins. As we have seen, the central role of specialist support in meeting the learning needs of children with disabilities was highlighted on the grounds that these specialists, such as IERTs, have better training, ‘Someone who can understand the child both emotionally and mentally, we need that’ (T, F, 16 years). One of the teachers went further on to argue that such specialist teachers should also provide after-school support to children at their home: A separate teacher should be appointed for them to go to home and teach them after school. We already spend some time with them in school, but it will help if someone can spend more time at home. Then there would be a chance to learn more- at home and school. (T, F, 9 years) Notably, this practice of asking children to get additional support after school hours is in many ways not very different from the practice of tuitions commonly followed across urban and rural India. With specific relation to children with autism, Johansson (2016) noted how private schools in Kolkata expected middleclass families to arrange home tuitions to help their child keep up with the school curriculum.

Concluding reflections In discussing the earlier findings and examining the potential of a dilemmatic perspective in helping shape future policies, it is important to begin by acknowledging the significant shifts evident at the micro level in relation to the education of children with disabilities. During the fieldwork, visits to schools, education offices and homes of children, the amplified focus on educating children with disabilities was clear. During these visits, head teachers shared the official forms that they were required to fill in, noting the number of children with different types of disabilities attending their school. During conversations with IERTs and teachers, there was ample evidence of pressure from the government to monitor and encourage increased enrolments. Many teachers described how before the start of the school year they were mandated to undertake door-to-door surveys

218  Nidhi Singal to identify all out-of-school children and were specifically required to encourage parents of children with disabilities to enrol them in school. Teachers also mentioned the greater ease in availability of wheelchairs and other assistive devices for children with disabilities. Teachers in this study were themselves sensitive to the increased focus on including children with disabilities and were mindful of not outrightly excluding them. This shift in teachers’ perceptions is also noted by other research studies in India. For example, findings from two different surveys, one conducted with 470 private secondary school teachers in New Delhi (Bhatnagar & Das 2013) and the second conducted with 560 government primary teachers in Gujarat (Shah et al. 2013), concluded that teachers largely held positive attitudes towards children with disabilities. However, in both cases they underlined important concerns about poor infrastructural resources in their schools and classrooms. Most significantly, teachers noted how they lacked essential knowledge about inclusive education and were unaware of how best to address the learning needs of children with disabilities. Thus, while teachers expressed support ‘in the theoretical possibilities of inclusion, they were still doubtful about the practical issues emerging from inclusion’ (Gafoor & Asaraf 2009, p. 11). Analysis of teacher interviews in this study suggests that they held highly deterministic and restricted views about children with disabilities. Teachers viewed differences as inherent and determined by biological factors, hence fixed rather than a result of interaction with the social and educational spaces that the children were located in. Teachers’ beliefs shaped their practices, wherein the focus was on fixing the child, such as: shifting the place they sit, reducing the number of tasks they were required to do, giving more time, asking an expert to teach them. None of these entailed reflecting on their own teaching practices and the possible mismatch between student learning needs and the adopted teaching style. These perspectives need to be understood in the wider context in which teachers were working. The realities of their classrooms were far from ideal, as is evident from the fact that they had limited teaching and learning resources, restricted physical space, pressures of covering vast amounts of syllabus and the challenges posed by most children being first-generation learners. Thus, the balance between pragmatism and the expectations arising from changing government policies, such as the Right to Education Act, was something that these teachers were continually struggling with. Importantly, one of the greatest challenges currently being faced is how to transform teacher practices rather than merely altering their perceptions towards children with disabilities. I have argued elsewhere that access is only the first step: guaranteeing a space in the classroom does not automatically result in learning (Singal 2015). Findings from this new research highlight the fact that teachers did not regard learning as being of central value for children with disabilities; rather they remained focused on letting children be together rather than children learning together. Teachers were conflicted in their views about the value of children with disabilities attending mainstream settings and a range of dilemmas were evident in their narratives. On the one hand, they appreciated fostering positive

Addressing dilemmas of difference 219 social interactions, but on the other hand, they did not believe that children with disabilities could easily learn in mainstream classrooms. Children with disabilities continued to be framed in a deficit-driven discourse where being disabled (and poor) positioned them as incapable and/or disinterested in learning. The intersectionality of poverty and disability shapes the experiences of children with disabilities from poor communities in other Southern countries (Department for International Development 2000), as is evident in this study wherein children were first-generation learners and belonged to families with low incomes. Educational research, Batra (2015) notes, has long seen poverty essentially as a barrier to schooling, and policies such as the building of more schools, or midday meals, have been framed to overcome the poverty barrier. However, very little attention has been paid to the processes that influence teaching and learning in schools where children of the poor study. Batra (2015, p. 1) powerfully concludes: [C]hildren of the poor are excluded from learning not because of the absence of conditions necessary for enabling participation and learning but because of the presence of conditions of capability deprivations that are found to characterise everyday classroom. The effects of poverty on schooling are further magnified for children with disabilities, especially for those from poor families. The poor quality of teaching and learning imparted in Indian schools has been of significant concern and under scrutiny for a long time, as noted by many researchers (e.g. Bangay & Latham 2013). Teachers in Indian classrooms have very few opportunities (and little time) to critically reflect on their practice (Majumdar & Mooij 2011). Given these challenges in the larger system, it is not surprising that children with disabilities, who have historically been denied a place in school, are now framed in the margins of the teaching and learning processes. Inclusion is not simply about upskilling teachers but also needs a deeper appreciation of the real barriers, including structural ones, that teachers face in everyday practice. Under Sarva Shiksha Abhiyan, even though the government has made clear provisions for para-professionals such as IERTs to support mainstream teachers in including children with disabilities, practices in the field are very different. The IERTs mentioned the high numbers of schools they were required to support, which resulted in their visits to each school being occasional and far from adequate. Additionally, while this professional cadre had been specifically trained and deployed to support teachers, they were regarded by teachers as being responsible for working directly with children with disabilities. There was no evidence of collaboration between these professionals and mainstream teachers. Thus, the lack of confidence amongst teachers to teach children with disabilities, the presence of experts but their inability to support teachers and the overwhelming focus on social inclusion resulted in a scenario where children with disabilities remained excluded from important school-based learning opportunities.

220  Nidhi Singal

Understanding teaching and learning processes The messages conveyed in international declarations and national policies give prominence to inclusive education, but they have also become a convenient cloak for preventing us from asking more difficult questions about how to achieve what is bring proposed. Fundamentally, inclusive education should relate to the dignity and well-being of each child (Reindal 2010), underpinned by values of social belonging, acceptance, respect and equality (Norwich 2013). In moving forward, current reforms in the Indian education system must be guided by a similar set of central principles rather than being driven by universalistic and idealistic notions of inclusive education. Pragmatism strongly anchored in key values and principles around respect and equity are important in shaping efforts aimed at educating children with disabilities. Efforts cannot be limited to overcoming barriers through the building of ramps, provision of basic aids and appliances scholarships for children with disabilities, etc. These are significant but not enough in themselves. There is a need to address processes in the classroom which are depriving children of the opportunity to learn. Rigorous research which focuses on deep investigation of classroom processes continues to remain thin in India, as is the case across many Southern countries (Hardman 2015). There is a growing need for educational researchers to develop a more nuanced engagement with understanding and developing effective teaching and learning opportunities for children with disabilities (Singal 2019b). Poverty and disability are not simply barriers to accessing schooling, but how these are embedded in everyday classroom processes also needs attention. Such systematic explorations will enable us to develop knowledge which can influence and support positive teacher practices. Teachers in mainstream classrooms are voicing some real concerns, not necessarily driven by a rejection of the child with disability. Rather these voices highlight an awareness of dilemmas or simply, the quandaries, they are faced with daily as they navigate teaching and learning in complex and resource-constrained classrooms. What is needed is greater clarity on the underlying values of what we seek from and through education, rather than continuing to uphold the rhetoric of inclusive education, a rhetoric that is clouded in ambiguity.

Notes 1 The team of researchers included Ms Aanchal Jain, who conducted the classroom observations and interviews with teachers and Ms Leelavathi (the translator). The research project was funded by CBM International. Access to schools and parents was facilitated by a local NGO. The data was transcribed, translated in India through a professional service and the analysis and writing up was undertaken in Cambridge. During the analysis, regular memos and updates were discussed with the field non-governmental organisation. Final analysis was also presented and discussed with the CBM International team through a webinar. 2 This refers to T – Teacher, female/male gender of the teacher (in this case female) and the number of years in the teaching profession. In cases where S1 is used, it refers to School 1; S2 refers to School 2.

Addressing dilemmas of difference 221

References Bangay, C., & Latham, M. (2013) Are we asking the right questions? Moving beyond the state vs non-state providers debate: Reflections and a case study from India. International Journal of Educational Development, 33(3), pp. 244–252. Batra, P. (2015) Quality of Education and the Poor: Constraints on Learning. [Online] TRG Poverty  & Education. Working Paper Series. Max Weber Stitfung. Available from: www.ghil.ac.uk/fileadmin/redaktion/dokumente/trg_india/Paper_4_ Poonam_Batra.pdf [Accessed 22 June 2020]. Bhatnagar, N.,  & Das, A. (2013) Nearly two decades after the implementation of persons with disabilities act: Concerns of Indian teachers to implement inclusive education. International Journal of Special Education, 28(2), pp. 104–113. Das, A.K.,  & Kattumuri, R. (2011) Children with disabilities in private inclusive schools in Mumbai: Experiences and challenges. Electronic Journal for Inclusive Education, 2(8). Department for International Development (2000) Disability, Poverty and Development. [Online] London: Department for International Development. Available from: https://hpod.law.harvard.edu/pdf/Disability-poverty-and-development.pdf [Accessed 15 July 2020]. Gafoor, K.A.,  & Asaraf, M.P. (2009) Inclusive Education: Does the Regular Teacher Education Programme Make Difference in Knowledge and Attitudes? Paper presented at the International Conference on Education, Research and Innovation for Inclusive Societies, March 19–21, Dravidian University, Kuppam, Andhra Pradesh, India. Government of Karnataka (n.d.) Primary Education. Department of Public Instruction. [Online] Available from: www.brlf.in/brlf2/wp-content/uploads/2018/05/ Elementary-Education-in-India_-2012-13.pdf [Accessed 15 July 2020]. Hardman, F.C. (2015) Making Pedagogical Practices Visible in Discussions of Educational Quality: Background Paper Prepared for the Education for All Global Monitoring Report 2015. [Online] UNESCO. Available from: https://unesdoc.unesco. org/ark:/48223/pf0000232449 [Accessed 22 June 2020]. Hoogeveen, J.G. (2005) Measuring welfare for small but vulnerable groups: Poverty and disability in Uganda. Journal of African Economies, 14(4), pp. 603–631. Johansson, S.T. (2016) Parents negotiating change: A middle-class lens on schooling of children with autism in urban India. Contemporary Education Dialogue, 13(1), pp. 93–120. Majumdar, M., & Mooij, J. (2011) Education and Inequality in India: A Classroom View. New Delhi: Routledge. Ministry of Human Resource Development (2009) Right to Education Act. [Online] Government of India. Available from: https://seshagun.gov.in/rte [Accessed 15 July 2020] Accessed 22 June 2020]. Ministry of Human Resource Development (2012) Sarva Shiksha Abhiyan. [Online] Government of India. Available from: https://seshagun.gov.in/ [Accessed 15 July 2020]. Ministry of Law and Justice (2016) Rights of the Persons with Disabilities (Equal Opportunities, Protection of Rights and Full Participation) Act, 1995. [Online] Government of India. Available from: http://niepmd.tn.nic.in/documents/ PWD%20ACT.pdf [Accessed 22 June 2020]. Minow, M. (1990) Making All the Difference. Inclusion, Exclusion and American Law. New York: Cornell University Press.

222  Nidhi Singal Mitra, S., Posarac, A., & Vick, B. (2012) Disability and poverty in developing countries: A multidimensional study. World Development, 41, pp. 1–18. Mizunoya, S., Mirta, S.,  & Yamasaki, I. (2016) Towards Inclusive Education. The Impact of Disability on School Attendance in Developing Countries. [Online] Office of Research – Innocenti Working Paper No. WP-2016–03. Available from: www. unicef-irc.org/publications/pdf/IWP3%20-%20Towards%20Inclusive%20Education.pdf [Accessed 23 June 2020]. National University of Educational Planning and Administration (2014) Elementary Education in India: Where Do We Stand? State Report Cards 2012 – -2013. [Online] Available from: www.brlf.in/brlf2/wp-content/uploads/2018/05/ElementaryEducation-in-India_-2012-13.pdf [Accessed 15 July 2020]. Neuman, W.L. (2013) Social Research Methods: Qualitative and Quantitative Approaches. Essex: Pearson Education. Norwich, B. (2002) Education, inclusion and individual differences: Recognising and resolving dilemmas. British Journal of Educational Studies, 50(4), pp. 482–502. Norwich, B. (2013) Addressing Tensions and Dilemmas in Inclusive Education: Living with Uncertainty. Abingdon: Routledge. Reindal, S.M. (2010) What is the purpose? Reflections on inclusion and special education from a capability perspective. European Journal of Special Needs Education, 25(1), pp. 1–12. Sarangapani, P.M. (2003) Constructing School Knowledge: An Ethnography of Learning in an Indian Village. Mumbai: Sage. Sawhney, S. (2015) Unpacking the nature and practices of inclusive education: The case of two schools in Hyderabad, India. International Journal of Inclusive Education, 19(9), pp. 887–907. Shah, R., Das, A., Desai, I., & Tiwari, A. (2013) Teachers’ concerns about inclusive education in Ahmedabad, India. Journal of Research in Special Educational Needs, 16(1), pp. 34–45. Singal, N. (2006) Inclusive education in India: International concept, national interpretation. International Journal of Disability, Development and Education, 53(3), pp. 351–369. Singal, N. (2015) Developments in Education of Children with Disabilities in India and Pakistan since 2000. Background Paper for the Education for All Global Monitoring Report 2015. [Online] Paris: UNESCO. Available from: https://unesdoc. unesco.org/ark:/48223/pf0000232424 [Accessed 15 July 2020]. Singal, N.  (2016). Schooling children with disabilities: Parental perceptions and experiences. International Journal of Educational Development, 50, pp. 33–40. Singal, N. (2019a) Challenges and opportunities in efforts towards inclusive education: Reflections from India. International Journal of Inclusive Education, 23(7–8), pp. 827–840. Singal, N. (2019b) Researching disability and education: Rigour, respect and responsibility. In Singal, N., Lynch, P., & Johansson, S.T. (Eds.) Education and Disability in the Global South: New Perspectives from Asia and Africa. London: Routledge. Singal, N., Ware, H.,  & Khanna-Bhutani, S. (2017) Inclusive Quality Education for Children with Disabilities. [Online] Report prepared for the World Innovation Summit for Education, Doha. Available from: www.wise-qatar.org/app/ uploads/2019/04/rr.6.2017_cambridge.pdf [Accessed 23 June 2020]. Srivastava, P., Noronha, C., & Fennell, S. (2013) Private Sector Research Study Sarva Shiksha Abhiyan. [Online] Report submitted to DFID India. Available from: www.

Addressing dilemmas of difference 223 prachisrivastava.com/uploads/1/9/5/1/19518861/srivastava_et_al._private_ sector_study_ssa_india.pdf [Accessed 23 June 2020]. UNESCO (2013) The Global Learning Crisis: Why Every Child Deserves a Quality Education. [Online] Available from: https://unesdoc.unesco.org/ark:/48223/ pf0000223826 [Accessed 23 June 2020]. UNESCO Institute for Statistics (2018) Education and Disability: Analysis of Data from 49 Countries. [Online] Available from: http://uis.unesco.org/sites/default/ files/documents/ip49-education-disability-2018-en.pdf [Accessed 23 June 2020]. United Nations (1948) Universal Declaration of Human Rights. [Online] Available from: www.un.org/en/universal-declaration-human-rights/ [Accessed 15 July 2020]. United Nations (2006) United Nations Convention on the Rights of Persons with Disabilities. [Online] Available from: www.un.org/development/desa/disabilities/convention-on-the-rights-of-persons-with-disabilities.html [Accessed 15 July 2020]. United Nations (2015) Sustainable Development Goals. Transforming Our Worldthe 2030 Agenda for Sustainable Development. [Online] Available from: https:// sustainabledevelopment.un.org/post2015/transformingourworld [Accessed 15 July 2020]. United Nations (2011) Development Goals. A Review of the MDG Process and Strategies for Inclusion of Disability Issues in Millennium Development Goal Efforts Agenda for Sustainable Goals. [Online] Available from: https://sustainabledevelopment. un.org/?menu=1300 [Accessed 23 June 2020]. Vijaysimha, I. (2013) ‘We are textbook badnekais!’: A Bernsteinian analysis of textbook culture in science classrooms. Contemporary Education Dialogue, 10(1), pp. 67–97.

12 Social distance, teachers’ beliefs and teaching practices in a context of social disadvantage Evidence from India and Pakistan1 Anuradha De and Rabea Malik In Southern country contexts, such as India and Pakistan, schools, teachers and the quality of teaching have been found to matter more than family background in positively influencing learning (Chudgar & Luschei 2009; Andrabi et al. 2007). The strong implication of this is that there is potential for teachers in these contexts to be able to compensate for disadvantages in student backgrounds and so to level the playing field. Having an accurate and complete understanding of the factors that can improve teaching and learning, particularly in contexts of social disadvantage, is crucial for making decisions about the types of policies to put in place to promote equality in education. Historically, the education production function approach has provided policy guidance on inputs considered necessary for ensuring learning (Hanushek 2008; Glewwe et al. 2011). The approach predicted that easily observable inputs such as qualifications and the experience of teachers could be used as a proxy for effective teaching. In this framework, the most effective teacher would be someone with the most years of experience and the highest qualifications. Yet, repeated applications of this framework to data from different contexts, including India and Pakistan, have established that these observable characteristics are not good predictors of teacher effectiveness (Andrabi et  al., 2007; Aslam et  al. 2019; Hanushek  & Rivkin 2012). Another framework, the value-added approach, predicates the notion of effective teaching on learning gains achieved by the teacher in a year and sets out to find characteristics of teachers and schools to which these gains can be attributed. Research that applies this framework has identified that latent variables such as contractual arrangements, subject content knowledge, motivation and pedagogical practices can have a stronger influence on children’s learning than observable characteristics (Aslam et al. 2019; Singh & Sarkar 2012). Teachers’ beliefs are considered important for shaping the learning environment and influencing student motivation and by doing so improving the education process (Organisation for Economic Cooperation and Development 2009). One aspect that potentially shapes teachers’ beliefs relates to their closeness to, and engagement with, students from different backgrounds and communities. A key question, therefore, is whether such social distance plays a role in shaping

Social distance and teaching practices: India and Pakistan 225 both pedagogic practice and learning outcomes or indeed contributes to inequalities in such outcomes. Some studies in India and Pakistan show that children taught by teachers of similar gender and/or caste perform better (Rawal & Kingdon 2010; Tamim & Haq 2015; Chudgar & Sankar 2008), however, others have produced conflicting evidence (Karachiwalla 2019). Furthermore, these studies reveal associations but do not elaborate on the mechanisms underpinning those associations (Rawal & Kingdon 2010; Sarangapani 2003).2 The proportion of children from marginalised backgrounds in mainstream classrooms is very high in India (Social Research Institute  – IMRB 2014), as well as in Pakistan to a lesser extent. Therefore, understanding ways in which social distance may impact learning becomes especially important. If teachers (who are from more privileged backgrounds and enjoy a higher social status than many of the children they are teaching) exhibit or express negative biases, are unaware of the challenges faced by children from disadvantaged backgrounds or are unable to support them, then classrooms in government schools can become sites for exclusion. In this chapter, our starting point is that such social distance is highly significant and that teachers’ beliefs are at the centre of the link between social distance and students’ successful or unsuccessful learning. This then raises the two key questions which we address in the chapter: (1) How, and to what extent, does social distance shape teachers’ beliefs and practices? (2) What is the likely effect of social distance on the education or learning of children from disadvantaged backgrounds? We begin by drawing on existing research to theorise a connection between social distance, teachers’ beliefs, their teaching practice and student learning. We then present empirical data from interviews with teachers and classroom observations in India and Pakistan to: (i) describe teachers’ beliefs, highlighting those particularly relating to children’s home backgrounds; (ii) note how teachers’ beliefs vary for different children and, (iii) describe observed teaching practice and teacher– student interaction. Our classroom observation tool allows us to differentiate between teaching practices in relation to high- and low-performing children. We find that while teachers are not actively biased against children from poor or low-caste households, they do believe family background to be the most significant determinant of learning. Furthermore, teachers’ conceptualisation of a good student in both countries comprises a set of characteristics that are much more likely to be present in children from middle- or upper middle-income families (those where parents are educated and in stable economic circumstances) or naturally gifted children. Teachers, through their classroom practice, end up favouring these good students. The likely outcome is that low-performing students or those from marginalised backgrounds may end up being excluded in the classroom.

226  Anuradha De and Rabea Malik

Social distance, teacher beliefs and teaching The notion of social distance, simply put, expresses the gap between social positioning of two individuals or groups. This gap may be due to class differences (rich and poor), ethnic or racial differences or gender differences. It is a sociological concept which takes multiple dimensions of identity into account in its expression of remoteness between individuals and groups. The concept of social distance can be traced back to Robert Park (1924) who defined it as ‘an attempt to reduce to something like measurable terms the grades and degrees of understanding and intimacy which characterise personal and social relations generally’ (cited in Wark & Galliher 2007, p. 383). The related concept of prejudice was defined in this work as ‘the instinctive disposition to maintain social distances’. These notions were applied to studies of race and ethnic relations in the United States by sociologist Emory Bogardus (1933). At the centre of his concept are differences in identities, where social distance is observed in the interactions (or lack thereof) between individuals belonging to different social groups. Social distance becomes particularly important when it results in exclusion from public services.3 This link between markers of identity such as social class and race and social exclusion is also to be found in the sociological discourses on social reproduction. Within these discourses, schools are thought to reproduce existing disadvantages through prioritising and valuing certain types of knowledge or in failing to ensure adequate resources and support to students from disadvantaged backgrounds (Bourdieu 1986). The Coleman Report in the 1960s is considered seminal for the empirical documentation of differential access to public resources in education by race, post segregation and the associations of these input inequalities with differential outcomes for African American students (Coleman et al. 1966). In another classic study, Bowles and Gintis (2002) document the ways in which schooling processes and trajectories determine unequal labour market outcomes.4 A framework on social identity and learning outcomes that elaborates on mechanisms has been developed by economists Akerlof and Kranton (2002). They draw on sociological concepts of identity, social categories and perceptions to develop a social category model that is also relevant to the way we are thinking about social distance. In this model, developed in the United States, schools impart an image of ideal students, in terms of characteristics and behaviour through activities like assemblies and day-to-day interactions in classrooms. Teachers praise and reward some students based on these perceived characteristics, while they disapprove of and punish others. Each of these identities (that of an ideal student or a poorly performing student) has expectations associated in terms of who is likely to succeed academically. Students determine the amount of effort they expend in school in response to how they think they are perceived by teachers and match their identity to these social categories. These choices are seen to be particularly poignant for

Social distance and teaching practices: India and Pakistan 227 students whose backgrounds (such as African Americans, Hispanics and other minorities) conflict with the school’s ideal. In choosing the level of effort they exert, the students match their identity to these constructed categories of good and poorly performing students. The expectations that teachers set for students through their pedagogic practice, interactions and even through what they say, form students’ expectations of their own success. Rosenthal and Jacobson (1968) tested through an experiment, in schools in the United States, the degree to which changes in teacher expectations produce changes in student achievement and found that for children in primary schools teachers’ expectations were a self-fulfilling prophecy.5 Students may perform less well if they expect to be discriminated against or if they have had low expectations set for them. McKown and Weinstein (2008) found that students in the United States whose teachers have high expectations for them performed better on achievement tests than students for whom teachers had low expectations. While the studies mentioned here are set in the United States, they offer us an insight into links between identity, practice and outcomes that is largely missing in the studies emerging from Southern countries. We use these insights to develop a framework for social distance, beliefs and outcomes in the next section. The salient features of identity in South Asia which relate to the notion of social distance include caste, gender, ethnicity and social class. Many of these are interrelated. In Pakistan, caste, occupational status and poverty are connected with each other and have been linked to the lack of educational mobility (Cheema  & Naseer 2013). In India, caste is a far more formal and visible marker of disadvantage,6 which also intersects with poverty and occupational status. Majumdar and Mooij (2012) highlight the increasing social mobility of teachers alongside the widening gap between their and students’ social status. The empirical work on teacher and student identity in South Asia therefore leads us to expect that teachers will have higher expectations from students from stable, high-income backgrounds (similar to their own) and low expectations from children from unstable, poor homes. However, there are very few studies that explicitly capture teacher expectations regarding students from disadvantaged backgrounds in this context.

Teachers’ beliefs and teaching practice Turning now to beliefs and practices, Figure  12.1 presents a stylised and simplified representation of factors that determine teacher practice and effort in classrooms. Knowledge of curricular content, pedagogy and incentives (career progression, rewards but also sanctions determined by the system) impact on teacher practice and their level of effort directly. However, whilst much has been written about these two factors, teacher beliefs are a relatively less explored component of the framework.

228  Anuradha De and Rabea Malik

Practice and Effort Knowledge

Education and Training

Beliefs

Identity

(socio-economic status, position, ethnicity, gender)

Incentives

Capacity Building

External Accountability

(parents, department, communities)

Figure 12.1 Stylised framework linking teacher practice and effort with identity, ­attitudes and beliefs, alongside other factors Source: Authors

Teachers’ beliefs about a wide range of aspects can impact on their teaching practice and their level of effort in teaching. These beliefs include those about the nature of learning and knowledge acquisition (epistemological beliefs), their own ability to convey knowledge (self-efficacy) and also – crucially – their beliefs about the ability of children to learn and the factors that influence this ability. The beliefs that teachers hold reflect their identity as individuals who are participants in society. As such, teachers may be influenced by cultural biases rooted deeply in the social fabric of which they are a part. Some beliefs may be formed, shaped and reshaped through teacher education whilst training and capacity-building support can be a source of knowledge about new alternative approaches that challenge beliefs and offer new tools that may mitigate these biases and institute positive values. Beliefs and practices are closely linked. In order to understand how beliefs impact practice, Pajares (1992) asks us to consider how a teacher responding to disciplinary problems and deciding on the appropriate management technique to use will make a series of judgements about the child when doing so. Her point is that such  – and other  – interactions in class involve continuous judgements and ‘evaluations of people, contexts and situation’ (p.  313). These evaluations are informed not just by professional but also by personal beliefs. Pedagogical practices that are deeply rooted in cultural beliefs about gender and caste may intersect with teachers’ beliefs about how children learn and which children can learn. Cultural biases and social inequities can then permeate student–teacher interactions (Brinkmann 2015). There is considerable evidence of the differential practices used by teachers in relation to students from disadvantaged backgrounds in South Asia which

Social distance and teaching practices: India and Pakistan 229 may be interpreted as bias. For example, Jha and Jhingran’s (2005) multi-state study in India of schooling of the most deprived children attributed the discrimination faced by children from families living in poverty partly to teachers’ cultural attitudes and their implications in the classroom. Similarly, in her 2015 survey of government elementary schools in India, Vasavi found that, for many teachers, their social and cultural background formed the basis of their beliefs which in turn impacted on their practices of classroom management in a way that reproduced biases and prejudices. Teachers’ subjective judgements have been demonstrated to impact on assessments as well (Rawal & Kingdon 2010). Jacoby and Mansuri (2015) report parents’ accounts of caste-based discriminatory practices, covering a range of strategies from the poor treatment in school of children from low-caste households to total exclusion from instruction altogether.7 Tamim and Haq (2015) also argue that exclusion is embedded in the political economy of social relationships and that it may occur deliberately and explicitly or silently. When these authors explored how caste results in such exclusion in Pakistani schools, they found that discriminatory practices included segregating children into different classrooms where there was lower teacher effort and poor treatment of students. They also found reports showing disregard for the learning of children from lower castes. The negative effects of social distance are apparent in acts of discrimination or prejudice or prejudicial beliefs. Discrimination manifests as negative stereotyping of a particular group of students and involves an association of low expectations from them. These are often expressed through teachers’ classroom behaviour (praise or criticism) and impact on student confidence.

Two channels that link beliefs and learning We combine insights from these studies to depict two pathways which could potentially link teachers’ beliefs and learning: (i) teachers’ expectations from students regarding how much they can achieve and (ii) teaching practice and the level of effort invested in, and attention paid to, students of different ability and home backgrounds (Figure 12.2). In the first pathway, teachers’ beliefs are linked to their expectations of how students learn. According to Akerlof and Kranton (2002), once stereotypes are created and expectations set, they will be apparent to students. Students will either work hard to meet those expectations, or in the case of low expectations fail to put in any effort. Expectations are likely to be high for good students and low for poorly performing students. These expectations are often expressed as the notion of who an ideal or good student is or the kinds of characteristics an ideal or good student possesses. High levels of student effort predict higher student learning. The second pathway predicts that teachers’ effort may vary for different students. Effort translates into support received by students. Teachers’ effort in relation to inclusive practices includes the attention paid to students who are

230  Anuradha De and Rabea Malik

Expectations of students and how they learn

>>> high for ‘good’ students >>> low for ‘poor’ students

Concept of a good Student

Student Effort

Teacher Beliefs

Learning

Effort through practice

>>> high for ‘good’ students >>> low for ‘poor’ students

Support received by student

Figure 12.2  Two channels linking teachers’ beliefs with student learning Source: Authors

struggling, the amount of time, feedback and support provided, how the classroom is organised and whether instruction is provided in ways that benefit the pupils who require more attention (Westbrook  & Croft 2015). Such inclusive practices matter because students from disadvantaged background are likely to receive lower home support and are more dependent on their teachers for learning. Also, if these students receive lower levels of support from teachers, their learning levels are likely to be even lower. Empirical data collected from teachers and classrooms in government schools in India and Pakistan provided us with an opportunity to interrogate these theorised pathways. We document, using survey and interview data, teachers’ expectations of students by asking who they believe to be good students and those that are easy to teach. We explored which teachers found it more difficult to teach and the factors they believed contributed to these difficulties. Using observational data, we then considered whether such teachers’ practice differed for students from different backgrounds.

Our research study This chapter uses data collected from the Teaching Effectively All Children (TEACh) Project8 which studied determinants of effective teaching through two rounds of surveys at 86 government primary schools (36 schools in India and 50 schools in Pakistan), including interviews with teachers and assessment of students, in rural Haryana in India and rural Punjab in Pakistan. These two sites were part of the same state under the colonial rule and have a similar school development history. They are also from similar agro-economic zone. Thirty

Social distance and teaching practices: India and Pakistan 231 villages were randomly selected from three districts in each site. The government primary schools in which the village children were primarily enrolled were chosen for the study (see Aslam et  al. 2019 for further details).9 Schools were visited at the start and end of the school year in order to conduct teacher and student quantitative surveys and learning assessments based on Young Lives and ASER tests.10 In total, 95 teachers in India and 129 teachers in Pakistan were surveyed regarding their beliefs about children’s capacity to learn, the challenges they faced during teaching and about their teaching practices. In addition, 12 of these 86 schools were selected for more detailed exploration. This qualitative aspect of the research involved classrooms observations in grades 4 and 5 in each of the schools, along with in-depth semi-structured interviews with teachers (17 in India, out of which 5 were female and 18 in Pakistan, of which 10 were female). The findings included in this chapter are from the surveys, the teacher interviews and classroom observations. The teacher interviews included specific questions about teachers’ perceptions about children’s abilities to learn (what factors support children’s learning or not learning, with probes and follow-up questions about identity markers such as social class, caste and gender), children they considered good students and specific children who struggled the most. Responses to these questions underpin the analysis we present in the next section. We conducted three sessions of observations in two classes in each of the 12 schools. There were two observers in the classroom, each with their own observation sheet – with one observer focused on the teacher and teaching practices and the other focused on four pre-selected children in the class. Scores on tests administered during the baseline school survey were used for selecting children: two high-performing students and two low-performing students. If a child with a disability had been identified in the class, they were included in the observation of students. The classroom observation tools recorded chronological developments throughout the class session, the seating positions of children, specific details about the classroom culture and negative and positive interactions. The observation tool recorded the activities of all the children and of the four pre-selected children. Data from these observations are the basis for the patterns described in the section on teacher practice. The socio-economic distance between teachers and parents in the sample schools was apparent in the gap between them in terms of levels of education and types of occupation. In Pakistan, 40% of the fathers and 64% of the mothers were illiterate. In India, 20% of fathers and 35% mothers were illiterate. In contrast, around 80% of teachers interviewed were graduates or postgraduates. The teachers worked in the government sector, and their salaries are regular and much higher compared to average income levels in rural areas. Very few parents had jobs comparable in nature and status to that of the teachers. The majority of mothers had no paid employment: 76% of the mothers in the Indian sample and 89% mothers in the Pakistan sample did not have paid work. The majority of the fathers in India were engaged in a daily wage or blue-collar jobs (40%), while

232  Anuradha De and Rabea Malik others were engaged in agricultural work (25%) or ran shops (11%). In Pakistan, while similar proportions were engaged in agriculture or ran shops, fewer fathers were involved in daily waged work. India has achieved near-universal primary education and consequently the number of children from the poorest households are in government schools (ASER 2018). By contrast, in Pakistan, the poorest children are still likely to be out of school (ASER 2019). With a high proportion of children from advantaged backgrounds in private schools, more so in India, the majority of the students in government schools are from disadvantaged backgrounds.11 The differences in education and occupation status between teachers and parents are quite high and very likely wider in India than Pakistan. Next we explore some of the consequences of such differences.

Do teachers have higher expectations of some students? In India, when teachers were asked to identify ‘good students’ in their class, they discussed three types of characteristics. The first way was in terms of those who were intelligent, had high IQs, could grasp concepts easily, were hard working and scored high marks in tests. The second was how students behaved in school, such as asking questions to clarify doubts, ensuring that their homework and classwork were checked regularly,12 whether they displayed higher levels of confidence, were more focused, well mannered, obedient and disciplined, had leadership qualities and could teach other students, if required. The third set of characteristics included their appearance and attendance, such as whether they were neat and tidy in their appearance, attended school regularly, were punctual, brought the required textbooks and stationery and regularly studied at home. The following are some of teachers’ perspectives of how they differentiated good students from others: [W]henever they are asked to any classwork, they [good students] continue to ask doubts till the concept is clear to them. Only when the concept is clear, they move ahead. Whatever is not clear, they don’t leave it and then move further. They don’t take any work given by the teacher lightly. They take it seriously, complete it and try to live up to the expectations of the teacher. Secondly, their habits are better than the others. Their habits and good manners differentiate them from the other students. That’s why they are good students. (Class 5 teacher, India) When something is being taught, if they don’t understand, then they ask again. They come properly bathed. They are disciplined. Students who remain discipline will always succeed. And the kids who are naughty or the one who is not disciplined their attention keeps diverting from one place to another and this is why they lag behind. (Class 5 teacher, India)

Social distance and teaching practices: India and Pakistan 233 They are good students not just because they are good at studies though they are tez [have high IQ levels], but also because they are responsible, they make the class form a queue and lead, they try to do their homework and class work, kehna maante hain [They do what I tell them], naha dho kar aate hain [They bathe before coming], they come on time, saaf safayi rakhte hain [They maintain cleanliness]. (Class 4 teacher, India) They [parents] do not think [about studies]. And there is definitely some difference because of the IQ level. Those children who are padhane mein hoshiyar [good at studies] are ahead in sports too. (Class 4 teacher, India) Teachers in Pakistan believed that a good student is someone who is intelligent, attends school regularly, is attentive in class, participates by asking questions, takes initiative by showing up to have his/her copies checked (class work or homework). A good student is someone who is obedient, neat and clean, and someone who does not copy work from other students: They are good students. They do their homework on time. They do their homework themselves. They don’t copy from other students. They participate in class and ask questions if they do not understand. I think they are genuinely interested in studying and that is why I think that they are good students. (Class 5 teacher, Pakistan) They are intelligent and hardworking and pay attention during the class. They are not careless, they ask me questions and if they are having any difficulty, they share it with me. And I also feel very happy that the child is taking interest and it motivates me to give more attention to these kids. (Class 4 teacher, Pakistan) There are around 10 students in my class who learn very quickly. I just need to tell them once and they understand. The rest of the students, they ask again and again. They need some time and they also waste other’s time too. Intelligence is also a factor and their home environment is also a factor. Their family members also take care of them. (Class 5 teacher, Pakistan) According to the teachers interviewed for the study in both India and Pakistan, good students are those who are viewed as being naturally intelligent, follow classroom teaching, are punctual and regular and study at home. According to the teachers, these good students are independent learners and quick to complete their work. On the flip side, those who take a long time to complete

234  Anuradha De and Rabea Malik class work or seek help from their peers are not good students. This has implications for the teaching and learning process in the classrooms: we found through our classroom observations that children who are struggling to keep up in class often rely on their peers, ask them questions, peek into their notebooks or copy from them because they are unable to receive the guidance needed in class from the teacher. While a student’s own ability is to a large extent intrinsic to them (though malnutrition and illness can have an adverse impact), it is more likely that children not doing their homework or receiving support from parents are ones whose parents are not literate themselves or do not have time because of their jobs. Regular attendance is closely linked with parental occupation. Children from families who do agricultural work and casual wage labour are more likely to get pulled out of classes and miss school days.

The significance of parental background In our surveys and interviews, we asked teachers about those children who had more difficulty learning (Figure  12.3). The vast majority of teachers surveyed (88% in the Pakistan, 90% in India) believed that children with parents who cannot read and write have more difficulty learning. The second largest category of those who had difficulty learning included children with disabilities (84% in both countries). When asked about children from poor backgrounds, 75% of teachers

84 84

Children with disabilities

Children who speak a different language

53

65

88 90

Children with illiterate parents

Children from poor households

75

41

Girls

28

4 0

10

20 Pakistan

30

40

50

60

70

80

90

100

India

Figure 12.3 Percentage of teachers who agree or strongly agree that specific background factors affect children’s ability to learn Source: TEACh School Survey 2017 Note: the responses were on a 5-point Likert scale and the percentages are of those who agreed, or strongly agreed.

Social distance and teaching practices: India and Pakistan 235 in Pakistan said they believed these children had greater difficulty, and a smaller but a substantial proportion (41%) in India said the same. In other words, almost all teachers expect children whose parents are not literate themselves and are from poor families to have lower capabilities of learning. Their reasoning is that, if parents are not able to help children review their classwork and augment the teaching done in class through instruction at home, children will not be able to learn. Parents who are poor or not educated were pinpointed by teachers, with their parenting practices classified by teachers as ‘neglectful’ or ‘disinterested’. The following quotes from teachers in Pakistan are examples of these views: The children have problems at home. Their mothers go to work as daily labourers. Sometimes picking potatoes, or arvi [Taro]. How will these mothers pay attention to their children? (Class 4 teacher, Pakistan) Only a few students are from the ‘stable’ economic background. Most of the students are from financially weak households. They attend school in the morning and work in evening time. They work in fields. (Class 4 teacher, Pakistan) Also, sibling or parents of few children are educated and therefore are able to help them. Moreover, a few children come from problematic families and face issues while learning at school as well. Children from poor families also sometimes face difficulty in learning since they feel that they do not possess all those things which their classmates have. (Class 4 teacher, Pakistan) In some cases, teachers were more understanding in realising that it is not a lack of interest on the part of the parents but due to problems they face in providing support, whether it is a result of time constraints or lack of literacy. A teacher from India said the other main reason for looking at parents’ financial background, especially in the context of the village where most people were poor, was that poorer parents leave for their work early in the morning, hence: They are unable to keep a track on their children, whether they are going to school on time, whether they are studying, not studying and so on. It is also not their fault because they leave at 7 in the morning, come back at 5 in the evening, are very exhausted by then and are unable to look after their children’s studies after that. (Class 5 teacher, India) If the home environment is good and there is importance given to studies then the child will also turn out to be good. And if some poor parents are

236  Anuradha De and Rabea Malik illiterate then they will obviously find it difficult to educate their children. Because they are not educated, they will not be able to tell their children what to do and what not to do. They can see that they are writing something, but who knows what they are doing and what that they tell their parents. This also happens. (Class 5 teacher, India) Family background and parental occupation also matter for teachers because in some instances students who attend school irregularly do so because their parents are taking them to work. These are most likely to be parents who were in casual labour or very poor and had not been to school themselves. A number of teachers mentioned irregular attendance during the harvest season. For example: When these children get done from school . . . they help in their parents’ jobs or around the house. In the harvesting season, the strength in the classes will reduce by nearly 50%. For example, if the class has 20 students, only 10 will come as the rest are out working in the fields with their parents. They have to do their own work. That’s why they can’t learn. (Class 3 teacher, Pakistan) They [children in families that work as contractual labour on brick kilns] are absent for three months [at a time]. [If they come back to school, we] need to start from the beginning. Yasira was fine for 3–4 months. When she leaves for contractual work at the kiln, I will have to start from the beginning again. (Class 4 teacher, India) While caste is an important indicator of social status, no teacher reported that the ability of children from disadvantaged castes was low: No, no direct impact of caste (on learning). It is possible that there is a particular caste which is backward in education, do not have livelihood options, have to go out for earning to brick kilns or farms, then automatically their children would lag in studies, and that is not because of their caste but family reasons. (Class 5 teacher, India) These quotes show that teachers’ expectations regarding the ability of students from poor backgrounds or from homes where parents do not have a stable economic situation, by and large, is low (with the exception of those from disadvantaged castes). In having low expectations of learning and achievement for students who are less likely to receive attention at home, teachers are reinforcing disadvantage, rather than countering it. Instead of delinking or muting the influence of family background, schools are potentially reinforcing their negative influence on those from disadvantaged backgrounds.

Social distance and teaching practices: India and Pakistan 237

Differentiated teacher practice Inclusive practices require teachers to provide a disproportionate amount of effort and attention to the students who are struggling. Inclusive classrooms are where all children are included in interactions and encouraged to participate. In this section we share findings from the classroom observation data collected in the 12 sample schools in India and Pakistan. As noted earlier, the design of the classroom observation tool involved teachers and selected high- and lowperforming students being observed separately. In India, there were variations in the extent to which teachers adopted inclusive practices in their classes. A few classes were observed where teachers used innovative methods to teach and attempted to involve all students. But there were other classes where teachers taught through reading and repeating and focused only on those students who were sitting nearby and following the class. On the whole, student–teacher interactions were pleasant and positive.13 Positive interactions were explicit and found with greater frequency with children who were better performing. Teachers were clearly stating their high expectations from such students. During one observation: The teacher asked her [Poonam] to answer a question that was relatively tougher than the previous ones, and the teacher said, ‘Stand up, Poonam! This question matches your level of intelligence’. The girl stood up and gave the correct answer too. The teacher praised her by saying, ‘well done!’ and confirmed her full faith in the girl’s intelligence as she said to her, ‘I had expected this from you’. (Observer note) There were few cases in Indian classrooms where teachers were observed actively trying to include all students, particularly the ones with lower learning levels. The low-attaining students being observed were often found to be disinterested and not following the class. But in certain cases when they made comments or wanted to respond, they were ignored by the teacher. Silent exclusion was observed in several classrooms. Teachers were seen to ignore or overlook raised hands of some students when they wanted to respond, while they acknowledged responses from others more readily. One teacher, however, stood out with her practice, when she specifically asked questions to the ones who were not ‘high-attaining’ students and explained carefully why the responses were correct or not. But this was quite an exception. And she, too, could not disguise differences in expectations. When she had to ask a more difficult question, she asked the high-attaining girl who was being observed, saying that this was a question worthy of her to respond. This general pattern of calling on the same high-attaining students and those seated in front was observed during multiple observations. One observer noted that: The teacher did not make any arrangement for other students to participate, for example, asking children who responded most to pause and ask others to

238  Anuradha De and Rabea Malik speak up. The teacher made himself available only to those who were sitting right in front of him. (Observer notes) It was also observed that the teachers were content with getting a quick answer to their question, even if it came from the same child, and others were not answering. Children seated at the back of the class were more disengaged, and the poorly performing students were seated in the far corners of the classrooms. These children were often seen talking amongst themselves. These children were often ignored by teachers, and even when the teachers did engage, it was not clear if effort had been made to ensure that all students had walked away from the interaction, having understood the concept being taught. Teachers were not questioning these students often or checking for understanding. In the schools in Pakistan, similar general trends were observed. Children seated at the front of the class interacted more with the teacher and had their work checked more often. One observer noted that the children towards the back of the class were almost invisible. However, it was not always clear if children seated at the front were those who needed more attention. For example, in boys’ schools, low-attaining boys were more likely to be seated at the back. In girls’ schools, low-attaining girls were equally likely to be in front or at the back. Students seated at the back received less attention. In their interactions, the Pakistani teachers tended to choose the same children for responding to the questions they asked. A higher frequency of positive interactions was similarly observed between the teachers and high-performing students and conversely, the frequency of negative interactions was higher with low-performing students. Teachers tended to get more easily frustrated with them, thus perhaps not surprisingly, low-attaining children were more likely to be overlooked in teacher interactions. These students seemed, in multiple observations, tense and confused. They did not receive support from the teacher. In one observation, a low-performing student approached the teacher multiple times with the intent to have his notebook checked, however he returned without doing so. The teacher did not pick up on this. The observer checked the child’s notebook to find that none of the previous work of this child had been corrected. It appears from classroom observations in both countries that the teachers do, through their actions, communicate their different expectations for the students. And they are more likely to provide greater effort for the students who actively participated in class and those they think of as ‘good’ students.

Discussion and conclusion In this chapter we have drawn on existing understandings of exclusion to frame pathways through which teachers’ beliefs about student ability may impact learning outcomes. The data collected from the two sites of the study indicate that social distance between teachers and students in government schools exists in both of the study sites. Demographic data from the schools and communities

Social distance and teaching practices: India and Pakistan 239 sampled confirm a generally spoken truism about government schools in India and Pakistan: a majority of children in these schools are from homes where parental literacy levels are low, their incomes are low and their occupations either blue collar or irregular. All these aspects combine to make the social distance between teachers and the students they teach quite wide. If teachers believe that children from poor families and those with illiterate parents cannot be taught, their practices will reflect this belief resulting in exclusion of children who arguably need the most help. We bring empirical findings to bear on two questions in this chapter: (i) how and to what extent does social distance shape teachers’ beliefs and practices? and (ii) what is the likely effect of this on children’s education and learning? In response to the first question, we find that teachers in the samples in both Pakistan and India strongly believe that parental background determines the pupil’s capacity to learn and to do well in class. The teachers’ conceptualisation of a ‘good student’ is someone who is intelligent, confident, interactive and attentive in classroom, attends regularly and is able to learn at home. These are all characteristics that are much more likely to be present in children from socially and economically stable and well-off homes, somewhat similar to the teachers’ own background. Teachers expect that these children will learn more and that they will do better. A majority of teachers agree that children whose parents are not literate have the most difficulty in learning. They expect children of parents with low levels of education or low income to have lower capability of learning. The exceptions are students who are exceptionally intelligent but from poor backgrounds. We also find that the link between the child’s background and his abilities is more strongly asserted by teachers in India. In Pakistan teachers are less likely to emphasise only family background. We interpret this to be an outcome of a larger number of poor children being present in government classrooms in India, intensifying the social distance between teachers and the students they teach. In response to the second question, we find that through these expectations and their practices, teachers may be reproducing disadvantages that the student arrives with. Evidence from classroom observation shows that rather than focusing more on students who have lower learning levels and who are less likely to participate in classroom process, teachers engage more with students who are more motivated and engaged in classroom. The first group is not openly excluded but rather there is more of a silent process of non-inclusion. Interactions with students also communicate to the students what the teachers’ expectations are about their performance. In other words, though the expectation is for teachers to contribute towards overcoming disadvantages that children face (and these can be multiple and intersecting in many ways), children who are perceived as not being intelligent or who do not have support from home may well be excluded from the learning process. Our findings imply that teachers may not be aware how their beliefs about the children’s ability may be biased and how that may impact on the students. Teachers appear to be prioritising children at the top of the learning distribution, and

240  Anuradha De and Rabea Malik those at the bottom of the learning distribution are being excluded. While not all, many of these children are those who will also not be receiving support at home. These findings have implications for contexts with very diverse classrooms, not just in terms of socio-economic background but also student ability. This has to do with teachers’ ability to cater to the needs of children from poor and marginalised backgrounds, respond to their cues in class and adapt their teaching to their needs. Survey data in this study also recorded teachers in both countries stating that they did not feel adequately prepared to support children from poor backgrounds. The emergent policy implication from this is that if social inequality and disadvantage are to be addressed, there is a need to introduce additional teacher-training components. This would include reflection on the social distance between teachers and the taught, the nature of teachers’ own beliefs about social distance and difference so that they do not have biased expectations from students and to provide teachers with knowledge of teaching strategies that can differentiate instruction in an ameliorative way. We identify teacher beliefs, practice and interactions in the classroom as an area for further study.

Notes 1 We are grateful for funding from Economic and Social Research Council and the Department for International Development for this research under the Raising Learning Outcomes grant ES/M005445/1. 2 Rawal and Kingdon (2010) indicate that the role model theory may explain why teachers of similar identities (gender, caste, socio-economic category) may benefit students: seeing individuals similar to them may improve students’ attitudes towards education. Sarangapani (2003) finds that belonging to the same community increases students’ comfort level with their teachers. 3 The link between identity and exclusion from public services in general (outside education) has been documented by political economists in Pakistan and India. Cheema and Mohmand (2007) document the disparities by caste and class in access to sanitation and safe drinking water in rural areas in Pakistan in a report about social structures. Banerjee et al. (2005) link social divisions to public goods access in India. 4 Bowles and Gintis (2002) have been critiqued for ignoring the role of family in their analysis. 5 They call this the Pygmalion effect. For older children the link between expectations and outcomes is less strong. 6 In India caste remains an important identity of households marking social status. With affirmative government policies towards historically disadvantaged castes, its relevance may have decreased, but the caste-based occupational segregation and economic inequality still continues to a varying extent (Desai and Dubey 2012). 7 As caste is not officially recognised as a form of exclusion in Pakistan, there are no mitigation policies in place. 8 See www.educ.cam.ac.uk/centres/real/researchthemes/teachingandlearning/ effectiveteaching/ 9 One school was chosen if it was co-educational. If not, a boys’ and a girls’ school were chosen for these villages. 10 Students were tested in mathematics, English and Urdu or Hindi. A combination of ASER and Young Lives assessments was administered. These tests were used for selecting high- and low-performing students for classroom observations.

Social distance and teaching practices: India and Pakistan 241 11 ASER data from India and Pakistan show that in 2015, 60% of children from the top income quartile in India were in private schools, while ASER 2018 in Pakistan shows only 30% of top quartile income children were in private schools. 12 During school visits in India, it was observed that in large classes teachers check homework and class work only of those students who bring them to the teacher for correction. 13 Teaching practices observed could be influenced by the presence of observers. This was anticipated and so the research team visited the school over a period of 7 to 12 days. Each teacher was observed in their class thrice so as to minimise the potential bias.

References Akerlof, G.A., & Kranton, R.E. (2002) Identity and schooling: Some lessons for the economics of education. Journal of Economic Literature, 40(4), pp. 1167–1201. Andrabi, T., Das, J., Khwaja, A.I., Vishwanath, T., & Zajonc, T. (2007) Learning and Educational Achievements in Punjab Schools (LEAPS): Insights to Inform the Education Policy Debate. [Online] World Bank. Available from: https://epod.cid.harvard. edu/sites/default/files/2018-02/learning_and_educational_achievements_in_ punjab_schools_leaps-_insights_to_inform_the_education_policy_debate.pdf [Accessed 29 June 2020]. ASER (2018) The Annual Status of Education Report 2018. [Online] Pratham. SafdarJung Enclave, New Delhi. Available from: http://img.asercentre.org/ docs/ASER%202018/Release%20Material/aserreport2018.pdf [Accessed 29 June 2020]. ASER (2019) The Annual Status of Education Report 2019. [Online] Idara-e-Taleemo-Agahi (ITA). ASER Pakistan Secretariat, Lahore. Available from: http://aserpakistan.org/document/aser/2019/reports/national/ASER_National_2019.pdf [Accessed 29 June 2020]. Aslam, M., Malik, R., Rawal, S., Rose, P.,  & Vignoles, A. (2019) Do government schools improve learning for poor students? Evidence from rural Pakistan. Oxford Review of Education, 45(6), pp. 802–824. Banerjee, A., Iyer, L., & Somanathan, R. (2005) History, social divisions, and public goods in rural India.  Journal of the European Economic Association,  3(2–3), pp. 639–647. Bogardus, E.S. (1933) A social distance scale. Sociology and Social Research, 17, pp. 265–271 Bourdieu, P. (2011) The forms of capital (1986). In Szeman, I., & Kaposy, T. (Eds.) Cultural Theory: An Anthology. Chichester: Wiley-Blackwell. Bowles, S., & Gintis, H. (2002) Schooling in capitalist America revisited. Sociology of Education, 75(1), pp. 1–18. Brinkmann, S. (2015) Learner-centered education reforms in India: The missing piece of teachers’ beliefs. Policy Futures in Education, 13(3), pp. 342–359. Cheema, A.,  & Mohmand, S. (2007) Decentralisation and inequality in Pakistan: Bridging the gap that divides. In Ali, S.M., &Saqib, M.A. (Eds.) Devolution and Governance: Reforms in Pakistan. Pakistan: Lahore: Oxford University Press. Cheema, A., & Naseer, M.F. (2013) Historical inequality and intergenerational educational mobility: The dynamics of change in rural Punjab. Lahore Journal of Economics, 18, pp. 211–231.

242  Anuradha De and Rabea Malik Chudgar, A.,  & Luschei, T.F. (2009) National income, income inequality, and the importance of schools: A hierarchical cross-national comparison. American Educational Research Journal, 46(3), pp. 626–658. Chudgar, A., & Sankar, V. (2008) The relationship between teacher gender and student achievement: Evidence from five Indian states. Compare, 38(5), pp. 627–642. Coleman, J.S., Campbell, E., Hobson, C., McPartland, J., Mood, A., Weinfeld, F., & York, R.L. (1966) Equality of Educational Opportunity Study.  [Online] United States Department of Health, Education, and Welfare. Available from: https:// files.eric.ed.gov/fulltext/ED012275.pdf [Accessed 29 June 2020]. Desai, S., & Dubey, A. (2012) Caste in 21st century India: Competing narratives. Economic and Political Weekly, 46(11), p. 40. Glewwe, P.W., Hanushek, E.A., Humpage, S.D., & Ravina, R. (2011) School Resources and Educational Outcomes in Developing Countries: A Review of the Literature from 1990 to 2010. [Online] National Bureau of Economic Research, Working Paper No. 17554. Available from: www.nber.org/papers/w17554.pdf [Accessed 29 June 2020]. Hanushek, E.A. (2008) Education production functions. The New Palgrave Dictionary of Economics, 2, pp. 749–752. Hanushek, E.A., & Rivkin, R. (2012) The distribution of teacher quality and implications for policy. The Annual Review of Economics, (4), pp. 132–159. Jacoby, H.G.,  & Mansuri, G. (2015) Crossing boundaries: How social hierarchy impedes economic mobility. Journal of Economic Behavior and Organization, 117, pp. 135–154. Jha, J., & Jhingran, D. (2005) Elementary Education for the Poorest and Other Deprived Groups: The Real Challenge of Universalization. New Delhi: Manohar Publishers. Karachiwalla, N. (2019) A Teacher Unlike Me: Social Distance, Learning, and Intergenerational Mobility in Developing Countries. [Online] International Food Policy Research Institute. Available from: www.journals.uchicago.edu/doi/pdfplus/10.1086/698131 [Accessed 29 June 2020]. Majumdar, M., & Mooij, J. (2012) Education and Inequality in India: A Classroom View. London: Routledge. McKown, C., & Weinstein, R.S. (2008) Teacher expectations, classroom context, and the achievement gap. Journal of School Psychology, 46(3), pp. 235–261. Organisation for Economic Cooperation and Development (2009) Chapter  4: Teaching practices, teachers’ beliefs and attitudes. In Organisation for Economic Cooperation and Development (Ed.) Creating Effective Teaching and Learning Environments: First Results from TALIS. Paris: Organisation for Economic Cooperation and Development Publishing, pp. 88–120. Pajares, M.F. (1992) Teachers’ beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62(3), pp. 307–332. Park, R.E. (1924) The concept of social distance: As applied to the study of racial relations. Journal of Applied Sociology, 8, pp. 339–334. Rawal, S., & Kingdon, G. (2010) Akin to My Teacher: Does Caste, Religious or Gender Distance between Student and Teacher Matter? Some Evidence from India. [Online] Institute of Education, Department of Quantitative Social Sciences, Working Paper No. 10–18. Available from: https://repec.ucl.ac.uk/REPEc/pdf/qsswp1018.pdf [Accessed 30 June 2020]. Rosenthal, R.,  & Jacobson, L. (1968) Pygmalion in the classroom.  The Urban Review, 3(1), pp. 16–20.

Social distance and teaching practices: India and Pakistan 243 Sarangapani, P.M. (2003) Constructing School Knowledge: An Ethnography of Learning in an Indian Village. New Delhi: Sage. Singh, R.,  & Sarkar, S. (2012)  Teaching Quality Counts: How Student Outcomes Relate to Quality of Teaching in Private and Public Schools in India. [Online] Young Lives, Working Paper No. 91. Available from: https://ora.ox.ac.uk/objects/ uuid:a66e6233-9263-418e-a9f4-ddb1ef25d97a [Accessed 30 June 2020]. Social Research Institute – IMRB (2014) National Sample Survey of Estimation of Out-ofschool Children in the Age 6–13 in India: Draft Report. [Online] Available from: https:// mhrd.gov.in/sites/upload_files/mhrd/files/upload_document/NationalSurvey-Estimation-School-Children-Draft-Report.pdf [Accessed 30 June 2020]. Tamim, T., & Haq, H. (2015) The intersection of caste, social exclusion and education opportunity in rural Punjab. International Journal of Educational Development, 43, pp. 51–62. Vasavi, A.R. (2015) Culture and life of  government elementary  schools. Economic and Political Weekly, 50(33), pp. 36–50. Wark, C., & Galliher, J.F. (2007) Emory Bogardus and the origins of the social distance scale. The American Sociologist, 38, pp. 383–395. Westbrook, J.,  & Croft, A. (2015) Beginning to teach inclusively: An analysis of newly-qualified teacher pedagogy in lower primary classes in Tanzania.  Teaching and Teacher Education, 51, pp. 38–46.

Index

Note: page numbers in italics indicate a figure and page numbers in bold indicate a table on the corresponding page. Page numbers followed by ‘n’ refer to notes. Abbink, J. 104, 105 Abdullah, A. 23 access to education 2, 5, 6, 7, 8, 9, 11, 19, 22, 25, 28, 29 – 30, 33, 37 – 54, 59 – 74, 79, 80, 82, 83, 84, 92, 98, 105, 106, 112, 113, 115, 123, 124, 127, 140, 147, 156, 160n5, 165, 168, 185, 187 – 188, 188, 190, 191, 202, 207, 218, 220; aid to education 37, 38, 46 – 52, 53 – 54; child labour and 62, 63, 64, 73; for children with disabilities 207; and demographic transitions 38 – 39; dropouts 39, 41 – 42, 79; and earnings 25; educational financing gaps 48 – 52; egalitarian education reforms 59 – 74; girls’ education 98, 105, 106, 106, 112, 113, 115; marginalised/ vulnerable children 22, 46, 79, 84, 201, 210; and non-state actors 81, 84; out-of-school children 7, 46, 79, 188, 188; patterns of enrolment 7, 39 – 42; patterns of participation by gender 42 – 45, 44; poverty and 37, 71, 100, 219, 220; public–private partnerships 8, 79 – 94; and returns to education 28, 29 – 30; school fees and 84, 86, 91, 92, 93, 155, 172, 177; and social inequalities 60; in sub-Saharan Africa 185; universal 37, 39, 48, 49, 50, 52, 60, 71, 123, 124; see also Ethiopia; Ghana; India; Pakistan; Sri Lanka; Uganda; passim Chapters 3, 4 Acemoglu, D. 31 Acharya, D.R. 145

Ackah, C. 31 activity based learning (ABL) 67, 69 Adivasis 73, 75n15 affordability 33, 84, 86 Afghanistan 41, 44, 45 Agarwal, B. 145 agency of women 145 – 146, 149, 154 Agyei-Quartey, J. 11, 185 Aham-Chiabuotu, C.B. 174, 179 Ahmed, Abiy 116n4 Ahmed, S. 125 Ahmed, S.S. 146, 147 Ahorlu, C.K. 166, 178 aided schools see grant-in-aid schools aid organisations 47; donors 47, 50; strategies 52 – 53 aid to education 37, 38, 46 – 52, 53 – 54 Aikman, S. 5 Aja, G.N. 174, 179 Akerlof, G.A. 226, 229; see also social category Akhtar, H.Z. 146 Akindele, O.A. 172 Akufo A. 53 Akyeampong, K. 3 Albania 42 Alcott, B. 87 Andrabi, T. 86, 224 Angrist, N. 32 Annual Status of Education Report (ASER) 79, 86, 87, 231, 232, 240n10, 241n11 Ansari, A. 91 Appadurai, A. 9, 10, 124, 125, 134, 139, 140; see also capacity to aspire

Index  245 Araya, M.W. 105 Arnot, M. ii, xx, xxi, 1, 14n2, 14n4, 125, 127 – 128, 131 – 132, 140 – 143 Arokiasamy, P. 146 Aromolaran, A.B. 31 Arshed, N. 23 Asadullah, M.N. 145 Asaraf, M.P. 218 Asegdom, A. 108, 116n4 Asian Development Bank 71 Aslam, M. 5, 8, 79, 81, 82, 83, 84 – 85, 87, 88, 89 – 90, 89, 91, 93, 94n6, 156, 224, 231 aspiration window 125, 132, 134, 138; see also capacity to aspire; Ray, D. Associates for Change 11, 185, 191, 204n4 Association for the Development of Education in Africa (ADEA) 54n1 Aturupane, H. 31, 65, 74, 74n12 Atuyambe, L.M. 169 autonomy 87, 88, 144, 145 – 146 Axinn, W.G. 154 Azid, T. 147 Bandaranaike, S.W.R.D. 64 Bandyopadhyay, M. 68, 69 Banerjee, A. 240n3 Bangay, C. 219 Bangladesh 30, 41, 44 Barr, N. 19 Barrera-Osorio, F. 82, 90 Barro, R.J. 25, 26 Bashir, S. 147 Basu, A.M. 145, 146 Basu, B.D. 61 Batra, P. 208, 219 Becker, G.S. 22 Benavot, A. 59 Benin 41, 44 Bhatnagar, N. 218 Bhatti, F. 5, 9, 10, 144, 149, 154, 160n3 Bhutan 42, 44 Biddlecom, A. 179 Billy, J.O.G. 146 Blystad, A. 104 Bogardus, E. 226; see also race; social distance Botswana 5, 8 Bould, S. 146, 147 Bourdieu, P. 2, 125, 127, 132, 134, 140; see also cultural capital; habitus,

Habitus Listening Guide; social capital; social class reproduction; symbolic violence; theory of practice Bowles, S. 226 Boyden, J., 102 Brahmanandam, N. 146 Brahme, M. 73 Breidlid, A. 5 Brinkmann, S. 228 British Raj 61 – 62 Bronfenbrenner, U. 165 – 166; see also ecological theory Brownrigg, R. 61 bullying/harassment 101; discipline 193 – 196; labelling/stereotyping 211, 229; male bullying 43; of schoolgirls by teachers 173; sexual violence and teenage pregnancy 174, 175, 178, 179 Burgess, G.L. 101, 104 Burkina Faso 41, 44 Burundi 41, 44 Caillods, F. 39 Cambodia 41, 44 Canning, C. 38 capability theory 2; see also Sen, A. capacity to aspire 9, 10, 124, 125, 140; see also Appadurai, A.; aspiration window Caribbean, the 81 Carvalho, S. 116n4 Casely-Hayford, L. 5, 11, 185, 186, 188, 201 caste 3, 6, 9, 61, 62, 63, 69, 70, 72, 73, 75n15, 126, 131, 132, 134, 136, 137, 138, 139, 143, 209, 225, 227, 228, 229, 236, 240n3, 240n6, 240n7 Central Advisory Board of Education, India 62 Central African Republic 41, 44, 45, 180n7 Central Statistical Agency (CSA) 102 Chad 41, 44, 45, 176 Chappell, L. 99, 114 charter schools 8, 83, 84 – 85, 90 Chau, K. 174, 179 Chaudhry, A. 127 Cheema, A. 126, 227, 240n3 child-centred pedagogy 66, 67 child labour 62, 63, 64, 69, 73 Chile, universal voucher programme in 85 China 38, 60

246 Index Chohan, Z. 166 Christian missionaries 61, 62, 72 Chudgar, A., 224, 225 church schools/education 61, 72 Chuta, N. 101, 102 Citizens for Democracy, India 66 Clapham, C. 103, 104, 105 Coady, D. 23, 32 cognitive skills, development of 7, 32 Colclough, C. 1 – 2, 4 – 6, 7, 8, 13 – 14, 14n4, 19, 21, 23, 25, 27, 31, 32, 33n1, 37 – 38, 39, 42, 46, 49, 50, 54n1, 71, 72, 80, 98, 100, 101, 144, 160n2 Coleman Report 226 colonialism 61 – 63, 72 Comoros 41 complementary basic education see Ghana completion, school: in LICs/LMICs 39, 40 – 42; in sub-Saharan Africa 165 conflict/crisis contexts, state-subsidised schools in 85 Congo 41, 44 Congress Party, India 66 Connell, R. 5 Constitution of Ethiopia (1950) 104 Constitution of India (1950) 63 Constitution of Uganda (1995) 167 – 168, 169 contraceptive prevalence rate, impact of girls’ schooling on 152 – 153, 153 corporal punishment 195 Cote d’Ivoire 41, 44 COVID-19 pandemic 13, 47, 50, 51, 54n2 Crawford, I. 19 Crawfurd, L. 91, 92 Crivello, G. 101 Croft, A. 230 cultural beliefs of teachers 229 cultural capital 129 – 130; see also Bourdieu, P. curriculum: indigenous knowledge 5; reforms, in Sri Lanka 67; sexual and reproductive health services 169 – 170, 175 – 176, 178; sexuality education 11, 169 – 170, 174, 178, 180; views of teachers 209, 212 – 215 curriculum dilemma 209, 212 – 215 Customary Marriage Act (1973), Uganda 169 Dalits 73, 75n15 Darling, M. 126

Das, A. 218 Das, A.K. 207 Das Gupta, M. 146 Datta, S. 84 Day Ashley, L. 3, 80, 81, 83, 84, 91, 92, 94n6, 94n9 De, Anuradha 5, 12, 71, 72, 224 Democratic Republic of Congo 41, 44, 47 demographic transitions, and access to education 38 – 39 Denzin, N.K. 5 Department for International Development (DFID) 52, 54n1, 94n6, 219 dependency theory 54 Derg regime, Ethiopia 103, 104 Desai, S. 240n6 Deshpande, R. 69, 74n2 De Silva, K.M. 62, 74n2 Development Assistance Committee (DAC) 46 – 47, 50, 53 Di John, J. 98 dilemmas of difference 12, 206, 208 – 210, 213, 215, 211 – 217; see also disabilities, children with; India; Norwich, B.; passim Chapter 11 disabilities, children with 11 – 12, 79, 206 – 208; curriculum dilemma 212 – 215; dilemmas of difference 208 – 210, 211 – 217; exclusion of 206; identification dilemma 211 – 212; learning inability of 212; location dilemma 215 – 217; and poverty 219; quality of education received by 208, 219; seating arrangements in classroom 213; special schools for 214 – 216; teaching and learning processes 220; tendency to lower expectations for 213; see also India; Pakistan; passim Chapter 11 District Information System on Education (DISE), India 86 District Primary Education Programme (DPEP), India 71, 207 Dizioli, A. 23, 32 domains of power framework 9, 98 – 99, 114; see also Hickey, S.; Hossain, N. Dougherty, C. 29 Drèze, J. 69, 139 Dribe, M. 159 dropouts, school: in LICs/LMICs 39, 40 – 42, 45, 49, 98, 79, 111, 136,

Index  247 174, 186, 187; and abuse by teachers 195; and livelihood patterns 190; negative consequences of 178; and teenage pregnancy 172, 177, 178; see also Ethiopia; Ghana; Pakistan; Uganda Dubey, A. 240n6 Dyer, C. 5, 67 Dyson, T. 145 early years education 7, 11, 19, 22, 39, 49, 50; see also primary (basic) education earnings: and demands for skilled labour 21, 23; distribution of 22 – 23; and education, relationship between 19 – 24, 27, 30; inequalities 7, 20, 21, 22 – 23, 26, 31, 32, 33, 71, 176; and laws of supply and demand 23; passim Chapter 2 Easterly, W. 48 East India Company 61 ecological theory 165 – 166; see also Bronfenbrenner, U. economic returns see returns to education education: achievement 41, 46, 59, 60, 62, 91, 101, 176, 227; achievement gaps 87; aspirations for taraqqi 9, 128, 129, 130, 131, 132, 134, 135, 136, 137, 139; and equity 6, 7, 8, 20, 22 – 23, 30, 37, 38, 46, 47, 66, 71, 79, 81, 83, 84, 87, 89 – 92, 102; fundamental 59, 60; and gender 8, 42 – 45, 66, 87, 98 – 116; girls’ 9, 10 – 11, 30, 42 – 45, 43, 84, 106, 92, 98 – 116, 130 – 131, 133 – 134, 138, 139, 144 – 160, 165 – 180, 191, 196, 197, 199 – 200, 203, 238; learning 3, 6, 13, 22, 23, 37, 48, 49, 52, 66, 83 – 87, 89 – 90, 92, 93, 98, 100 – 101, 106, 178, 192, 193, 194, 201, 202, 207, 212, 213, 217, 218 – 220, 224 – 225, 234 – 236, 238 – 240; literacy and numeracy 6, 186, 192; outcomes of 2, 4, 10, 21, 37, 48, 49, 52, 80, 83 – 86, 88, 89 – 90, 92, 93, 99, 100 – 101, 105, 126 – 127, 132 – 139, 186, 201, 202, 225, 226, 238; out-of-school children 7, 11, 13, 31, 38, 46, 79, 123, 134, 136, 166, 177 – 178, 185, 186, 187, 188, 188, 189, 190, 206,

218, 232; participation in 42 – 45, 50, 105, 106, 185, 191, 198; quality of 5, 6, 8, 11, 19, 23, 31 – 33, 38, 41, 48, 49, 59, 66, 69, 82, 83, 84, 86, 92, 98, 106, 186, 187, 191, 202, 206, 208, 219, 224; rates of return to 7, 19 – 33, 123, 140; retention in 9, 91, 106, 115, 186; social norms in 136, 172, 174 – 176, 178; and teenage pregnancy 10 – 11, 165 – 174, 176 – 180, 196; see also access to education; passim Chapter 3 Education 2030 Framework for Action 47, 50 educational financing 4, 7, 8, 22, 37, 38, 82, 54n2, 63, 82, 88; aid 37, 46 – 52, 53; compensation 48; cost saving reforms 49; cost shifting reforms 49; and domestic revenue 47, 48, 51, 53; efficiency and effectiveness gains 52 – 53; and enrolment of girls 21; fiscal reforms 37, 51 – 52, 53; gaps 48 – 52, 53; government subsidy 83, 85, 90; issues in progress towards sustainable financing 50 – 52; in LICs/ LMICs 47 – 48, 49; quality enhancing reforms 49; and returns to education 19; sub-prime borrowers 47, 48; targeted financial programmes 30; voucher schemes 8, 83, 85 – 86, 89, 90 – 92; see also Sri Lanka educational systems 3, 4, 6, 8, 12, 13, 22, 31, 65, 66, 67, 68, 70, 74, 80, 82, 88, 123, 124, 127, 132, 188, 190, 193, 206, 207, 208, 220; aid to 37, 38, 46 – 52, 53 – 54, 88; egalitarian reforms 7 – 8, 59 – 74; equity and 37 – 38, 85, 98, 102, 103 – 104, 108, 132; finance 7, 37, 38, 82; income inequality and 7, 33, 71; labour markets and 31 – 32; planning 7, 45, 52, 68; politics of 9, 98 – 116; poverty alleviation 4, 5; resistance to change 62, 63 – 64, 65, 67, 69, 72, 73 Education and Training Policy (ETP), Ethiopia 104, 106 Education Commission 22 Education for All (EFA) 1, 3, 8, 13, 22, 37, 48, 54n1, 54n4, 59, 60, 68, 69, 70, 71 education policy 1, 2, 8, 13, 23, 59, 60, 104; colonial 61 – 63; compulsory schooling 62, 63, 64, 69, 70, 124,

248 Index 158; educational expansion 7, 23, 62, 63 – 65, 72, 73, 105, 124, 134, 136; educational financing gaps 48 – 52; egalitarian reforms 7 – 8, 59 – 74; free schooling/fee free education 11, 62, 63, 65, 69, 71, 90, 167, 196, 197, 203; gender equality reforms/policy 107 – 109, 114, 115; girls’ education policy 98, 99, 104, 105 – 107, 112, 113, 114, 115, 144; neoliberal education policy 2, 80; political drivers and interests, history 7, 59 – 74; and political settlement 98 – 99, 105; post-colonial 63 – 72; pre-colonial 61; public– private partnerships 88, 93, 94; related to children with disabilities 207, 214; and returns to education 7, 19; Sarva Shiksha Abhiyan 69, 71, 207, 219; stakeholders’ views/ attitudes 109 – 113, 110; teenage pregnancy and education inequality 167 – 171; women’s power to affect 107 – 109, 115 education production function approach of teaching 224 Education Sector Development Framework and Program (ESDFP), Sri Lanka 71 Education Sector Development Plans (ESDPs), Ethiopia 106 – 107, 106, 108, 111 Education Voucher Scheme, Pakistan 90 – 91 Elacqua, G. 81 Elementary Education Bill (1911), India 62 elementary education see primary (basic) education Eligu, C. 116n4 elites 3, 13, 61, 72, 73, 126, 127 Eritrea 41, 44 Ethiopia 6, 8 – 9, 41, 44, 46, 47, 98 – 99; access to education 6, 8, 98 – 100, 115; aid to education for 47; civil society organisations in 104 – 105; dominance of girls’ education in political priorities 103 – 105; dropouts, school 98; early marriage in 102; educational reform, politics of 102 – 103; Ethiopian People’s Revolutionary Democratic Front (EPRDF) 103, 104, 116n5; and focus on economic development 105;

and gender bias 109, 112, 113; and gender norms 99, 100, 101 – 102, 111, 113; and gender parity 107, 111 – 112, 115; gender quotas 109, 113; idea of developmental state 103; informal institutions 99, 107 – 114, 115; lack of attention and support 111; masculinity 101, 167; mixed views of stakeholders 109, 110, 112 – 113; narrow views of stakeholders 109, 110, 111 – 112; outof-school children in 46; patriarchy/ patriarchal norms 9, 101; policy legacies and strategies 99, 105 – 107, 106; political settlement of 98 – 99, 103; politics of 114 – 116, 114; poverty in 100, 101, 103, 106; primary school enrolment in 100, 100, 101; progress in female education and gender equality in 100 – 102; progressive views of stakeholders 109, 110 – 111, 110, 113; teachers in 106, 106; women’s power/representation in policy domain 107 – 109, 115; passim Chapter 6 ethnicity 4, 65, 72, 79, 189, 190, 227 – 228 Faridi, M.Z. 147 Farrell, J.P. 185 feminism 104 fertility attitudes/behaviour of women 10; contraceptive prevalence rate 152 – 153, 153; desired family size 150 – 151, 150, 155, 159; desired sex composition of children 151 – 152, 159; impact of girls’ schooling on 150 – 153, 159; and maternal schooling 145 – 147 Fiseha, A. 105 forced marriage, and teenage pregnancy 173, 175, 177 – 178 Forum for African Women Educationalists (FAWE) 8 Foundation Assisted Schools programme, Pakistan 90 Framework for Action to Meet Basic Learning Needs 48 Frey, C.B. 32 Froerer, P. 160 Gafoor, K.A. 218 Galliher, J.F. 226 Gambia 41, 44

Index  249 Gandhi, Indira 66 Gandhi, Mahatma 62, 63 Gandhi, Rajiv 66, 68 Gazdar, H. 126 Gebresilassie, F.H. 105 gender 3, 8 – 9, 10, 69, 73, 87; adolescent boys/girls 165 – 167, 171 – 172, 173 – 175, 177, 178, 179, 180; and enrolment in higher education 43; equality 100 – 102, 104 – 105, 107 – 109, 110, 111 – 112, 113; and fertility 145; gap, in school enrolment 21; gender relations and structure 130, 131, 139, 144, 189, 207, 209, 225, 226, 227, 229; inequalities, intra-household 153 – 155; mainstreaming 104, 107; norms 98, 99, 100, 101 – 102, 111, 112, 113, 131, 138, 173, 177; parity 43, 44, 84, 111 – 112, 115, 165; patterns of participation by 42 – 45, 44; and primary (basic) schooling 28, 42, 43, 44, 45; and primary school enrolment 100, 101; and returns to education 25 – 26, 28 – 30 sexuality 3, 10, 170, 173, 174, 178, 180; violence/harassment 101, 167, 173, 174, 175, 178, 179; see also Ethiopia; Ghana; India; Pakistan; Uganda Gender Parity Index 43 Georgia 42 Ghana 5, 6, 31, 42, 44, 53; and absence of primary schools 202; adaptability of 201, 203; commitment level of facilitators 194 – 195; community experiences of education 191 – 193; complementary basic education (CBE) in 11, 185 – 187; contextual factors affecting 189 – 190; and costs of primary education 186, 201, 202; decision-making within families 198 – 200; dropouts, school 186, 187, 190, 195; and ethnic tensions 190; and family size 198; financial benefits 196 – 197, 202; food insecurity 200; girls’ education 199 – 200, 203; government/public schools in 191, 202; Gushegu district, in Northern Region, in 188, 189, 190, 191, 192, 193 – 194, 195 – 197, 198, 200, 201; incidences of out-of-school children and access to primary schools 188; influence of peers 200; linking

decision-making and schooling choices 196 – 200; and livelihood patterns 190, 196, 202, 203; marriage in 185, 199, 203; parental support for 187; patriarchy in 198, 201; pedagogic and disciplinary approaches 193 – 196; and poverty 187, 201, 202, 203; research study 187 – 189; Pusiga district in Upper East Region 188 – 189, 190, 191, 193, 194 – 195, 197, 198 – 199, 200, 201; religion in 187, 189; teachers in 185 – 186, 189, 191 – 198, 202 – 203; and teenage mothers 196; tension in enrolment of older and younger children 200; passim Chapter 10 Ghana Population and Housing Census (GPHC) 189, 191 Ghana Statistical Service (GSS) 186, 187, 190 Ghartey, A. B. 11, 185, 201 Ghimire, D.J. 154 Gintis, H. 226 girls’ education 9, 84, 199 – 200, 203; access 9, 106, 115; daughters’ aspirations 9, 130 – 131, 133 – 134, 138, 139; educational interventions 30, 170, 180; enrolment of girls 42 – 45, 43; and fertility preferences 10, 144 – 160, 174 – 175; girls’ schooling 30, 147, 150 – 155, 191, 197, 238; learning 92, 98, 100 – 101, 106, 107, 178; and patriarchy 101, 127, 146, 166, 174 – 176; politics of 98 – 116; retention 9, 106, 115; sex education 10 – 11, 168, 170, 174, 178; and teenage pregnancy 10 – 11, 165 – 180, 196; see also Ethiopia; Ghana; India; Pakistan; Uganda Glewwe, P.W. 224 Global Gender Gap Index 141n3 Global Monitoring Report 1, 8, 54n4 Global North 2, 6, 12; high income countries 21, 171; PPPs in 83 Global Partnership for Education (GPE) 46, 47, 50 Goetz, A.M. 99, 105 Gokhale, G. K. 62 Goldin, C. 20, 23, 31, 32 government/public schools 8, 62, 65, 71, 75n13, 80, 85, 86 – 88, 92, 93, 131, 156, 158, 191, 210, 214, 225, 230, 231, 232, 238 – 239; and

250 Index learning achievement 86 – 87; parental abandonment of 86; Partnership for Management arrangement 90; see also Ghana; India; Pakistan; Sri Lanka; teachers, practices of; Uganda government subsidy, and PPPs 83, 85, 90 Govinda, R. 68, 69 Gram, L. 146 grant-in-aid schools 62, 87 – 88; see also India; Pakistan Green, A. 49 Green Revolution, India 23 Grindle, M. 74n2 Growth and Transformation Plan (GTP), Ethiopia 105 Guilmoto, C.Z. 146 Guinea 41, 44, 176 Guinea Bissau 44 Guyana 44 Guzzo, K.B. 147 Haas, B. 174, 179 habitus 125, 127, 129, 134; Habitus Listening Guide 125, 132, 141n1; see also Bourdieu, P. Hagmann, T. 104, 105 Hagos, Belay 116n4 Haiti 41 Hakim, A. 145, 146 Hansen, W.L. 22 Hanushek, E.A. 224 Haq, H. 225, 229 Hardman, F.C. 220 Härmä, J. 70, 74n12 Hart, C.S. 125 Hartwell, A. 185, 186 Hastings, J. 19 Heath, R. 146 Heyneman, S.P. 21 Hickey, S. 9, 98 – 99, 114, 115; see also domains of power framework higher education 7, 19, 165; access to 22, 29, 43, 49, 50, 68; equity impact for 22 – 23; and gender 43, 172 – 173; returns to 19, 21, 25, 27, 28, 30, 31, 32; skills training 7, 19, 31, 32; universities 21, 22 – 23, 27, 63; vocational colleges/education 21, 22, 167, 215 Hilgers, M. 126 Hindus/Hinduism 62, 74 Himaz, R. 31 Home-Based Volunteers (HBVs) 210 Honduras 42, 44

Hoogeveen, J.G. 206 Hossain, N. 9, 98 – 99, 114, 115; see also domains of power framework human capital 20, 37; human capital theory 7, 20 – 21, 23 human rights 37, 84, 167, 207; see also Universal Declaration of Human Rights Hussain, A. 127 Hussain, R. 146 Hutter, I. 174, 179 ICF International 170, 172, 178 Idara-e-Taleem-o-Aagahi (ITA), Pakistan 94n4 identification dilemma 209, 211 – 212 Ilie, S. 22 IMC Worldwide 186 inclusion/exclusion: children with disabilities in 11 – 12, 206 – 220; disciplinary practices 228 – 229; exclusionary practices 206, 225, 229, 237; good child/student, the 225, 227, 229 – 230, 231, 232 – 234, 238, 239; inclusionary practices 213 – 214, 230, 237; labelling 211; out-of-school children 206, 218; see also India; passim Chapter 11 Inclusive Educational Resource Teachers (IERTs) 210, 214, 217, 219 income inequality 7, 20, 21, 22 – 23, 26, 31, 32, 33, 71, 176 India 5, 6, 42, 51, 52, 240n3, 240n6, 241n11, 241n12; from 1990 to present 68 – 70, 71, 72; child labour 61; child-centred approach 66; children from disadvantaged castes 236; colonial rule 61, 62, 72; compulsory and free primary education in 62; conceptualisation of ‘good student’ 232 – 234, 239; curriculum dilemma 212 – 215; demographic transition in 38; differentiated teacher practice 237 – 238; dilemmas of difference 208 – 210, 211 – 217; Direct Payment Agreement (1972), Kerala 87; diversity in 74; drivers and interests shaping education reforms 72 – 73; drivers for education expansion (from independence to 1970) 63 – 64, 65; egalitarian education reforms in 7 – 8, 59 – 60, 73 – 74; elementary education in 60, 62, 63, 66, 69, 71; evolution of

Index  251 public–private partnerships in 87 – 88, 89; expectations of teachers 232 – 236; free and compulsory education 63, 69; and gender 66, 69, 73, 209, 225, 227, 231; girls’ education in 62, 63, 79, 92; government/public schools in 71, 86, 87 – 88, 92, 93, 225, 230, 232, 238 – 239; grant-in-aid schools in 87, 88; Green Revolution 23; identification dilemma 211 – 212; inclusive education for children with disabilities in 11 – 12, 206, 207, 208, 209, 220; Inclusive Education Resource Teachers 214, 217, 219; instability and struggle for universal education (1970–1990) 66 – 67, 68; intergenerational mobility in 147; languages 64; learning inability 212; learning of children with disabilities 218 – 219; literacy rate in 59, 61, 63; location dilemma 215 – 217; mass education 61, 62, 64; non-state sector in 86; Operation Blackboard 66 – 67; out-of-school children in 79; parental background, significance of 234 – 236; pedagogy 224 – 225, 227, 228; political and economic instability in 72 – 73; poverty in 93, 227, 229; pre-independence period 61 – 63; primary school-age population in 94n5; private schools in 65, 70 – 71; private tuition in 217; public–private partnerships in 8, 79 – 94; quality of teaching and learning 219; religion in 59, 61, 65, 73; research study 210 – 211, 230 – 232; resistance and promotion by interest groups 73; Salary Disbursement Act (1971), Uttar Pradesh 87; sameness, notion of 211 – 212; school enrolment in 70, 72, 74, 93, 210; seating arrangements in classroom 213; secondary education in 59, 72; shifts at micro level 217 – 218; size and population of 60; social distance 225, 227, 238 – 239; social groups 73; special schools 214 – 216; student attendance and parental occupation 236; support from external agencies 71 – 72; teachers’ beliefs and teaching practices, link between 229; teachers’ perceptions 218; teaching practices in 12, 224, 225; tensions between centre and states 67; tensions

regarding public–private partnerships in 93; universal franchise 62, 63, 72; voucher schemes 85 – 86, 91 – 92; passim Chapters 4, 5, 11, 12 Indian Education Commission 62 Indian Peace Keeping Force 68 Indo-Sri Lanka Peace Accord 68 inequality/ies 1 – 3, 5 – 6, 9 – 13; gender 9, 10, 105, 107, 112, 113, 114, 115, 127, 145, 147, 165, 201; income 7, 20, 21, 22 – 23, 26, 31, 32, 33, 71, 176; intergenerational mobility 146, 147, 156 – 158, 159, 185; intrahousehold 153 – 155; kinship 9, 126, 131, 132, 134; in low- and lower-middle income countries 37 – 54; patronage 9, 126, 127; race 4, 226; see also Bogardus, E; rural 2, 65, 67, 73, 123 – 141; social class 6, 70, 71, 73, 139, 209, 226, 227; social conflict 8, 73, 74; social mobility 9, 37, 124, 125, 127, 128, 129, 134, 140, 147, 155, 157, 159 – 160; and teenage pregnancy 165 – 180; urban 2, 65, 67, 73; see also caste; disabilities, children with; ethnicity; landownership; patriarchy; poverty; power; religion Institute of Development Studies (IDS), Sussex University 1, 2, 5, 54n1 intergenerational mobility 147, 156 – 158, 159 International Commission on Financing Global Education Opportunity 22 International Finance Facility for Education (IFFEd) 48, 50 International Monetary Fund (IMF) 47 intrahousehold inequalities, and girls’ schooling 153 – 155 investment, educational 19, 20 – 21, 23, 24, 29 – 30 Isenman, P. 74n74 Islam, K.M.M. 145 Islamic education 141n2, 141n5 Iyer, L. 61 Iyer, P. 116n4 Jacobson, L. 227 Jacoby, H.G. 229 Jain, A. 220n1 Janatha Vimukthi Peramuna (JVP), Sri Lanka 67, 68 Jayasuriya, J.E. 61, 62, 63, 74n2 Jayaweera, S. 65, 74

252 Index Jayewardene, J.R. 68 Jeffery, P. 139, 145, 160n1 Jeffery, R. 1, 9, 10, 144, 145, 146, 149, 154, 160n1, 160n4 Jeffrey, C. 2, 132, 139 Jejeebhoy, S.J. 146 Jelili, M.O. 172 Jensen, R. 19, 21 Jewkes, R. 166, 170 Jha, J. 229 Jhingran, D. 229 Jin, H. 22 Jodhka, S.S. 63 Johansson, S.T. 217 Jolly, R. 4 Jomtien 7, 59, 69, 71 Jones, P.W. 21 Juhn, C. 21 Juneja, N. 65 Kabeer, N. 69, 145 – 146 Kandiyoti, D. 127, 146 Kannangara, C.W.W. 62 Kantur Consulting 186 Karachiwalla, N. 225 Kasarda, J.D. 146 Kattumuri, R. 207 Katz, L.F. 23, 31, 32 Kenya 42, 44, 98, 172 Khan, S. 127 Khurshid, A. 145 Kingdon, G.G. 5, 8, 29, 65, 69, 74n2, 74n74, 79, 84, 86, 87, 88, 94n1, 225, 229, 240n2 Kotalawala, D.E.M. 65 Kothari Education Commission 64 Kranton, R.E. 226, 229; see also social category Krugu, J.K. 166 Kumar, K. 62, 64, 74n2 Kumaratunge, C. 70 Kumari, J. 91 Kyomuhendo, G.B. 10, 165 Kyrgyzstan 42 labour market 23, 100, 123, 124, 136; female labour force participation 147; and income equality 23; returns to higher education 19, 31, 32; and secondary education completion 179; see also Pakistan Lady Health Workers (LHW) 149, 158 Lajpat Rai, L. 62

landownership 9, 62, 64, 73, 126, 129, 137, 138 Langa, M. 166 language 63, 64 – 65, 68, 70, 72, 74, 92, 192; see also mother tongue Lao People’s Democratic Republic 41, 44 Lassibille, G. 20 Latham, M. 219 learning 6, 13, 23, 48, 224 – 225, 238 – 240; achievement, of government schools 86 – 87; CBE 192, 193, 194, 201, 202; and beliefs of teachers, channels linking 229 – 230, 230; and children with disabilities 207, 212, 213, 215, 217, 218 – 220; crisis 98, 100, 207; gaps 13, 22, 86 – 87; girls’ education 92, 98, 100 – 101, 106, 107, 178; minimum level of learning 66; and non-state actors 81; and public– private partnerships 83, 84, 85 – 86, 89, 90, 92, 93; outcomes 2, 4, 10, 21, 37, 48, 49, 52, 80, 83 – 86, 88, 89 – 90, 92, 93, 99, 100 – 101, 105, 126 – 127, 132 – 139, 186, 201, 202, 225, 226, 238 Learning Metrics Task Force 48 ‘leave no-one behind’ 3 Lee, J.W. 25, 26 Leftwich, A. 74n2 Lemma, G. 116n4 Lensick, R. 48 Lesotho 42, 44, 172 Levy, B. 98, 99 Levy, F. 21 Lewin, K. M. 1, 2, 3, 4, 7, 9, 37 – 38, 39, 40, 41, 42, 43, 43, 44, 45, 46, 47, 49, 50, 51, 52, 68, 70 Liberia 41, 44 Little, A. W. 7 – 8, 49, 59, 65, 68, 73, 74, 74n2, 74n12 livelihood, impact on access to education 190, 196, 202, 203 Lopez Calix, J.R. 176 low- and lower-middle income countries (LICs/LMICs) 38; aid to education in 47 – 52; demographic transitions 38 – 39; economic growth in 47; educational financing in 49 – 52; enrolment in 39 – 42, 40, 41 – 42; evolution of LICs as fiscal states 52; GDP allocation to education in

Index  253 50 – 51; out-of-school children 46; patterns of participation by gender 42 – 45, 44; percentage of girls enrolled by grade in 43; and revenue generation 51 – 52; total public expenditure on education in 50 Loxley, W. 22 Lubaale, Y.M.A. 172 Luschei, T.F. 224 McDonald, P. 145 McGillvray, M. 71 McGregor, J.A. 130 McKown, C. 227 McNay, L. 127 Macpherson, I. 3 Macro International Inc. 170 Madagascar 41, 44 Majumdar, M. 219, 227 Malawi 41, 44 Mali 41, 44, 46, 176 Malik, A. 90, 91, 126 Malik, R. 5, 12, 90, 93, 123, 224 Mangez, E. 126 Manor, J. 3, 4, 37 Mansoor, S. 156 Mansuri, G. 229 Mari Bhat, P.N. 146 Marin, A. 23 marketisation in education 3 marriage 101, 102, 131, 134, 139, 169, 185, 199, 203; and girls’ schooling 154, 157 – 158, 199 – 200; and teenage pregnancy 173, 175, 177 – 178; see also Ethiopia; Ghana; Pakistan; Uganda Mason, K.O. 145 mass education 37, 60, 61, 62, 64 maternal schooling, and fertility attitudes/behaviour of women 145 – 147 Mathur, K. 66 Mauritania 41 medium of instruction 63, 64 – 65, 69 middle space 9, 98, 99; see also North, A.; Unterhalter, E. Millennium Development Goals 3, 22, 59, 206 Miller, E. 43 Mincer, J. 20, 23, 24 minimum level of learning 66 Minow, M. 12, 208, 209; see also dilemmas of difference

Mirza, R. 126 Mitchell, R. 5 Mitra, S. 206 Mizunoya, S. 206 Mkwananzi, S. 172 Mohmand, S.K. 126, 240n3 Moldova 42 Mongolia 42 Montenegro, C.E. 19, 24, 25, 26, 27, 27, 28, 29, 30 Mooij, J. 219, 227 Moon, B. 3 Moore, M. 51, 145 Moore, R.J. 62 Morgan Committee 62 Morrow, V. 102 mother tongue 11, 65, 186, 193, 201; see also language Mozambique 41, 44 Muhanguzi, F. K. 5, 10, 165, 174, 178 Mumtaz, Z. 145, 146 Muralidharan, K. 30, 91, 92 Murnane, R.J. 21 Myanmar 41, 44 Naik, J.P. 63 – 64 Naqvi, Z.F. 147 Narayan, K.R. 71 Naseer, M.F. 126, 227 National Action Plan on Gender Equality, Ethiopia 104 National Development Plan, Uganda 167 National Education Commission, Sri Lanka 70 National Learning Assessment, Ethiopia 100 National Literacy Mission, India 69 National Planning Commission, Ethiopia 104, 105 National Policy on Education, India 64, 66, 69 National Policy on Women (1993), Ethiopia 104 National Strategy for Girls Education (NSGE), Uganda 170 National University of Educational Planning and Administration, India 69, 207 Naveed, A. 5, 9, 123, 124, 125, 126, 127, 128, 131, 132, 134, 136, 140, 141n5, 142 Nechyba, T. 92 Nehru, J. 63

254 Index Nepal 41, 44 Neuman, W.L. 211 Nicaragua 42, 44 Niger 41, 44, 46, 176 Nigeria 41, 44, 46, 172, 174 Ninsiima, A.B. 170, 174 non-state education providers 81 – 82, 82, 86; and equity in access to quality education 84; government subsidy to 83; and learning outcomes 83 – 84; see also Ghana; India; Pakistan; private sector schools; passim Chapter 5 Noronha, C. 91 North, A. 3, 9, 98, 99; see also middle space Norwich, B. 208 – 209, 220; see also curriculum dilemma; dilemmas of difference; identification dilemma; location dilemma Ochan, W. 169, 174, 175, 176, 177 Odimegwu, C. 172 Oesterheld, J. 62 Official Languages Act of 1956, Sri Lanka 64 Ogundari, K. 31 Operation Blackboard (OB), India 66 – 67 option value 22, 30, 32 ORC Macro 170 Organisation for Economic Cooperation and Development 51, 224 Osborne, M.A. 32 out-of-school children 7, 11, 13, 31, 79, 123, 133, 134, 136, 166, 189, 190, 218, 232; in Africa 185; and children with disabilities 206; complementary basic education 186, 187, 188, 188, 196; demand factors 185; flexible time frames 11; in LICs/LIMCs 46; and livelihood patterns 194; supply factors 185; and teenage pregnancy 177 – 178; see also Ethiopia; Ghana; India; Pakistan Oyefara, J.L. 172 Pajares, M.F. 228 Pakistan 5, 6, 9 – 10, 41, 44, 87, 94n4, 123 – 124, 144, 160, 229, 230, 240n3, 240n7, 241n11; aid to education for 47; caste 9, 126, 131, 132, 134, 136, 137, 138, 139, 143, 225, 227, 229, 240n3,

240n7; children with disabilities in 79, 234 – 235; conceptualisation of ‘good student’ 233 – 234, 239; contraceptive use in 147 – 148; costs and value of children 155 – 158, 159; cross gender-intergenerational mobility, aspirations for 156 – 158, 159; daughters/daughters-in-law 128, 130 – 139, 146, 151 – 152, 154, 156 – 159; decline in fertility rates in 147 – 148; delivery of education to underserved children 90; desired family size 149, 150 – 151, 150, 155, 159; desired sex composition of children 151 – 152, 159; differentiated teacher practice in 237, 238; dropouts, school 79; educational aspirations and realities 132 – 139; educational strategies 123 – 124; employment 157, 158; evolution of public–private partnership 88 – 89; expectations of teachers 233 – 236; family research 127 – 128; and fertility attitudes/behaviour of women 144, 150 – 153, 159; girls’ schooling in 10, 144 – 145; good women, roles of 154; government/public schools in 86 – 88, 90, 93, 225, 230, 232, 238 – 239; grant-in-aid schools in 87, 88; intergenerational bargain 130; karobar (in Punjab) 129, 131, 134; kinship 9, 126, 131, 132, 134; land ownership 9, 126, 129, 137, 138; marriage 131, 134, 154, 157 – 158; meanings, norms and values of schooling 129 – 131; mothers 9 – 10, 145, 146, 154, 158, 159, 231; navigational capacity 139; out-of-school children in 79, 123; parental background, significance of 234 – 236; parrhalikha (schooled) and the unparrh (without schooling) in Punjab 129; Partnership for Management arrangement 90; patriarchy 9, 127; pedagogy 224 – 225, 227, 228; pentagonal rural social structure 9, 126 – 127, 132, 140; poverty in 93, 227, 229; primary (basic) education in 232; private schools in 79 – 94, 131, 147, 155 – 156, 158; private tuition in 156; public–private partnerships in 8, 79 – 94; relational dynamics of schooling 136; relationship of

Index  255 patronage 9, 126, 127; religion 9, 126 – 127, 130, 134, 139, 141n2; research study 147 – 150, 230 – 232; returns to education 123, 140; rural social structure in Punjab, Pakistan 124 – 127, 128, 139; school enrolment in 86, 93, 148 – 149; secondary education in 147 – 149, 156; social distance 225, 227, 238 – 239; social mobility 9, 124, 125, 126 – 127, 128, 129, 131, 134, 140, 146, 147, 154 – 155, 156 – 157, 159 – 160, 227; social stratification 123; son preference in 146; student attendance and parental occupation 236; taraqqi 9, 128, 129, 130, 131, 132, 134, 135, 136, 137, 139; teachers’ beliefs and teaching practices, link between 229; teaching practices in 12, 224; tensions regarding public–private partnerships 93; transformations in intrahousehold inequalities 153 – 155; typologies of public–private partnerships 89; voucher schemes 85 – 86, 90 – 91; women’s literacy and labour force participation in 147; passim Chapters 5, 7, 8, 12 Pakistan Demographic and Health Survey (PDHS) 148 Pakistan National Institute of Population Studies 148, 148, 150, 150, 153, 153, 154, 160n5 panchayati raj in India 68 Pankhurst, A. 8, 98, 102 parents 11, 67, 126, 154, 156, 159 – 160, 231 – 232; attitudes to teenage pregnancy 172, 173, 175, 176, 177; families and religion 189; mothers’ role 10, 154, 158; parental choice of school 70, 85, 86, 91, 147, 156, 186, 187, 193 – 200, 201, 202; parental education 145, 146, 231 – 232, 234 – 236, 239; parental values 129, 130, 131, 134, 140 – 141; and poverty 134, 172; son preference of 146; see also Ghana; India; Pakistan; Uganda; passim Chapter 10 Park, R. 226 Partnership for Management model 90 Pasha, G. 145 patriarchy 9, 101, 127, 145, 198; and agency of women 146; in relation to girls’ education 101, 127, 146, 166,

174 – 176; in sub-Saharan Africa 166; and teenage pregnancy 11, 174 – 176; see also Ethiopia; Ghana; girls’ education; Pakistan; Uganda; women Patrinos, H. A. 7, 19, 21, 23 – 25, 25, 26, 26, 27 – 28, 27, 28, 29, 30, 31, 82, 83, 85 pedagogy 193 – 196; child-centred pedagogy 11, 66, 67, 201; classroom culture 229, 231; local knowledge 5; teacher–pupil interactions 228 – 229, 237, 238, 239; see also Ghana; India; Pakistan; passim Chapters 10, 11, 12 Peiris, K.M. 68 Perera, W.J. 65 philanthropic schools 84 Population Secretariat, Uganda 170 poverty 1, 3, 4, 6, 9 – 10, 11, 13, 23, 41, 62, 63, 71, 72, 123, 124, 132, 140, 144, 147 – 148, 179, 185, 186, 187, 201, 202, 203, 210, 225, 227, 229, 232; and access to education 37, 38, 64, 69, 71, 100, 219, 220; alleviation 4, 5; capacity to aspire 9, 10, 124, 125, 140; and children with disabilities 219, 220; and dropouts, school 42; and gender 43; and girls’ education 100, 101 – 102; and Global South 1, 4, 13, 219, 144, 219; livelihood patterns 190; and parental education 234 – 236, 239; and parents 134, 172; and patronage 127; and PPPs 84 – 85, 87, 91, 92, 93, 94; and private schools 69, 84, 87, 91, 156; and school dropout/completion rates 41; socio-economic status 65, 69, 80, 87, 107, 131, 167, 172, 208; and teenage pregnancy 10 – 11, 172, 176, 178; wealth 2, 42, 45, 70; see also Ethiopia; Ghana; India; lowand lower-middle income countries (LICs/LMICs); Pakistan; Uganda power 2, 3, 9, 13, 98, 99, 105, 107 – 109, 114, 115, 124, 125, 126 – 127, 129, 139, 140, 145, 189 Prakash, N. 30 Prata, N. 146 Premadasa, R. 68 primary (basic) education 4, 8, 11, 13, 59, 60, 62 – 74, 74n5, 75n13, 94n5, 94n6, 99, 116n1, 141n5, 169, 174, 179, 185 – 203, 206 – 220, 231, 232; and access to jobs 31 – 32; demand

256 Index for 32; enrolment 27, 32, 81, 100, 100, 101; enrolment, in LICs/LIMCs 39, 41, 43, 45; equity impact for 22; expectations of teachers 227; and gender 28, 42, 43, 44, 45; and PPPs 85, 90, 92; quality of 33; returns to 7, 19, 21, 23, 24, 27, 28, 30, 31, 32, 33; universalisation of 3, 7, 31, 32, 49, 50, 59, 60, 62, 66, 70, 72, 74, 85, 203, 232; see also Ethiopia; Ghana; India; Pakistan; Sri Lanka; Uganda; passim Chapters 4, 10 Pritchett, L. 3 private returns to education 21, 24, 25, 27, 28, 123 private schools 65, 69, 71, 80, 81, 82, 83, 84, 86 – 89, 91, 147, 155, 156, 158; contested role of 93; government subsidy to 83, 85; and learning outcomes 86, 92, 93; see also India; Pakistan; Sri Lanka; passim Chapter 5 private tuition 156, 161, 217; see also India; Pakistan privatisation of education 3, 38, 84, 93 Programme of Action (1986), India 66 progressive universalism 22 Promoting Low-Cost Private Schooling in Rural Sindh programme, Pakistan 90 Psacharopoulos, G. 19, 21, 22, 23 – 25, 25, 26, 27 – 28, 31, 32 public–private partnerships (PPPs) 8, 79 – 94; charter schools 8, 83, 84 – 85, 90; contested role of 93; in contexts of crisis and conflicts 85; contracts 83; effectiveness, factors behind 93 – 94; enabling environment for 93 – 94; enhancement of equity 92; government subsidy 83, 85, 89; grant-in-aid schools 87 – 88; impact on learning outcomes 93; Partnership for Management arrangement 90; parental choice of 86; private school fees 80, 81, 84, 86, 89, 91, 92, 93; private schools 80 – 89, 90, 91, 92, 93, 94; privatisation 84, 93; regulation of 84, 90; vouchers 8, 83, 85 – 86, 90 – 92; passim Chapter 5 Public Report on Basic Education in India (PROBE) report 69 Punjab Education Foundation, Pakistan 88, 90 – 91 Putzel, J. 98

Raina, V. 66 Raju, D. 82, 90 Ramachandran, V. 73 Ranaweera, M. 71 Rashtriya Adarsh Vidyalaya schools programme, India 91 Rashtriya Madhyamik Shiksha Abhiyan (RMSA), India 52, 70, 71 Rawal, S. 83, 84, 94n6, 225, 229, 240n2 Ray, D. 125, 138, 140; see also aspiration window Reindal, S.M. 220 relational well-being 125 religion 9, 59, 61, 63, 65, 73, 80, 126 – 127, 130, 134, 139, 167, 187, 189 – 190, 204n12; beliefs, and sexuality education 170, 174; Buddhism 62, 64, 74; Christianity/ Christians 61, 189, 190; Hinduism/ Hindus 62, 74; Islam/Muslims 62, 141n2, 141n5, 189, 190; missionaries 61, 62, 72, 73; see also Ghana; India; Pakistan; Sri Lanka reproduction: arranged marriages 146, 154; early marriage 167, 168, 173, 176, 177; family size 10, 149, 150 – 151, 150, 152, 155, 159, 160, 172; fertility 10, 144 – 160; fertility decline 10, 146, 147, 148 – 149, 150, 160; forced marriage 173, 175, 177 – 178; marriage choices 154, 157; teenage pregnancy 10 – 11, 165 – 180, 196; see also Pakistan; Uganda reproductive autonomy/agency of women 145, 146, 149 Research Consortium on Educational Outcomes and Poverty (RECOUP) 5, 7, 8, 9, 11, 12, 128, 141n4, 148, 149, 160n2 Research for Equitable Access and Learning (REAL) Centre, University of Cambridge 14n3 research methodology: case study research 80, 93, 139, 141, 167, 210; family research 124, 125, 132, 144 – 145, 148; Habitus Listening Guide 125, 132, 141n1; longitudinal research 124; narrative research 125, 210, 211; relational ontology 125, 132; voices of the poor 125, 128, 132

Index  257 Research on Improving Systems of Education (RISE) Ethiopia programme 102 – 103, 116n2, 116n4 Resnick, J. 59 Results for Development (R4D) 83, 84 retention 9, 91, 106, 115 returns to education 7, 19 – 20, 24, 26, 123; causes of pattern changes 31 – 32; changes since 1980 28, 29; earnings and education, relationship between 19 – 24, 27, 30; and educational expansion 23; and gender 25 – 26, 28 – 30; and increasing coverage at lower levels of schooling 31; and laws of supply and demand 23, 31, 32; and level of education 23, 25, 26 – 27, 28, 30, 32; private 21, 24, 25, 27, 28, 123; and quality of schooling 32, 33; by region 24, 28 – 29, 31; social 21, 22; and technological change favouring higher-order skills 7, 19, 31, 32; time trends of 24 – 32; and years of schooling 25 – 26, 25, 26, 27, 27; see also Pakistan; passim Chapter 2 Right of Children to Free and Compulsory Education Act, India 207 Rights of the Persons with Disabilities Act, India 207 Right to Education Act (RTE) 2009, India 69, 87, 88, 91, 93, 94 Rivkin, R. 224 Robinson-Pant, A. 5 role models, for prevention of teenage pregnancy 179 role model theory 240n2 Rose, P. 1, 8, 22, 33, 47, 87, 98, 108, 123 Rosenthal, R. 227 Rosenzweig, M.R. 23 Roy, T. 61 Rwanda 41, 44 Saba, A. 145 Sabates, R. 37 Sachs, J.D. 3 Saeed, S. 146 Saleem, A. 145 Salway, S. 145 Samari, G. 145, 146 Sankar, V. 225 Sao Tome and Principe 42, 44 Sarangapani, P.M. 213, 225, 240n2

Sarkar, S. 224 Sarup, A. 66 Sarva Shiksha Abhiyan programme, India 69, 71, 207, 219 Sathar, Z.A. 145, 146, 147 Sawhney, S. 207 Sayajirao Gaekwad III 62 Sayed, Y. 3 school enrolment 6, 7, 8, 11, 22, 24, 38, 50, 149, 190, 210; in aided schools 88; of children with disabilities 207, 217 – 218; complementary basic education 196, 199, 200, 202; cream-skimming of students 80; deterrents of 203; and earnings 25; and farming activities 190, 196; by gender, in LICs/LMICs 42 – 45; and gender gaps 9, 115; of girls 21, 100, 100, 101, 105, 107, 112, 115; and grant-in-aid schools 62; impact of PPPs in 90; and income 22; in LICs/LMICs 39 – 42, 40, 41 – 42; primary level 27, 32, 39, 42, 43, 65, 67, 68, 72, 81, 100, 100, 101; and private/non-state sector 70, 71, 81, 83, 86; and public–private partnerships 84, 85, 90, 93; secondary level 27, 29, 39, 42, 43, 46, 65, 68, 72, 165; and tensions between older and younger children 200; tertiary level 28; universal 70, 72, 74; and voucher schemes 85; see also Ethiopia; Ghana; India; Pakistan; Uganda School for Life, Ghana 186 Schweisfurth, M. 3 Sebudde, R.K. 177, 178 secondary education 7, 33, 44, 50, 64, 65, 67, 70 – 74, 136, 174, 179; and access to jobs 31 – 32; and affordability 33; enrolment 27, 165; enrolment, in LICs/LIMCs 39, 41, 43, 45, 46; expansion of 33; option value of 30, 32; and reduction in teenage pregnancy 178; returns to 19, 27, 28, 29, 30, 32; sub-Saharan Africa 84; and teenage pregnancy 178, 179; universalisation of 70; see also India; Pakistan; Sri Lanka; Uganda Seers, D. 54 Sekiwunga, R. 171, 173 Semela, T. 104 Sen, A. 2, 74n74, 139; see also capability theory

258 Index Sen, G. 99, 114 Senegal 41, 44 Sergeant Plan 62, 63 Sethuraman, K. 175, 176, 178 sexual and reproductive health (SRH) services 169 – 170, 175 – 176, 178 sexuality education 11, 170, 174, 178, 180 Shafiq, M.N. 22, 30 Shah, R. 218 Shami, M. 127 Shani, O. 63 Sharma, G. 146, 147 Sierra Leone 41, 44 Sindh Education Foundation 90 Singal, N. 1, 5, 11 – 12, 73, 206, 207 – 208, 210, 216, 218, 220 Singh, A. 69 Singh, R. 224 skilled labour 21, 23, 105 Smith, Adam 20 social capital 179; see also Bourdieu, P. social category 226 social class reproduction 226 social distance 6, 12, 224 – 225, 226, 227, 229, 231, 238 – 239, 240; passim Chapter 12 social mobility 9, 37, 124, 125, 127, 128, 129, 131, 134, 140, 146, 147, 154 – 155, 156 – 157, 159 – 160, 227; educational mobility 126, 154, 227; passim Chapter 7 Social Research Institute – IMRB 225 social returns to education 21, 22 social structure 131; caste 3, 6, 9, 61, 62, 63, 69, 70, 72, 73, 75n15, 126, 131, 132, 134, 136, 137, 138, 139, 143, 209, 225, 227, 229, 236, 240n3, 240n6, 240n7; elites 3, 13, 61, 72, 73, 126, 127; ethnicity 4, 65, 72, 79, 189, 190, 227 – 228; interest groups 13, 60, 70, 73, 74; kinship 9, 126, 131, 132, 134; land and landowners 9, 62, 64, 73, 126, 129, 137, 138; patronage 9, 126, 127; pentagonal social structure 9, 126 – 127, 132, 140; power 2, 3, 9, 13, 124, 125, 126 – 127, 129, 139, 140; rural social structure 124 – 127, 128, 139; socio-economic classes/ groups 65, 69, 80, 87, 91, 107, 126, 131, 138, 147, 167, 172, 208; see also poverty; religion

Sointu, E. 125 son’s education 123 – 141, 156, 157, 159; see also Pakistan South Asia 5, 6, 28, 43, 79, 80, 94n5, 140, 145, 146, 160n1, 160n2, 227, 228; see also India; Pakistan; Sri Lanka South Sudan 41, 46, 176 Sri Lanka 6, 7 – 8, 31, 59 – 60, 73 – 74,, 74n1, 74n12, 75n13; from 1990 to present 68, 70 – 72; ’1997 Education Reforms, The’ 70; children of plantation workers 73; colonial rule in 61, 62 – 63, 72; curriculum reforms 67; diversity in 74; drivers and interests shaping 72 – 73; drivers for education expansion (from independence to 1970) 64 – 65; economic instability 72 – 73; and ethnicity 65; fee-free access to education in 63, 65; government/ public schools in 65, 71; instability and struggle for universal education (1970–1990) 67 – 68; insurrections 67 – 68, 72; and language 64 – 65, 70; literacy rate in 59, 62, 63; mass education 61; parish schools 61; political instability 67 – 68, 72 – 73; post primary English-medium schools 63; pre-independence period 61 – 63; Presidential Commission on Youth 68; primary (basic) education in 60, 65, 70 – 72, 74; private schools 65, 70, 71; religion in 59, 61, 65, 73, 74; resistance and promotion by interest groups 73; secondary education in 59, 65, 67, 70, 72; size and population of 60; Tamil Eelam 68; social groups 73; support from external agencies 71 – 72; teachers in 61, 64, 67, 73; universal franchise 62, 72; passim Chapter 4 Srivastava, P. 3, 91, 147, 210 State Council, Sri Lanka 62 – 63 Stavropoulou, M. 102 stigma: attached to special schools 216; and teenage pregnancy 177 sub-Saharan Africa (SSA) 5, 6, 8, 10, 48, 52, 53, 94n6, 102; aid to 51; access to education in 185; child population growth rates in 39; countries, debt distress of 47; demographic transition in 38, 39; gender parity in 165; out-of-school

Index  259 children in 46, 185; patterns of participation by gender in 42 – 43, 45; poor quality education across 186; private and non-state actors in 81; public–private partnerships in 84; returns to schooling in 28; tax avoidance and evasion in 51; teenage pregnancy in 166, 171, 172, 176 – 177, 178, 179; see also Ethiopia; Ghana; Uganda successful globalisation 49 Sucharita, V. 91 Sujatha, K. 91 Sundararaman, V. 91, 92 Sussex University 1, 54n1 Sustainable Development Goals (SDGs) 3, 6, 22, 49, 50, 59, 60, 79, 144, 206 Sutoris, P. 140 symbolic violence 134; see also Bourdieu, P. Tagore, R. 62 Tajikistan 42 Tamil Eelam 68 Tamim, T. 225, 229 Tan, J.P. 20 Tan, X. 146 Tanzania 6, 42, 44, 46, 180n7 taraqqi 9, 128, 129, 130, 131, 132, 134, 135, 136, 137, 139; passim Chapter 7 tax, as source of educational financing 51, 52 teachers 3, 11 – 12, 49, 61 – 62, 64, 67, 87, 88, 186, 195, 196, 202, 224 – 241; of aided schools in India 87, 88; assistant teachers 49; complementary basic education 191, 195, 197, 198, 202 – 203; and caste 126; curriculum, views of 209, 212 – 215; expectations of 227, 229 – 230, 232 – 236, 239; female 173 – 174, 179; head teachers 189, 191, 198, 210 – 211, 213, 217; identification of 209, 211 – 212; Inclusive Educational Resource Teachers (IERTs) 210, 214, 217, 219; interviews with 6, 189, 198, 210 – 212, 213 – 214, 218, 225, 231, 234; location of 209, 215 – 217; private school 86; recruitment 49, 87 – 88; sexual harassment by 173; and social distance 6, 12, 224 – 225,

226, 227, 229, 231, 238 – 239, 240; training 63, 67, 90, 137, 138, 202 – 203, 207, 240; see also Ethiopia; Ghana; India; Pakistan; Sri Lanka; Uganda teachers, beliefs of 224 – 225, 238, 239 – 240; and learning, channels linking 229 – 230, 230; links between children’s background and learning ability 234; parental background, significance of 234 – 236; and social distance 239; and teacher practices 218, 227 – 229; passim Chapter 12 teachers, practices of 12, 210, 213 – 214, 216, 217, 219, 220, 224 – 225, 231; and beliefs 218, 227 – 229, 228; bias in 229, 239 – 240; differentiated 237 – 238; dilemmas of difference 12, 206 – 220; inclusive 230, 237; silent exclusion 237, 239; social distance 224 – 241; student–teacher interactions 237, 238, 239; passim Chapters 11, 12 Teaching Effectively All Children (TEACh) research project 12, 14n6, 230 – 231, 234 technological change favouring higherorder skills, and returns to education 7, 19, 31, 32 teenage pregnancy 10 – 11, 165 – 166, 171, 196; and boys 177 – 178; and education 166; and lack of parental guidance/monitoring 173; and marriage 173; and peer pressure 174; and poverty 172, 178; programmes for reducing 170; role of education in ending 178 – 180; and schools 173 – 174; sexual and reproductive health services 169 – 170, 175 – 176; and sexual violence 174, 175; stigma and shame associated with 177; in sub-Saharan Africa 171; and transactional sex 172, 176; and travel to/from school 174; see also Ghana; Uganda; passim Chapter 9 Tefera, T. 116n4 tertiary education, returns to 25, 26, 28, 31; see also higher education theory of practice 125 – 126; see also Bourdieu, P. Timor Leste 41, 44 Tinbergen, J. 31, 32 Togo 41, 44

260

Index

transactional sex, and teenage pregnancy 172, 176 Tuhiwai Smith, L. 5 Uganda 5, 6, 10 – 11, 41, 44; access to education 5, 6, 165, 172, 177; dropouts, school 174, 177 – 178; government/public schools in 170, 172, 173 – 174, 179; impact of teenage pregnancy on education attainment 176 – 178; lack of parental guidance/monitoring 173; legal age for marriage 169; legislative and policy commitments to address 167 – 171; and moral decay 171, 177; National Legal and Policy Framework 168 – 169; patriarchal social norms and practices 11, 174 – 176; poverty in 172, 176, 178, 179; pressures, risks and powerlessness 171 – 174; primary (basic) education in 174; re-entry into school for teenage mothers 169; research study 166 – 167; role of education in ending teenage pregnancy 178 – 180; school enrolment in 165, 169; secondary education in 165, 178, 179; sex education 10, 166, 167, 170, 173 – 176; teachers in 167, 170, 173 – 174, 177, 179; trends in teenage pregnancy rates 170; value on women’s fertility 174 – 175; passim Chapter 9 Uganda Bureau of Statistics 170, 172, 178 UN Charter of Human Rights 1, 37 UNESCO 1, 3, 8, 13, 37, 43, 46 – 47, 48, 49 – 51, 52, 54n1, 59, 60, 72, 74, 98, 165, 170, 179, 185, 207 UNESCO Institute for Statistics (UIS) 27, 28, 39, 46, 206 UNICEF 42, 43, 54n1, 71, 185, 186, 204n2 United Nations 37, 60, 206 United Nations Convention on the Rights of Persons with Disabilities 206 United Nations Fund for Population Activities 165, 170, 171, 172, 178 United Progressive Alliance, India 69–70 Universal Declaration of Human Rights 206 universalisation of primary education (UPE) 3, 7, 31, 32, 49, 50, 59, 60, 62, 66, 70, 72, 74, 85, 203, 232

university education, expansion of 22 – 23; see also higher education; tertiary education, returns to University of Cambridge 1, 5, 7, 8, 9, 11, 12, 14n3, 128, 141n4, 148, 149, 160n2; see also Research Consortium on Educational Outcomes and Poverty (RECOUP); Research for Equitable Access and Learning (REAL) Unterhalter, E. 2, 3, 9, 98, 99; see also middle space Usher, A.D. 48 Uzbekistan 42 Vasavi, A.R. 229 Vaughan, S. 104 Vawda, A.Y. 23 Verger, A. 3 Verspoor, A. 39 Vietnam 42, 44 Vijaysimha, I. 213 Vision 2040, Uganda 167 Vlassoff, C. 152 voucher schemes 8, 83, 85–86, 89, 90–92 Vyborny, K. 127 Wales, J. 81, 83, 84, 94n6 Walford, G. 3 Walker, J.A. 166 Walker, M. 2 Ward, M. 71 Wardha conference, India 62 Wark, C. 226 Watson, C. 174 Waylen, G. 99, 114 Weiner, M. 74n74 Weinstein, R.S. 227 Weisbrod, B.A. 22, 30 Welch, F. 23 Westbrook, J. 230 White, H. 48 White, S.C. 125 Whyte, S.R. 171, 173 Wijesundara, S. 65 Wikramanayake, D.H. 72 Williams, T.P. 104 Winkvist, A. 146 Wodon, Q. 165, 176, 178, 179 Woldehanna, T. 105 Woldemariam, H. 116n7 Woldetsadik, G. 101 Wole, D. 116n4 Wolf, P.J. 91, 92

Index  261 Women’s Affairs Office, Ethiopia 104 women 21; access to higher education 29; agency of 145 – 146, 154; empowerment 104, 105, 115; power/ representation in education policy domain 107 – 109; reproductive autonomy of 145; returns to education 25 – 26, 26, 28 – 30, 29; status, in patriarchal family systems 145; taraqqi of 130 – 131; see also Ethiopia; fertility attitudes/behaviour of women; gender; girls’ education; Pakistan Wood, A. 14n1 Wood, G. 124, 127 Wood’s Despatch 62 World Bank 3, 22, 23, 31, 52, 71, 72, 98, 147, 185 World Conference on Education for All (WCEFA) 37, 48, 54n1, 59

World Education Forum (WEF) 7, 37, 46, 50 World Health Organisation 171 World Inequality Database on Education 42, 45 Wotipka, C.M. 146, 147 Yemen 41, 44 Yorke, L. 8, 98 Zaidi, B. 147 Zambia 42, 44 Zenawi, M. 103, 105 Zewde, S-W. 115 Zewide, G. 103, 108, 109 Zhang, T. 60 Zimbabwe 42 Zuanna, G.D. 146 Zubairi, A. 33, 47