Social sector in India: issues and challenges 9781443877992, 1443877999, 9781443884839, 1443884839

Citation preview

Social Sector in India

Social Sector in India: Issues and Challenges Edited by

Himanshu Sekhar Rout and Padmaja Mishra

Social Sector in India: Issues and Challenges Edited by Himanshu Sekhar Rout and Padmaja Mishra This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Himanshu Sekhar Rout, Padmaja Mishra and contributors

Department of Analytical and Applied Economics (UGC-Centre of Advanced Study in Economics) Utkal University, Bhubaneswar, Odisha, India

All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-7799-9 ISBN (13): 978-1-4438-7799-2

TABLE OF CONTENTS About the Book.......................................................................................... vii About the Editors...................................................................................... viii Contributors ................................................................................................ ix Preface ........................................................................................................ xi Acknowledgements .................................................................................. xiii Introduction and Overview .......................................................................... 1 Himanshu Sekhar Rout and Padmaja Mishra Chapter One ............................................................................................... 12 Growth, Poverty and the Social Sector in Odisha: An Overview Santosh Chandra Panda Chapter Two .............................................................................................. 25 Efficiency of Social Sector Expenditure in India Brijesh C. Purohit Chapter Three ............................................................................................ 60 Dejected Social Sector Development in India during the Reform Period: Trends, Patterns and Determinants Amit Kumar Basantaray and Anirudha Barik Chapter Four .............................................................................................. 82 Social Sector Expenditures and Economic Growth in India: An Empirical Analysis Anirudha Barik Chapter Five .............................................................................................. 96 Public Expenditure on Healthcare in Assam: Posing Questions for the National Health Bill of India 2009 Rajshree Bedamatta and Nirmala Devi

vi

Table of Contents

Chapter Six .............................................................................................. 115 Utilization Pattern of Outpatient Care and Delivery of Health Care Services in Odisha Sarit Kumar Rout Chapter Seven.......................................................................................... 128 Inter- and Intra-State Disparity in the Health Status of Odisha vis-a-vis India Sonali Chakraborty Chapter Eight ........................................................................................... 143 Intellectual Property Right Protection and Health Sector Innovation: Concerns for India Vani Aggarwal and Jyotsna Chowdhury Chapter Nine............................................................................................ 163 The Threat Posed by Industrial Pollution to Human Health: A Case Study Mrutyunjay Mishra and Nirmal Chandra Sahu Chapter Ten ............................................................................................. 176 The Distant Goal of Universalising Elementary Education: Case Study of a Tribal Village of Odisha Amarendra Das and Himanshu Sekhar Rout Chapter Eleven ........................................................................................ 200 Gender Disparity in Employment in Odisha: Some Evidence from NSS Data Smrutirekha Mohanty Chapter Twelve ....................................................................................... 217 Participation of the Central Government in Rural Development of India: An Analysis of the Union Budget (1981–82 to 2008–09) Animesh Kumar Chapter Thirteen ...................................................................................... 236 The Determinants of Child Labour in India: A Probit Analysis Chandan Kumar Mohanty

ABOUT THE BOOK

This book is an outcome of the national conference on “the Social Sector in India: Issues and Challenges” organized by the Centre of Advanced Studies in Economics, Utkal University, Bhubaneswar, India on March 29–30, 2013. Of the thirty papers presented at the conference, thirteen are included in this volume, after peer review. As education and health are two major areas of concern in the context of social sector development and human development achievements, this book is a humble attempt to look into the situation in India. The liberalisation of the Indian economy has no doubt increased the growth rate of its Gross Domestic Product (GDP). The economic growth of the country jumped from the so-called Hindu growth rate of 3.5% to 8–9% per annum. The literacy rate increased to 74.04% in 2011 from 12% in 1947. The universalization of elementary education has been achieved to a great extent, and dropout rates have decreased. However, despite considerable progress, exclusions and wide disparities still exist. Combining access with affordability and ensuring quality with good governance and adequate finance still remain great concerns. On the health front, significant achievements have been made. Smallpox and guinea worm disease have been eradicated from the country, while polio is on the verge of elimination and leprosy, kala azar and filariasis are expected to be eliminated in the near future. There has been a substantial drop in the Total Fertility Rate (TFR) and Infant Mortality Rate (IMR). Life expectancy has gone up from 36.7 years in 1951 to 67.14 in 2011. The infant mortality rate is down from 146 in 1951 to 46.7 in 2012. Crude birth rate has been reduced from 40.8 in 1951 to 20.6 in 2012, and the crude death rate from 25.1 to 7.43 in the same period. These achievements are impressive, but at the same time our failures appear even more glaring. In this connection, the present volume brings together research papers on different social issues such as linkages between growth, poverty and the social sector; the efficiency of social sector spending in India; the disparity in health statuses; IPR protection in health innovations; pollution and health; the universalisation of elementary education; problems faced at the higher levels of education; and issues of child labour. Eminent scholars from institutes of national repute have contributed to and enriched this volume.

ABOUT THE EDITORS

Himanshu Sekhar Rout PhD is a Reader in the Department of Analytical and Applied Economics, Utkal University, Bhubaneswar and Deputy Co-ordinator at the Centre of Advanced Study (CAS) in Economics of the University Grants Commission (Government of India) Special Assistance Programme (SAP) in the same department. He has 15 years of teaching and research experience, and has eight books and 36 research papers published in national and international journals to his credit. Dr. Rout is Editor-in-Chief of the biannual journal Manthan: Journal of Commerce and Management and is a Member of the Editorial Board of the journals Utkal Economic Papers and the International Journal of Interdisciplinary Current Researches. His teaching domain includes Macroeconomics, Public Economics, Health Economics, the Economics of Education, Indian Economic Planning and Development, and Research Methodology. His main areas of research are Economics of Social Sector (Health, Education, and Gender) and IPR (GIs). Padmaja Mishra PhD is a Professor in the Department of Analytical and Applied Economics, Utkal University, Bhubaneswar and Coordinator, Centre of Advanced Study (CAS) in Economics in the same department. She has more than 25 years of teaching and research experience. She has been the recipient of international fellowships including the Commonwealth Academic Staff scholarship. She has many books and research papers published in national and international journals to her credit. Professor Mishra is the Editor-in-Chief of the journal Utkal Economic Papers. Her teaching domain includes Environmental Economics, Public Economic and Agricultural Economics, and her main areas of research are Environmental Economics, Agricultural Economics and Gender Studies.

CONTRIBUTORS

Name Amarendra Das

Address Department of Analytical and Applied Economics, Utkal University, Vani Vihar, Bhubaneswar – 751 004, Odisha, India

:

Amit Kumar Basantaray

:

Department of Economics and Public Policy, Central University of Himachal Pradesh, Dharamshala, Himachal Pradesh – 176215, India

Animesh Kumar

:

Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi, India

Anirudha Barik

:

Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi, India

Brijesh C. Purohit

:

Madras School of Economics, Behind Govt. Data Centre, Gandhi Mandapam Road, Kottur-600025, Chennai, India

Chandan Kumar Mohanty

:

Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi

Himanshu Sekhar Rout

:

Department of Analytical and Applied Economics, Utkal University, Vani Vihar, Bhubaneswar – 751 004, Odisha, India

Jyotsna Chowdhury

:

Centre for WTO Studies, Indian Institute of Foreign Trade, New Delhi, India

x

Contributors

Mrutyunjaya Mishra

:

Department of Economics, Banaras Hindu University, Varanasi-221005, Uttar Pradesh, India

Nirmal Chandra Sahu

:

Department of Economics, Berhampur University, Berhampur, Ganjam, Odisha, India

Nirmala Devi

:

Department of Humanities and Social Sciences, Indian Institute of Technology Guwahati, India

Padmaja Mishra

:

Department of Analytical and Applied Economics, Utkal University, Vani Vihar, Bhubaneswar – 751 004, Odisha

Rajshree Bedamatta

:

Department of Humanities and Social Sciences, Indian Institute of Technology Guwahati, India

Santosh Chandra Panda

:

Professor and Dean Faculty of Economics, South Asian University, Akbar Bhawan, Chanakyapuri, New Delhi

Sarit Kumar Rout

:

Public Health Foundation of India (PHFI), New Delhi, India

Smrutirekha Mohanty

:

Centre for the Study of Regional Development Jawaharlal Nehru University New Delhi, India

Sonali Chakraborty

:

School of Management, KIIT University, Bhubaneswar, India

Vani Aggarwal

:

Centre for WTO Studies, Indian Institute of Foreign Trade, New Delhi, India

PREFACE

The liberalisation of the Indian economy has no doubt increased the growth rate of its Gross Domestic Product (GDP). The economic growth of the country jumped from the so-called Hindu growth rate of 3.5% to 8.9% per annum. But how has India fared with respect to social sector achievements? Being a consistent low rank holder as per the human development index values constructed by the UNDP, it seems that social sector developments have not received due attention in the growth process of India. No doubt, improvements have been made. The literacy rate increased to 74.04% in 2011 from a meagre 12% in 1947. Enrolments have increased and dropout rates have declined. Concerted efforts have been made for the universalisation of elementary education and expansion of access to secondary and higher levels of education. But, expansions have not been matched by equity and excellence. Similarly, in the health scenario, smallpox and guinea worm disease have been eradicated from the country; polio is on the verge of elimination, while leprosy, kala-azar and filariasis are expected to be eliminated in the near future. Life expectancy has gone up from 36.7 years in 1951 to 67.14 in 2011. There has been a substantial drop in the Total Fertility Rate (TFR) and Infant Mortality Rate (IMR). Infant mortality rate is down from 146 in 1951 to 46.7 in 2012. The crude birth rate has been reduced from 40.8 in 1951 to 20.6 in 2012 and the crude death rate from 25.1 to 7.43 in the same period. Unfortunately, these achievements have been accompanied by glaring failures. According to the NFHS III of 2005–6, 48% of children under five were stunted and 23.7% were severely stunted, 19.8% were wasted and 6.4% were severely wasted, and 42.5% were underweight and 15.8% were severely underweight. At present, India has the largest prevalence of TB, accounting for one fifth of the global incidence, and contributes 77% of the total number of malaria patients in Southeast Asia. The financing of education and health in a country like India is a big challenge. India lags far behind the other BRIC countries in social sector achievements due to its low spending on education and health. According to a study by the Associated Chambers of Commerce and Industry of India, the Bhore Committee recommended that India should spend 6% of its GDP / 15% of its total expenditure on health. However, in 2011–12 it

xii

Preface

spent only 1.3% of its GDP and 4.8% of its total expenditure on these sectors. In this connection, the present volume has brought together research papers on various social issues, such as: how growth, poverty and the social sector are linked; the efficiency of social sector spending in India; the disparity in health statuses; IPR protection in health innovations; pollution and health; the education sector; and issues of child labour. Eminent scholars from institutes of national repute have contributed research papers for this volume. We hope that the volume will provide insights for academics, researchers and policy makers. HSR and PM

ACKNOWLEDGEMENTS

We express our deep sense of gratitude to all who presented their papers at the National Conference on the Social Sector in India: Issues and Challenges, organized by the Centre of Advanced Study in Economics, Utkal University, Bhubaneswar, India on March 29–30, 2013. We are extremely thankful to the authors for their meticulous revision and rerevision of their papers after these were peer reviewed, and for their patience in waiting for more than two years to see this book published. We are indebted to Professor Jatin Kumar Nayak, Dr. Mitali Chinara, Moumita Deroy and Graham Clarke for editing the chapters. We express our warm appreciation to our colleagues and ministerial staff of the department for their kind cooperation. We would like to express our sincere thanks to Cambridge Scholars Publishing Limited, United Kingdom, for accepting our proposal to publish this volume and extending active cooperation at every stage to bring it to its present form. HSR and PM

INTRODUCTION AND OVERVIEW HIMANSHU SEKHAR ROUT AND PADMAJA MISHRA

I. Introduction The main objective of this book is to acquaint the readers with the social sector (mainly the education and health sectors) in India. The social sector is usually defined as dealing with social and economic activities carried out for the purposes of benefiting society, and in the main nonprofit, not-for-profit, philanthropic and mission based and nongovernmental organizations are associated with this sector. However, in the context of this book, the social sector is considered under “social services” in Indian budgets. The focus is on education and health, as these are two major components and have wider positive externalities for other sectors of the economy and society as a whole. They help increase the overall expansion of the economy by enhancing productivity and output, and may be funded by private or public agencies. Indian plan documents also discuss health and education under the social sector, expressing a lot of concern. Given India’s demographic dividends, with a larger amount of young people in the productive age group, education and health assume great significance on account of their real contribution to production by ensuring rapid and inclusive growth. India is on the brink of a demographic revolution with the proportion of the working-age population between 15 and 59 likely to increase from approximately 58% in 2001 to more than 64% by 2021, adding approximately 63.5 million new entrants to the working age group between 2011 and 2016, the bulk of whom will be in the relatively younger age group of 20–35 (GoI 2013). The country report also explains that demographic dividends can be reaped only if this young population is healthy, educated and skilled. To ensure this, a large investment in the education and health sectors by the central and state governments is essential. Existing capacity and resources are inadequate, particularly for

2

Introduction and Overview

education and health (Murthy, Hirway, Panchmukhi & Satia 1990). Substantial enhancement in budgetary support for the development of the social sector needs to be addressed by the government on a priority basis. India’s social service provisions are seldom used by the poor. The health and education of the poor have improved but not for the population as a whole. Physical access to education and health services has improved but inequalities exist because of inadequacies and biases in locating the facilities. The access of the poor to housing, social security and social welfare services has been limited because these services are in adequate and inappropriate relating to their needs, and due to the services leak to the non-poor. To track the social development in the decade of economic reforms from 1990–1 to 1999–2000, Ray (2008) considered thirteen social indicators of India and constructed the country’s social development index (SDI) as a certain weighted average of the selected indicators. The movement of SDI in the 1990s has been compared with the movements of India’s per capita income and plan expenditure on the social sector. The results show that, though there has been growth in the social sector, it is not encouraging and more perhaps needs to be done in the social sector (Ray 2008). Focusing on the impact of economic reforms in the social sector in India by comparing the data of the pre-reform and reform periods, Panchamukhi (2000) notices a declining trend in the budgetary allocations of both the central and state governments for various sub-sectors of the social sector, especially health and education.

II. Social Sector in India The performance of the social sector is far from satisfactory in India and could have been much better (Dreze & Sen 1995). India’s new economic policy has been quite successful in creating a favourable environment for rapid economic growth, but adequate attention has not been paid to basic health care and education. The education and health status of a vast majority of the population continues to remain poor even after a decade of reforms (Joshi 2006). India has made noticeable improvements in key social indicators on education and health since the 1980s, but even so the country lies at the bottom of the ladder of human development with a HDI rank of 136 of 187 countries (UNDP 2013) at 0.554, which is below the South Asian average of 0.0558. Though the HDI value is improving for India, the rank remains more or less stagnant, and neighbouring countries like China and Sri Lanka do far better.

Social Sector in India: Issues and Challenges

3

II.1 Education Sector The role of education in overall social and economic progress is widely recognized. The right to education has been enshrined as a fundamental right in the Constitution of India, which states that: “the State shall provide free and compulsory education to all children aged six to fourteen years in such a manner as the state may, by law, determine.” The literacy rate in India has been constantly rising, improving from 64.8% in the 2001 census to 74.04% in the 2011 census. Both the central and the state governments have been paying increased attention to the need to provide “education for all.” The Indian government has placed lots of emphasis on primary or elementary education. The Right of Children to Free and Compulsory Education Act became operative on April 1, 2010, since which enrolments have increased to become near universal and dropout rates have decreased. Despite this, the provision of quality education with value addition still remains a distant dream. Secondary education covers children aged 14–18, and provides for more than ninety million children. The SSA has been extended to secondary education in the form of Rastriya Madhyamik Shiksha Abhiyan, with special emphasis on the inclusion of disadvantaged sections and profession-based vocational training. Despite the higher levels of enrolment at all levels of education, actual value addition has been unsatisfying, as revealed by poor learning outcomes. Evidence suggests that learning trajectories for Indian school children are almost flat and are far below the corresponding class levels in other comparable countries. Dropout rates at secondary and higher levels remain high and much higher for socially and economically marginalized groups. The complex nature of the problems of poor quality, inequality and exclusion poses challenges to the Indian education system. In this context, the importance of infrastructure, institutional framework and the governance systems cannot be exaggerated. Higher education in India is the third largest in the world after China and the United States. By 2030, India will be among the youngest nations and the need for higher education will be even more urgent. Unfortunately, access to education beyond higher secondary schooling in India is a mere 10% among the university-age population, with huge regional and social disparities (US-India Policy Institute 2014). The system has many issues of concern at present, such as access, equity, relevance and quality. The problems of financing and management are also to be dealt with. All these issues are of very crucial concern as the country seeks to use higher and technical education as a powerful tool to build a knowledge-based economy of the twenty-first century (UGC 2003).

4

Introduction and Overview

Recognising the constraints and need for greater government attention, public spending on education has increased and education expenditure as a percentage of GDP has increased from 3.3% in 2004–05 to over 4% in 2012–13. Education is a subject in the concurrent list, and both central and state governments have their own responsibilities. The central government spending grew at a rate of 25% per year, and the spending incurred by the state governments grew at 19.6% per year during the 11th plan. While about 43% of the total was spent on elementary education, 25% was spent on secondary education and 32% on higher education. For elementary education, the focus remains on the government institutions. The potential for innovative partnerships between the public and the private sectors is explored in secondary education. Gross enrolment at the secondary level in India, though close to the average for all developing countries, remains substantially lower than that of emerging countries like China, Brazil, Indonesia and Thailand. Combined with wide regional and social disparities, it causes considerable concern. Similarly, less than 1/5th of the eligible 120 million students are enrolled in higher education (well below the world average of 26%), and skill acquisition is far less. The “three Es”—expansion, equity and excellence—are to be achieved and the whole system is to be guided by these objectives. Institutions, governance and finance are the major areas in need of revamping and the papers in this volume throw some light on these issues. There is a broad range of challenges facing the education sector in India. Given the inadequate infrastructure, poor learning outcomes, wide variations across states and social and economic categories, the four main priority areas in our education planning are access, equity, quality and governance, and the papers included in this volume discuss these issues.

II.2 Health Sector There is no doubt that India has achieved a good deal in the health sector during the last sixty-five years. The Government of India’s (GOI) initiatives in the public health sector have recorded some noteworthy successes over time—smallpox and guinea worm disease have been eradicated from the country; polio is on the verge of being eradicated; leprosy, kala azar and filariasis can be expected to be eliminated in the foreseeable future. There has been a substantial drop in the total fertility rate and infant mortality rate. These achievements are very impressive and we have every reason to be proud of them, but it is also clear that our failures are even more glaring.

Social Sector in India: Issues and Challenges

5

To respond efficiently to the health needs of the people, particularly the poor and under-privileged, we had hoped to provide “Health for All by the year 2000 AD” through the universal provision of comprehensive primary health care services. In retrospect, it is observed that the financial resources and public health administrative capacity, which it was possible to marshal, were far short of that necessary to achieve such an ambitious and holistic goal. Second, the present health care system in India provides services mostly to the urban populace. Third, the health care system is still weighted in favour of curative programmes in spite of the clear conviction that, in our present situation, it is the preventive, socio-economic and educational aspects that are most significant. Fourth, access to, and benefits from, the public health system have been very uneven between the better-endowed and the more vulnerable sections of society. This is particularly true for women, children and the socially disadvantaged. Fifth, in India, present-day doctors are commercial physicians, motivated by the desire for profit. Sixth, the multiple systems of health care services in India—allopathy, Ayurveda, homoeopathy, unani, siddha; various types of ownership patterns, public (central and state governments, municipal and panchayat local government), private (for-profit and not-for-profit); and different kinds of delivery structures such as teaching hospitals, secondary level hospitals, first-level referral hospitals (CHCs or rural hospitals), dispensaries, PHCs, sub centres, health posts, and occupational groups such as ESIS, defence, CGHS, posts and telegraphs, railways and mines— have resulted in a complex plurality that makes the development of an organized system difficult. Seventh, inappropriate policies, a poor governance structure and inadequate financial arrangements in the supply side, and a high incidence of poverty, ignorance and traditional practices and cultural factors operating in the demand side, influence health outcomes in India (Rout & Panda 2007). India also bears a disproportionately heavy burden of the world’s diseases. Measured in terms of healthy years lost to illness, the World Health Organization estimates that Indians, who make up 17% of the world’s population, suffer 28% of the world’s total number of years lost to respiratory infections, 25% to tuberculosis, 24% to diarrheal diseases, 21% to measles, and 45% to leprosy. In addition, 2.5 million people in India are living with HIV/AIDS (WB 2007). The Government of India has recognized that significant improvements can be made in the health sector. Public health spending constitutes only 1% of the GDP, placing India below most low-income countries and at the bottom 20% of all countries. Nearly 80% of India’s health spending comes

6

Introduction and Overview

from individuals’ out-of-pocket payments at the point of service. Private funds finance 93% of all hospitals, 64% of all hospital beds and 80% of all doctors nationwide. Being unable to pay, the poor are shunted to inadequately funded public hospitals or refrain from seeking medical care. As a result, the poorest 20% of Indians suffer mortality and malnutrition rates more than twice those suffered by the wealthiest 20% (WB 2007).

III. Overview of the Volume The present volume brings together scholarly articles on important issues in the social sector in India presented at the national conference on the Social Sector in India: Issues and Challenges held on March 29–30, 2013 under the auspices of the Centre of Advanced Study in Economics, Utkal University, Bhubaneswar, India. The volume comprises thirteen chapters highlighting the major issues and challenges arising from the social sector in India. Growth, Poverty and Social Sector in Odisha—An Overview is an important contribution by Santosh Chandra Panda, focusing on the change in educational and health outcomes in the poor but rapidly developing state of Odisha, particularly during the last decade. The outcomes in the field of education and health must have been impacted by higher growth and a reduction in poverty. In his view, though significant improvements have taken place in education and health in the last ten years, the progress is still far from satisfactory. There is enough scope for expanding higher education in the state but it has to invest in human resources to enhance the quality in education. Health facilities have also been improved but remain unsatisfactory. The state has to increase public expenditure on health to create a strong health provisioning infrastructure base. For covering expenses on major illness and for helping the poor, appropriately designed health insurance schemes can be provided and the state can actively participate in the process. The Efficiency of Social Sector Expenditure in India by Brijesh C. Purohit analyses the social sector efficiency focusing on two major aspects, namely health and education. His analysis, covering the major states in India, uses both parametric and non-parametric approaches. Besides the individual merits of each approach, although both provide benchmarks to judge relative efficiency across states, the former provides a yardstick more at an aggregative level without parametric restrictions whereas the latter is used to focus on the health care sector. Thus, using Free Disposal Hull (FDH) analysis as well as the stochastic frontier model, Purohit evaluates the performance of the social sector. His results indicate

Social Sector in India: Issues and Challenges

7

considerable state-level disparities. Despite differing budgetary expenditures across states, the outcomes in health and education could be substantially improved through a mix of strategies involving reallocations within the sector and mobilizing additional resources through enhanced budgetary emphasis or encouraging more private sector participation. This may enhance efficiency and increase availability and equity across lowperforming and poorer states in general. In Dejected Social Sector Development in India during the Reform Period—Trends, Patterns and Determinants, Amit Kumar Basantaray and Anirudha Barik examine India’s social sector spending priorities since the reform period to find whether the far-reaching economic reforms that began in 1991 had any significant impact on the level and trend of these social development expenditures, and what the factors behind the ensuing changes were. This paper therefore analyses trends and patterns in expenditure by the central government through annual budgets, expenditure of all states and combined expenditure of the centre and the states. A recent revamp in the social sector expenditure observed may be partly due to the fiscal stimulus given to the economy to enable it to emerge from the global economic recession. Within the social sector, education, health and rural development have remained the major expenditures. At the state level, per capita social sector expenditure has increased in all the states over time, and special category states account for a higher per capita social sector expenditure as compared with most of states in the general category. Although the per capita expenditure has increased over time, by international standards India’s social sector spending in crucial areas like education and health is far below that of other countries. Adopting a simple utility maximization framework, this study reveals that fiscal capacity represented by own-source revenue positively stimulates efficiency and became a major determinant of social sector expenditures during the reform period. In his paper Social Sector Expenditures and Economic Growth in India—An Empirical Analysis, Anirudha Barik investigates the growth effects of social sector expenditure in India over the period from 1990 to 2011. He disentangles social sector expenditures, and takes three sectors into account: education, health and rural development. Johansen & Juselius multivariate co-integration analysis has been used to examine the effect of each sector on economic growth. Barik finds that government spending on education and health has a positive effect on growth, both in the long and short-terms. Spending on rural development is also positively and significantly related to growth in the long and short-terms.

8

Introduction and Overview

The paper Public Expenditure on Healthcare in Assam— Questioning the National Health Bill of India 2009 by Rajshree Bedamatta and Nirmala Devi looks at the pattern of public health expenditure of Assam for the period 1996–97 to 2011–12. Analysing the pattern of public expenditure on health emerging during this period, accompanied by Assam’s health outcome indicators, they assess the implications of the policy environment visualised by the National Health Bill 2009. Considering that Assam’s revenue expenditure on the health sector is very high, spending on infrastructure and delivery systems has been declining for a very long time. The need of the hour is therefore to increase government attention on the health sector. The public-private partnership mode encouraged by the National Health Bill or the insurancebased systems of healthcare will not be able to handle the problems faced by the health sector in the state. In The Utilization Pattern of Outpatient Care and Delivery of Health Care Services in Odisha, Sarit Kumar Rout examines the socioeconomic characteristics of patients, the causes and costs of hospitalization, and looks into patients’ perceptions of the quality of health care in select healthcare institutions in two districts of Odisha. Overall, people from the low socio-economic strata, especially scheduled castes (SC) and scheduled tribes (ST), utilized public health care institutions more than people belonging to the higher castes. In spite of the difference in socio-economic status, the proportion of SC in the public hospitals was more than their share in the total population of the district. It was observed that patients incurred huge out-of-pocket expenditure in the public hospitals. The expenditure on medicine constituted three-fourths of the total out-of-pocket expenditure. In addition to spending on medicine, patients spent money on diagnostic tests and made informal payments to the doctors. The variations in expenditure are quite sharp in spite of the availability of similar facilities in the health care institutions of both districts. Around 47% of the patients had to borrow money in order to meet hospital expenses. This indicates that public hospitals, even when utilized by a majority of the poor, could not provide financial risk protection to a great number of people. Sonali Chakraborty, in Inter- and Intra-State Disparity in the Health Status of Odisha vis-a-vis India, focuses on the development of the healthcare system in Odisha, where in inter-and intra-district disparities persist. The people of Odisha suffer from multiple diseases and the death rate is the highest (nine per 1,000 people) in comparison to other states.

Social Sector in India: Issues and Challenges

9

Intellectual Property Protection and Health Innovation—Concerns for India by Vani Aggarwal and Jyotsna Chowdhury examines the issues of health innovation in the pre- and post-TRIPS eras. The study endeavours to balance both the legal and economic aspects of health innovations and the patent regime in India. Further, the research and development expenditure of the top 20 domestic pharmaceutical industries is analysed to capture the impact of the transition from process patent to product patent on health innovations. In developing countries like India, rising healthcare costs are a cause of concern for a government already burdened with high fiscal deficits. The regime change in the patent system has also not been very supportive with respect to ensuring access to medicines. It is estimated that Indian pharmaceutical companies would get the necessary incentives to invest more in innovation; however, the results arrived at through a thorough data analysis present a different picture. A stricter patent regime has appeared to neither help health innovation nor provide access to medicines at economically viable prices. Although India has tried its best to incorporate all the flexibilities provided under TRIPS in its patent laws, Indian policymakers need to continue this flexible patent regime and not give into the pressure of the developed countries to strengthen its patent laws. In The Threat Posed by Industrial Pollution to Human Health—A Case Study, Mrutyunjaya Mishra and Nirmal Chandra Sahu demonstrate the calculation of air pollution-related (morbidity) health costs in AngulTalcher in Odisha, India. The study makes use of the damage function approach employing the dose-response technique, and the economic value associated with air pollution health damage is calculated using the cost of illness method based on a household survey. The study concludes that the annual health loss due to air pollution amounts to `1.06 million, which is approximately 12% of the per capita income of the people. Amarendra Das and Himanshu Sekhar Rout, in the Distant Goal of Universalising Elementary Education—a Case Study of a Tribal Village of Odisha, attempt to answer the question of whether the Right to Education Act 2009 of the Government of India has been successful in ensuring the rights of every school-going child. In this regard, they undertook a study in a tribal hamlet 20 kilometres from the capital city Bhubaneswar in Odisha. The survey of 173 households in the village reveals the unrealised goals of Sarva Siksha Abhiyan. In the surveyed village, 22.5% of the children aged 6–14 were found to be illiterate, and 40.32% of their parents opined that poverty acts as a major barrier to their education. Fear of education remains a major factor for school dropout. Parents’ lack of awareness of the importance of education and

10

Introduction and Overview

transportation facilities is also a reason for 14.52% and 9.68% of drop outs, respectively. On the supply side, the researchers found that the nonavailability of teachers, classrooms and hostel rooms comprise a major barrier to providing quality education. Similarly, insufficient sanitary facilities also raise the issue of children’s dignity. It is observed that tribal students perform well in sports, but due to a lack of sports teachers their potential is not realised. In her paper Gender Disparity in Employment in Odisha—Some Evidence from NSS Data, Smrutirekha Mohanty brings forth the determinants of female workforce participation in Odisha and analyses gender disparity in employment. It is found that caste, religion, marital status and education are significant variables in defining the employment status of a woman. It has been found that ST women are most likely to be employed compared to all others. Likewise, Hindu women are less likely to participate in the workforce. In comparison with rural women, urban women are more likely to be unemployed in Odisha. Analysis also reveals that women workers are paid less than their counterparts for the same type of work in regular salaried jobs and the wage differentiation is even more pronounced in case of casual jobs. In Participation of the Central Government in the Rural Development of India—an Analysis of the Union Budget, 1981–2 to 2008–9, Animesh Kumar seeks to find the pattern of central government’s budgetary allocations to the rural sector to examine changes in the government’s priorities and spending over time. Given the importance of the government’s expenditure in fulfilling social development and poverty alleviation objectives, the study analyses the central government budgetary expenditure in India during the period 1981–2 to 2008–9, with special reference to the rural sector. He concludes that, while there has been some improvement in recent years, the rural sector has largely been ignored during the post-reform period. There is a consensus that investment in the education of children will enhance the productive capacity of the economy and guarantee a better future for society. Unless checked in time, the high incidence of child labour will jeopardize the functioning of society, engulfing the economy in a low human capital-low economic development trap. Thus, it is necessary to address the dynamic factors of child labour at the macro level and more intensively at the micro (household) level. In this, Chandan Kumar Mohanty, in Determinants of Child Labour in India—A Probit Analysis, highlights the incidence of child labour in India and identifies the micro household level factors that determine the probability of a child participating in the labour force. The results reveal that, although many of

Social Sector in India: Issues and Challenges

11

the individual variables in the regression model like age and gender explain the probability of a child participating in the labour force, the poverty profiles of the household and the education level of its head are important. The study confirms that the education of the head of the household is the most important variable in explaining the participation probability of the child in the labour market. Hence, the policy to eradicate child labour lies in imparting education to the parents and elders and calls for changes in the broader societal attitudes.

References Dreze, J. & A. Sen. 1995. Economic Development and Social Opportunities. Oxford University Press. New Delhi. Government of India. 2013. SAARC Development Goals: India Country Report 2013. Central Statistics Office, Ministry of Statistics and Programme Implementation, New Delhi Joshi, S. 2006. Impact of Economic Reforms on Social Sector Expenditure in India. Economic and Political Weekly. 412:358–365. Murthy, N. I. Hirway, P. Panchmukhi & J. Satia. 1990. How Well do India’s Social Service Programs serve the Poor? World Bank Policy Research Working Paper WPS491. World Bank, New Delhi. Panchamukhi, P. R. 2000. Social Impact of Economic Reforms in India: A Critical Appraisal. Economic and Political Weekly. 3510:836–847. Ray, A. K. 2008. India’s Social Development in a Decade of Reforms: 1990–91/1999–2000. Social Indicators Research. 873:409–425. Rout, H. S. & P. K. Panda. 2007. Introduction and Overview. In Health Economics in India, edited by H. S. Rout and P. K. Panda, 1–12. New Century Publications. New Delhi. University Grants Commission. 2003. Higher Education in India: Issues, Concerns and New Directions, http://www.ugc.ac.in/oldpdf/pub/he/heindia.pdf. UNDP. 2013. Human Development Report 2013—The Rise of the South: Human Progress in a Diverse World. United Nations Development Programme, New York. World Bank. 2007. Detailed Implementation Review: India Health Sector, 2006–2007, Volume I. Department of Institutional Integrity, World Bank, Washington DC.

CHAPTER ONE GROWTH, POVERTY AND THE SOCIAL SECTOR IN ODISHA: AN OVERVIEW SANTOSH CHANDRA PANDA

I. Introduction The last ten years have been an exciting phase for Odisha’s economy. Since 2003–4, the state has been growing at an average rate of nearly 8%, which is higher than the national average. This also led to a significant reduction in poverty between 2004–5 and 2009–10. Going by the 61st round (2004–5) and 66th round (2009–10) of consumer expenditure data, and using the Tendulkar Committee estimates of the poverty line, the Head Count Ratio for Rural Poverty in Odisha declined from 60.81% to 39.19%. In urban Odisha, it declined from 37.58% to 25.91%. The decline in poverty in Odisha is higher than the decline in poverty for all of India. The per capita income in the state has also gone up, though it still trails behind the national average. All this must have an impact on the social sector, particularly education and health. The purpose of this paper is to focus on the change in educational and health outcomes in the state, particularly in the last ten years, which higher growth and reduction in poverty must have affected. What we find is that, though significant improvements have happened in education and health in the last ten years, the progress is still far from satisfactory. In section II below we summarize our findings on growth and poverty. Section III looks at education and section IV evaluates the state of health care. We conclude in section V.

Growth, Poverty and the Social Sector in Odisha: An Overview

13

II. Growth and Poverty We look at the growth rates of GSDP from 1981–2 to 2012–13. To keep the data more manageable we have reproduced the growth rates from 1998–9 to 2012–13 in Table1.1 below. If we look at the growth rates before 2003–4, they are pretty low and fluctuating. An upsurge in growth happens in 2003–4 and is more or less sustained. In 2012–13, the growth rate is 9.14, which is well above the national average. There are several factors that might have contributed to this turnaround in growth in 2003–4. The state government started implementing the liberalized policy of industrialization around this time, and several mine-based industries in steel, aluminium and ferromanganese were established. Secondly, the state followed a path of fiscal discipline and became a surplus revenue state in 2003–4. This would have curtailed resources being used for debt financing and would have augmented growth. Thirdly, the disaster management policy adopted by the state around this time mitigated the negative impact of natural disasters on economic growth. Table 1.1. Growth Rate of GSDP Year 1998–9 1999–2000 2000–1 2001–2 2002–3 2003–4 2004–5 2005–6 2006–7 2007–8 2008–9 2009–10 2010–11 2011–12 2012–13

Growth Rate at 2004–5 prices 2.84 8.59 -1.72 4.82 -0.08 12.84 12.82 5.68 12.85 10.94 7.75 4.55 7.5 4.92 9.14

Source: Odisha Economic Survey (2012–13).

Chapterr One

14

Whetherr the turnarounnd in growth in i 2003–4 wass significant or o not, we ran a stattistical signiificance testt. The equaation estimaated was wheere Dt is a duummy variab ble taking value 0untill 2002–3, andd value 1 afterr that. We fouund that the change c in the rate of ggrowth is statisstically signifficant at the 5% % level of sig gnificance using Neweyy-West standaard errors. The summary off results is preesented in Table 1.2 beelow. Table 1.2. S Summary of Results—Gro R owth Rate of GSDP Dependent Variable

1981 1–2 to 2002– 3 (ȕ ȕ)

20033–4to 20122-13 (ȕ + +ȕ )

Ch hange in Rate R of Gro owth (ȕ )

3.4 4 3.38

7.888 8.114

4.49 4.75

0

GSDP GSDP 3 yeaar moving aveerage

0

1

1

Source: Calcuulated by the Auuthor.

Next, w we look at thhe per capita real NSDP of the state. This is presented inn Table 1.3 bellow. Table 1.3. Per Capita Real R NSDP of o Odisha 20004–5 to 201 12–13 (at 2004–5 pricces) Year 2004–5 2005–6 2006–7 2007–8 2008–9 2009–10 2010–11 2011–12 2012–13

Per Cap pita Real NSD DP (O) 177650 18194 200194 21640 222963 222846 23875 244134 25584

All India P Per Capita NSD DP (I) 24143 26015 28067 30332 31754 33901 36342 38037 39143

O//I x 100 73% 70% 71% 71% 72% 67% 65% 63% 65%

Source: Centrral Statistical Organization. O

What wee find is that the t rate of gro owth in per caapita income in i Odisha in the last yeear is 6.01% as a against the national figurre of 2.91%. However, H the per capiita income inn Odisha is on nly 65% of tthe national per p capita income and has steadily declined d from m 2004–5. Odiisha remains the t fourth poorest statee in terms of per p capita inco ome, and onlyy Assam, UP and a Bihar

Growth, Poverty and the Social Sector in Odisha: An Overview

15

are below it. So, even though the state has performed well on the growth front, it has remained poor. Next, we look at how poverty has changed in Odisha during 1993–94 and 2009–10. We have estimated poverty using consumer expenditure data from NSS. As we know, NSS data is collected under Uniform Reference Period (URP) and Mixed Reference Period (MRP). In URP the recall period is one month for all consumable items, and under MRP the recall period is one month for frequently consumed items (such as food) and one year for durables. The Tendulkar committee recommended that we should use NSS data under MRP. Following their recommendation and their poverty line, we estimate poverty in the state using four measures. What is noteworthy is the fact that between 1993–4 and 2009–10, the decline in poverty has been significant. If we look at HCR, the decline in poverty is to the tune of 21 percentage points. The story is similar for urban Odisha. The decline in poverty between 2004–5 and 2009–10 has been very significant, no matter what poverty measure we use. These results are presented in Tables 1.4 and 1.5 below. Table 1.4. Poverty in Rural Odisha Poverty Measures Head Count Ratio % Poverty Gap Ratio % Sen Index*100 FGT Index*100

1993–4 63.16 16.03 21.760 5.683

(all figures in `).

2004–5 60.81 17.38 23.15 6.638

2009–10 39.19 8.99 12.41 3.003

2004–5 37.58 9.60 13.15 3.50

2009–10 25.91 5.31 7.55 1.70

Planning Commission Poverty line for Rural Odisha: 1993–4: 224.20; 2004–5:407.8; and 2009–10: 567.1

Table 1.5. Poverty in Urban Odisha Poverty Measures Head Count Ratio % Poverty Gap Ratio % Sen Index*100 FGT Index*100

1993–4 34.76 8.35 11.45 2.86

Planning Commission Poverty line for Urban Odisha (in `) 1993–4:279.3 295.9 all India 2004–5:497.31 578.8 all India 2009–10:736.00 859.6 all India

Next, we compare Odisha with all India in terms of poverty reduction. The results are presented in Table 1.6 below, showing that the decline in

Chapter One

16

poverty in Odisha between 2004–5 and 2009–10 is much more than the all-India figure. Looking at the Head Count Ratio, we see that there is a 21% decline in rural poverty in Odisha as against the 8% drop for all India. The story is similar for urban Odisha. Table 1.6. Poverty Comparison—Odisha and All India (HCR)

RURAL

URBAN

1993–4

2004–5

2009–10

Odisha

63.16

60.81

39.19

All India

65.79

41.83

33.81

Odisha

34.76

37.58

25.91

All India

39.09

25.74

20.88

III. Education In the previous section we saw that Odisha has grown at an appreciable rate in the last ten years. We also know that economic growth, though important, cannot be an end in itself. Higher standards of living, as well as of developing opportunities for all, coming from greater resources generated through growth, should be the ultimate aim of a development policy. In order to have higher standards of living, education plays an important role. This section examines how the state has performed on the education front. Given the fact that economic growth has been high and there has been significant reduction in poverty, one expects positive outcomes on education. To assess the performance in education, we use some standard parameters such as literacy rate, the gross/net enrolment ratio at various levels (primary, upper primary, secondary, higher secondary and higher education), dropout rates, gender parity index and transition rate. While all these figures throw light on the quantitative expansion of education, they do not reveal the quality of education being imparted. We will therefore now assess the quality of education in the state, particularly at the level of higher education. Table 1.7 below presents the literacy rates for Odisha and all India.

Growth, Poverty and the Social Sector in Odisha: An Overview

17

Table 1.7. Literacy Rates for Odisha and All India (%) Year

Odisha Male Female 1951 27.32 4.52 1961 34.68 8.65 1971 38.29 13.92 20.6 1981 46.39 34.68 1991 63.09 2001 75.35 50.51 82.40 64.36 2011 Source: Census of India (1951–2011).

Total 15.8 21.66 26.18 33.62 49.09 63.08 73.45

All India Total 18.33 27.30 34.45 43.57 52.21 64.8 74.04

Two points are to be noted here: (i) Literacy rate in the state is steadily rising and is close to the national average. (ii) The gender gap in literacy levels has been declining over the years but is still far from the desired level. In Kerala, the overall literacy rate in 2011 was 93.91%, with 96.2% of males and 91.98% of females being literate. Now we turn to the performance of the state in primary/upper primary education and then to secondary/ higher secondary / higher education.

III.1 Primary and Upper Primary Education In view of the Right to Education Act adopted by the state, primary education and upper primary education assume utmost importance. First, we look at the Gross Enrolment Ratio (GER) and Net Enrolment Ratio (NER) in primary education and upper primary education. As we know, the GER for primary education gives us the ratio between the number of children enrolled at primary level and the number of children in the desired age group (6–11 years) who should be enrolled for primary education. One can see that this ratio can exceed 100 if children from the higher age group are enrolled in primary education. NER takes care of this difficulty by looking at the number of children enrolled in primary education from the age group 6–11 years divided by the total number of children in the state in this age group. The results are presented in Table 1.8 below.

Chapter One

18

Table 1.8. GER and NER in primary and upper primary education Year Primary

2004–5 104.26 2005–6 92.25 2006–7 90.98 2007–8 96.66 2008–9 96.92 2009–10 94.04 2010–11 99.06 2011–12 99.69 Source: OPEPA.

GER Upper Primary

Total

Primary (6–11)

81.29 83.3 100.31 106.09 99.06 104.11 105.45 104.93

92.77 87.78 95.65 99.47 98.27 99.79 101.32 101.31

93.13 78.58 82.06 84.23 92.28 92.88 91.83 93.3

NER Upper Primary(11– 14) 69.04 71.84 63.11 76.62 85.52 85.68 83.84 93.24

Total (6–14) 73.65 78.58 82.02 82.03 90.47 91.29 90.78 93.27

What we see from Table 1.8 above is that the GER and NER have shown marked improvement over the last ten years. NER has improved from 73% to 93%, which is close to the national average. We must also look at some interesting facts regarding the number of schools and teachers recruited to manage them. As per the Odisha Economic Survey 2012–13, the number of primary schools increased from 52,972 in 2009–10 to 55,106 in 2011–12, but the number of teachers declined from 137,833 to 133,262. Similarly, the number of upper primary schools increased from 22,209 in 2009–10 to 23,239 in 2011–12, but the number of teachers declined from 56,758 to 53,264. This is somewhat puzzling, but the non-appointment of teachers has been a general malaise that the state has to address at all levels. Table 1.9 below looks at the dropout rates. It is satisfying to note that the dropout rates at the primary and upper primary levels are now negligible (6–14 years), having gradually reduced from a high in 2003–4. With a reduction in poverty there is less pressure on families to withdraw children from school and force them to join the workforce. Increased per capita income with a reduction in poverty must have affected this result. The other factor that might have influenced this is the provision of midday meals in schools, meaning that for poor families it may be better to keep their children in school than to pull them out. None of these observations apply at the secondary level where the dropout rate is still very high.

Growth, Poverty and the Social Sector in Odisha: An Overview

19

Table 1.9. Dropout rates for selected years Year

Primary Upper Primary Girls Total Girls Total 2003–4 35.4 33.6 58.60 57.50 2006–7 10.72 10.53 18.47 18.05 2008–9 4.89 4.95 8.43 8.42 2010–11 2.86 2.60 7.31 7.23 2011–12 0.62 0.43 2.23 3.07 Source: Odisha Economic Survey (2012–13).

Secondary Girls Total 66.7 64.4 62 61 60.6 59.3 52.1 51.0 51.8 49.5

To keep it simple, we did not include SCs and STs in Table 1.9. The dropout rates at the primary and upper primary levels are slightly higher if we consider the 2011–12 figures.

III.2 Secondary, Higher Secondary and Higher Education Between 2005–6 and 2011–12, the number of high schools in the state grew from 7,300 to 8,280, and the number of teachers grew from 61,900 to 64,800. The number of students grew from 13.46 lakhs to nearly 15 lakhs. To universalize secondary education, the state adopted Rashtriya Madhyamik Sikha Abhiyan (RMSA) in 2009–10, a Central Government scheme. Under this scheme, effort will be made to make quality education available, accessible and affordable to all children in the age group 14–18. As per the NUEPA figures, in 2009–10 the transition rate for the higher secondary level in Odisha was 48.6% as against 78.7% in Tamil Nadu and 74.64% in West Bengal. This could be because of many factors, an important one being that there are not enough seats at university level to meet the demand. In such a case, the state should take care to expand higher education to provide the opportunity for students to pursue higher education. For that purpose, if the state does not have the resources then private players may be invited to establish universities to pursue liberal education. Quantitative expansion cannot be the sole criterion of measuring success in educational attainments. Imparting quality education should be a major intent behind expanding education. One of the key elements of maintaining quality in education is to attract, appoint and retain good faculty. No educational institution can attain excellence without good faculty. The performance of the state, on this parameter, has been rather disappointing. Even a premier university like Utkal University has many postgraduate departments where 50–70% of the posts are vacant. This is not just limited to this institution alone. Regular posts are being filled up

20

Chapter One

by the contractual appointments with paltry salaries and they have no motivation to work. It is time for the state to realize that the long-term damage it will have on the quality of education is going to be irreversible and should take corrective actions very soon. When we look at technical education in the state, we find that in 2010– 11 there were eight government engineering colleges and 141 private engineering colleges; 13 government MCA colleges and 61 private MBA colleges; three government medical colleges and three private medical colleges; and one government pharmacy college and 32 private pharmacy colleges. Private players have entered the market to meet the demand for more engineers, managers, computer professionals, doctors and pharmacists. This is consistent with the growth story. With an increase in growth and manufacturing, more engineers and managers are required. The growth of the pharmaceutical industry needs more pharmacists. If the government has no resources to train such professionals, private players will come in. The government’s regulatory role should be limited to ensuring minimum infrastructure when they start up. By the market mechanism, if they cannot maintain quality they will perish.

IV. Health In order to assess the performance of the state on health, we look at some standard indicators such as crude birth rate (CBR), crude death rate (CDR), infant mortality rate (IMR), under-five mortality rate (U5MR), life expectancy at birth and total fertility rate (TFR). We also look at the health infrastructure the state has created over the last ten years and whether it is spending enough on health. We also look at alternative ways of providing health facilities to the population. Given the poor performance of the state in providing health facilities, providing health insurance to all may be a better way of ensuring health facilities. First, we look at some basic health indicators. For the sake of comparison, we have included one neighbouring state, West Bengal. In every indicator Odisha lags behind the all-India figure, as well as West Bengal. Clearly, the state needs to do more to improve health conditions. All indicators have improved over the last five years. If we look at Odisha only, IMR declined from 75 in 2005 to 57 in 2011; CDR fell from 9.5 to 8.5; and CBR fell from 22.3 to 20.1 during the same period. So, things have improved but much more still needs to be done.

Growth, Poverty and the Social Sector in Odisha: An Overview

21

Table 1.10. Some basic indicators Indicator

Odisha

India

West Bengal

1. Life expectancy at birth (Female) 2006–10 2. Life expectancy at birth (Male) 2006–10 3. Infant Mortality Rate 2011 (Per 1,000 live births) 4.Crude Birth Rate 2011 (Per 1,000) 5. Crude Death Rate 2011 (Per 1,000) 6.Under Five Mortality Rate 2010 7.Total Fertility Rate 2010 8. Sex Ratio 2011 (No. of females/1,000 males)

63.9

67.7

71

62.2

64.6

67.4

57

44

32

20.1 8.5 78 2.3 978

21.8 7.1 59 2.5 940

16.3 6.2 ----947

Source: Economic Survey (2012–13).

IV.1. Health Infrastructure As per the government of India norm, there should be one sub-centre for every 5,000 people, one primary health centre (PHC) for every 30,000 people, and one community health centre (CHC) for 120,000 people. The staff will be commensurate with this requirement. As per the calculation in Rout & Hota (2015), the shortfall in sub-centres is 17.8%, and in PHCs is 6.12%. When it comes to manpower, there is a 43.97% shortfall in surgeons at CHCs, an 89% shortfall in physicians, a 73% shortfall in paediatricians, a 57% shortfall in doctors at PHCs, a 73% shortfall in nursing staff, an 89% shortfall in radiographers at CHCs, and a 79% shortfall in laboratory technicians. In fact, the shortfall list goes on. Add to this the fact that Odisha has six medical colleges with 750 seats, amounting to one medical seat per 55,930 people (2011 census). The allIndia figure for this is one medical seat per 30,000 people. This convinces us that there is a huge shortfall in human resources to adequately cater to the health needs of the population. This is true of all-India situations also. No wonder that the number of OPD patients using the public facilities is falling over time (Odisha Economic Survey 2012–13). In 2006, 3,238.8 thousand used OPD, which declined to 2,056 thousand in 2010 and 1,322.6 thousand in 2011.

Chapter One

22

IV.2. Public Expenditure on Health First, we look at the public expenditure on health in India vis-à-vis some developed and emerging economics (as percentages of GDP) in table 1.11 below. Table 1.11. Public expenditure on health as a percentage of GDP in India vis-à-vis some developed and emerging economics Country

Australia UK US Mexico Brazil Russia China South Africa India

Expenditure on health (2010 or latest available year) Public

Private

Total

6.2 8.0 8.5 2.9 4.2 3.2 2.7 3.9 1.2

2.9 1.6 9.1 3.3 4.8 1.9 2.4 5.0 2.9

9.1 9.6 17.6 6.2 9.0 5.1 5.1 8.9 4.1

Source: Economic Survey (2012–13).

As is clear from the table, India’s expenditure on health as a percentage of GDP is very low. Public expenditure on health is less than half of the private expenditure, and India is the only country with this feature. Odisha also falls into the same pattern. In 2004–5, private expenditure on health was 2.7 crore, whereas public expenditure was 70 lakh, almost a quarter, while public expenditure on health was only 0.88% of GSDP. Health expenditure as a percentage of GSDP varied from 0.56% to 0.88% between 2004–5 and 2012–13; 2012–13 BE puts it at 0.69% of GSDP. The abysmal infrastructure we see in the provision of health is clearly due to a lack of adequate investment in the infrastructure. To the extent that financial resources permit it, a modern welfare state must strive to provide a minimum level of healthcare to those unable to afford it. Seeing the failure of the government to meet this requirement, one is tempted to conclude that one should hand this over to the private sector. So, to provide support for health care, public expenditure has to be stepped up to the tune of at least 3% of GDP.

Growth, Poverty and the Social Sector in Odisha: An Overview

23

IV.3. Insurance As against the OPD treatment of minor health problems, major illnesses are less frequent, more expensive and need hospitalization, and the associated expense is unpredictable. These characteristics make the market for the care of major illness a perfect candidate for insurance. We face here an adverse selection problem—those hit by a major illness would seek insurance while those in good health would avoid it. The common solution to the problem is Group Insurance. Since the poor cannot pay for it, a government subsidy to cover them will be required. A beginning in this direction has been made. Rashtriya Swasthya Bima Yojana (RSBY), launched by the Government of India on April 1, 2008, falls into this category. The scheme is funded in a 3:1 ratio by central and state governments and is available to households below the poverty line. A long list of illnesses is covered up to a maximum expenditure of 30,000 rupees per year for a family of five. As per government statistics, on December 8, 2011, 25.6 million smart cards were in circulation across 23 states. This seems to be a step in the right direction. Bhagwati & Panagariya (2012) go on to calculate that by spending 0.4% of GDP we can cover half the Indian population, which will provide an insurance of 50,000 rupees to a household of five members.

V. Concluding Remarks Odisha’s growth has been impressive since 2003–4, and this has also led to a substantial reduction in poverty. One expects that a reduction in poverty should lead to improved school enrolment, and the data supports this hypothesis. GER up to upper primary was close to 100% for 2011 and the dropout rate negligible at 2%. There is scope for expanding higher education facilities, but the state has to invest in human resources to maintain the quality in education. Health facilities have improved but remain unsatisfactory. The state has to increase the expenditure on healthcare in order to create a health provisioning infrastructure worthy of the name. For covering expenses on major illness, insurance cover seems to be a good idea and the state can actively participate in this.

24

Chapter One

References Bhagwati, J. & A. Panagariya. 2012. India’s Tryst with Destiny. Collins Business. Government of India. 2013. Economic Survey. Government of Odisha. 2012-1. Odisha Economic Survey. Nayak, N. C. S. Dhar & P. K. Mahakur. 2015. Educational Outcomes in Odisha: Trends and Issues. In The Economy of Odisha: A Profile. edited by Nayak P. B, Panda S.C. & Pattanaik P. K. New Delhi: Oxford University Press. Panda, S.C. 2015. Economic Growth, Inequality and Poverty in Odisha, 1993-2010 InTrend, Magnitude and Dimensions of Inequality in PostReform India. edited by Panda, R.K. and Meher, R.K. New Delhi: Concept Publications. Rout, H. S. & A. K. Hota. 2015. Health and Healthcare in Odisha. In The Economy of Odisha: A Profile. edited by Nayak P. B., Panda, S. C. & Pattanaik P. K. New Delhi: Oxford University Press.

CHAPTER TWO EFFICIENCY OF SOCIAL SECTOR EXPENDITURE IN INDIA BRIJESH C. PUROHIT

I. Introduction The social sector is an important item in state budgetary expenditure. It has remained around 5.8% of gross domestic product (GDP) and its share of total state expenditure has varied between 36.8% from 1990 to 1995 and 39.2% from 2010 to 2011. Within the social sector, education, sports and art constitute a major share of expenditure (nearly 57%). A good share is spent on culture (46.1%) and medical and public health (10.5%). Family welfare, water supply and sanitation, housing, urban development, welfare of scheduled castes (SCs), scheduled tribes (STs) and other backward classes (OBCs), labour and labour welfare, social security and welfare, nutrition, natural calamities and the rest comprise 1.7%, 5%, 3%, 8.9%, 6.9%, 1.1%, 9.6%, 3.7%, 1.3% and 2.3% of the social sector expenditure, respectively. It becomes pertinent, therefore, to find out whether the sectors with major expenditure like health and education are performing satisfactory according to the criteria of efficiency. Several approaches to measure the efficiency of government expenditure have been proposed. In general, these approaches are broadly categorized into four types: (i) studies that have concentrated on gauging and enhancing efficiency by focusing on certain types of government spending in a specific country; (ii) researchers who use data on inputs only, neglecting the output of government spending in quantitative terms; (iii) those that use data on outputs only, not on inputs; and (iv) those that use data on both inputs and outputs. These studies, however, have not made a consistent comparison of the efficiency of government spending among countries. Thus, the issue of gauging and enhancing the efficiency of government spending continues to interest policymakers and researchers alike (Chu &

26

Chapter Two

Hemming 1991; Chu et al. 1995; Gupta, Honjo & Verhoeven 1997; Purohit 2010a). This interest received a boost with the initiation of wideranging institutional reforms by some nations,1 which aimed at improving the efficiency of their public sectors. These reforms were to separate policy formulation from policy implementation, create competition among government agencies and between government agencies and private firms, and develop output-oriented budgets using a wide array of output indicators. This practice of following result-oriented public expenditure management has generated huge information on how to control production processes within the government and how to enhance their efficiency. For instance, using regression analysis and focusing on inputs, a twenty-year study of OECD member countries analysed the efficiency of healthcare systems. It shows that public-reimbursement health systems, which combine private provision with public financing, are associated less with public health expenditures and higher efficiency than with publicly managed and financed healthcare systems (Gerdtham et al. 1995).2 Another school has focused on the differences in social indicators among countries (used as indicators of government output) after netting out the effect of economic development on these indicators.3 These studies do not explicitly analyse the relationship between government spending and social indicators. There are certain studies that analyse both input and output. For instance, Harbison & Hanushek (1992) provide an overview of 96 studies on education production functions in developing countries and 187 studies on education production functions in the United States, and investigate the relation between input and output in education. It is found that in most of the studies of developing countries, teacher education, teacher experience and availability of facilities have a positive and significant impact on the output of education, and half of the studies show that the effect of expenditure per pupil is significant; the pupil-teacher ratio and teacher salary have no discernible impact on the output of education. Likewise, Jimenez & Lockheed (1995) also assess the relative efficiency of public and private education in several developing countries by taking into account both input and output. Yet another type of analysis, 1

For instance New Zealand in the late 1980s and the OECD in later years. See Scott (1996); Oxley et al. (1990); OECD (1994). 2 This is traced to the high incidence of gauge efficiency by looking at factors associated with relatively expensive in-patient care and the lack of a mechanism to restrain demand for specialized healthcare. Countries without ceilings on in-patient care were also found to have higher public health expenditure. 3 For example, Isenman (1980); Sen (1981); Aturupane, Glewwe & Isenman (1994); Kakwani (1993).

Efficiency of Social Sector Expenditure in India

27

by Tanzi & Schuknecht (1997), assesses the incremental impact of public spending on social and economic indicators, for example on real growth and the mortality rate in industrial countries, and concludes that higher public spending does not significantly improve social welfare. The growing importance of efficiency in resource utilization, particularly in the healthcare sector, has also attracted research in the area. A number of studies have emphasized the overall performance of health systems and their impact on health (WHO 2000; Murray & Frank 1999). As such, studies have focused on hospitals, nursing homes, HMOs and district health authorities (Worthington 2004, Hollingsworth & Wildman 2002; Jamison et al. 2001; Salomon et al. 2001; Evans et al. 2001; Wang et al. 1999; Sankar & Kathuria 2004). Recently, a better yardstick was developed to evaluate the economic performance of health systems. A number of studies have been conducted in the healthcare sectors, employing either non-parametric approaches like FDH or DEA4 or parametric approaches like SFA. In the former category one may include, for instance, studies by Aubyn (2002)5 (using FDH covering both the health and education sectors in Portugal), Felix Masiye (2007)6 for Zambian hospitals, Hofmarcher et al. (2002)7 for an Austrian province, Jaume Puig-Junoy (1998)8, a report on district hospitals in Namibia (2004),9 Gannon (2004)10 for Ireland, Mathiyazhgan (2006)11 for hospitals

4

DEA constructs an efficiency frontier that reveals the lowest input requirement for obtaining a given output level, or, from an output-oriented framework, the highest output obtainable from a given input set. 5 Miguel St. Aubyn. Evaluating efficiency in the Portuguese health and education sectors (February 2002). 6 Felix Masiye. Investigating Health System Performance. An Application of Data Envelopment Analysis to Zambian Hospitals. (Masiye; licensee BioMed Central Ltd., 2007). 7 M. M. Hofmarcher. I. Paterson. & M. Riedel. Measuring Hospital Efficiency in Austria--a DEA Approach. Healthcare Manag Sci. 5 (1) (2002): 7–14. 8 Jaume Puig-Junoy. Technical Efficiency in the Clinical Management of Critically Ill Patients. Health Economics. 7 (3) (1998): 263–277. 9 Government Of The Republic Of Namibia, Ministry Of Health And Social Services, The Technical Efficiency Of District Hospitals In Namibia. (July 2004). 10 Brenda Gannon, Technical Efficiency Of Hospitals In Ireland, Economic And Social Research Institute. Working Paper 18. (November 2004). 11 Maathai K. Mathiyazhgan, Cost Efficiency of Public And Private Hospitals: Evidence From Karnataka State In India. ISAS Working Paper No. 8. (January 27, 2006).

28

Chapter Two

in Karnataka State in India, Mirmirani (2008)12 for transition economies of the former Socialist bloc including Albania, Armenia, Russia and others, Kittelsen & Magnussen (1999)13 for Norway, Li & Wang (2008)14 relating to Chinese public acute hospitals, Hajialiafzali et al. (2007)15 relating to Iran, Magnussen (1996)16 on Norway, Suraratdechaac & Okunadeb (2006)17 for Thailand, Jeffrey & Coppola (2004)18 relating to the United States, Bates et al. (2006)19 also to the United States, Kontodimopoulos et al. (2003)20 for Greek hospitals and Spinks & Hollingsworth (2005).21 22 In 12 Sam Mirmirani. Healthcare Efficiency In Transition Economies: An Application of Data Envelopment Analysis. International Business & Economics Research Journal. 7 (2) (2008): 47–56. 13 S. A. C. Kittelsen & Jon Magnussen. Testing DEA Models of Efficiency in Norwegian Psychiatric Outpatient Clinics. Health Services Research, University of Oslo, Health Economics Research Programme. Working Paper (1999). 14 Ling Li & Jian Wang. Relative Efficiency of the Chinese Public Acute Hospitals: An Empirical Data Envelopment Analysis application. Department of Management, The Hong Kong Polytechnic University, Hong Kong. 15 Hossein Hajialiafzali. J. R. Moss. & M. A. Mahmood. Efficiency Measurement for Hospitals Owned by the Iranian Social Security Organisation. Journal of Medical Systems. 31 (3) (2007). 166–172. 16 Jon Magnussen. Efficiency Measurement and the Operationalization of Hospital Production, Health Services Research. (April 1996). 17 Chutima Suraratdechaac & Albert A. Okunadeb. Measuring Operational Efficiency in a Healthcare System: A Case Study from Thailand. Health Policy. 77 (1) (2006): 2–23. 18 P. Jeffrey & M. Nicholas Coppola. Efficiency of Federal Hospitals in the United States, Journal of Medical Systems. 28 (5) (2004): 411–422. 19 Laurie J. Bates. Kankana Mukherjee & Rexford E. Santerre. Market Structure and Technical Efficiency in the Hospital Services Industry: A DEA Approach. Medical Care Research and Review. 63 (4) (2006): 499–524. 20 N. Kontodimopoulos. P . Nanos & D. Niakas, Balancing Efficiency of Health Services and Equity of Access In Remote Areas in Greece. Health Policy. 76 (1) (2003): 49–57. 21 Jean Spinks & Bruce Hollingsworth. Health Production and the Socioeconomic Determinants of Health In OECD Countries: The Use of Efficiency Models. Centre for Health Economics. Monash University, Working Paper 151. (January 2005). 22 Questions of the application of data envelopment analysis (DEA) for crosscountry comparisons of the technical efficiency of health production. The authors argue that, in theory, such an analysis utilizes measures of the socio-economic determinants of health relevant to all social policies, not just the health policy. Using OECD and WHO panel data, particularly for 2004, this paper critically analyses a number of outstanding theoretical questions regarding the use of DEA in this setting. It concludes that until such questions like data quality, availability

Efficiency of Social Sector Expenditure in India

29

the latter type of studies one may include, for instance, Schmacker et al. (2008)23 relating to the United States, Evans et al. (2000),24 a crosscountry comparison, Evans et al. (2000a),25 Greene (2002),26 a crosscountry comparison, Farsi et al. (2007)27 relating to Switzerland, Wang et al. (2006)28 for New South Wales, Kris et al. (200129; 200730) relating to Texas, the United States, Rosko (2001)31 relating to the United States, Yong & Harris (1999)32 relating to Australia, Hollingsworth & Wildman

and comparability, missing variables and agreement on the most indicative social determinants of health are addressed; the resultant implications for policy will be based on misleading information. Using a framework for the socioeconomic determinants of health, the study concludes that policy makers should be aware of the limitations and uncertainty of using such techniques in the production of health settings. 23 E. R. Schmacker & N. L. McKay. Factors Affecting Productive Efficiency in Primary Care Clinics. Health Serv Manage Res. 21 (1) (2008): 60–70. 24 David B Evans, Ajay Tandon. Christopher J. L. Murray & Jeremy A. Lauer. The Comparative Efficiency of National Health Systems In Producing Health: An Analysis of 191 Countries. GPE Discussion Paper Series: No. 29. EIP/GPE/EQC, World Health Organization, 2000. 25 Ajay Tandon. Christopher J. L. Murray, Jeremy A. Lauer, David B. Evans. Measuring Overall Health System Performance for 191 Countries,” GPE Discussion Paper Series: No. 30. EIP/GPE/EQC. World Health Organization, 2000a. 26 William Greene. Fixed and Random Effects in Stochastic Frontier Models. Department of Economics, Stern School of Business. New York University. (October 2002). 27 Mehdi Farsi. Massimo Filippini & Diego Lunati. Economies of Scale and Efficiency Measurement in Switzerland’s Nursing Homes. (June 2007). 28 Jian Wang, Zhong Zhao & Amir Mahmood. Relative Efficiency, Scale Effect and Scope Effect of Public Hospitals: Evidence from Australia. University of Newcastle, Australia, IZA Discussion Paper No. 2520, December 2006, Bonn, Germany. 29 Kris J. Knox. Eric C. Blankmeyer & J. R. Stutzman. Efficiency of Nursing Home Chains and the Implications of Non-profit Status: A Comment. The Journal of Real Estate Portfolio Management. Apr-Jun (2001). 30 Kris J. Knox. Eric C. Blankmeyer & J. R. Stutzman. Technical Efficiency in Texas Nursing Facilities: A Stochastic Production Frontier Approach. Journal of Economics and Finance Spring. (2007). 31 Michael D. Rosko. Cost Efficiency of US Hospitals: a Stochastic Frontier Approach. Health Economics. 10 (6) (2001): 539–551. 32 Karen Yong. & Anthony Harris. Efficiency of Hospitals in Victoria under Casemix Funding: A Stochastic Frontier Approach. Working Paper 92. Centre for Health Program Evaluation, Monash University, Australia (June 1999).

30

Chapter Two

(2002),33 a cross country comparison, Hollingsworth & Wildman (2002),34 Mortimer & Peacock (2000)35 relating to Australia, Jayasuriya & Wodon (2002)36 and Purohit (2010a; 2010b; 2010c; 2010d; 2010e). The second type of studies has deployed frontier efficiency measurement techniques, which involve a production possibility frontier depicting a locus of potentially efficient technical output combinations that an organization or health system is capable of producing at any point in time. An output combination below this frontier is termed “technically inefficient” (Fried, Lovell & Schmidt 1993; Charnes et al. 1995; Coelli, Rao & Battese 1998). Despite their nascent application in the healthcare sector, an exhaustive review of studies applying these methods has been attempted, which provides the steps and empirical problems highlighted by researchers (Hollingworth, Dawson & Maniadakis 1999; Worthington 2004). Notably, there are very few studies in the context of developing countries and fewer, particularly in the context of India, which have focused on this aspect. The literature is marked by the absence of studies on the recent period.

II. Non-Parametric Approach—Free Disposal Hull (FDH) In this paper we use two types of techniques—non-parametric and parametric—that allow for a direct measurement of the relative efficiency of government spending among countries or states within a nation. In the first type we apply FDH analysis, which assesses the relative efficiency of production units in a market environment. This analysis consists of, first, establishing the production possibility frontier representing a combination of best-observed production results within the sample of observations (the "best practices"), and second measuring the relative inefficiency of 33

Bruce Hollingsworth. & John Wildman. The Efficiency of Health Production: Re-estimating the WHO Panel Data Using Parametric and Nonparametric Approaches to Provide Additional Information. Working Paper 131 Centre for Health Program Evaluation, Monash University, Australia, (May 2002). 34 Bruce Hollingsworth. and John Wildman. Efficiency and Cross Efficiency Measures: A Validation Using OECD Data. Working Paper 132 Centre for Health Program Evaluation, Monash University, Australia, May, 2002. 35 Duncan Mortimer. & Stuart Peacock. Hospital Efficiency Measurement: Simple Ratios vs Frontier Methods. Working Paper 135, Centre for Health Program Evaluation, Monash University, Australia (August 2000). 36 Ruwan Jayasuriya. & Quentin Wodon. Measuring and Explaining Country Efficiency in Improving Health and Education Indicators. The World Bank, (April 2002).

Efficiency of Social Sector Expenditure in India

31

producers inside the production possibility frontier by the distance from the frontier. The major advantage of FDH analysis is that it imposes only weak restrictions on the production technology while allowing for a comparison of efficiency levels among producers. The only assumption made is that input and/or output can be freely disposed of, so that it is possible to lower output with the same production technology while maintaining the level of input and increase the input while maintaining output at the same level. This assumption guarantees the existence of a continuous FDH or production possibility frontier for any sample of production results. Thus, FDH analysis provides an intuitive tool that can be used to identify best practices in government spending and to assess how governments fare vis-à-vis these best practices.37 In our analysis using FDH, the term "producer" is used as including government. A producer is relatively inefficient if another producer uses less input to generate as much or more output than the former. A producer is relatively efficient if there is no other producer that uses less input to generate as much or more output. In Fig. 2.1 below, this is illustrated in the case of one input and one output. Producer B uses more input to produce less output than producer A, and is therefore relatively inefficient in comparison with producer A. Producer C is relatively efficient; there is no producer in the sample that has a lower input as well as a higher output. Producers A and B have lower inputs but also lower outputs than producer C; producer D has a higher input as well as a higher output than producer C. By the same reasoning, producer D is also relatively efficient. If a producer is engaged in the production of multiple output using more than one input, it becomes more difficult to establish relative efficiency. In such a situation (of multiple inputs), it is postulated that a producer is relatively inefficient if they use as much as or more input to generate as much as or less output than all other producers, with at least one input being strictly higher or one output being strictly lower. This analysis (namely FDH) establishes the degree of efficiency in the given way. The first step is to identify the relatively efficient production results in the sample. In Fig 2.1, the relatively efficient producers are A, C and D. Given that producer A's production result is feasible and there is free disposal, all production results where at least as much input is used to generate the same level of output, or less, are also feasible. These relatively inefficient production possibilities are marked by the rectangular area to the right and below producer A, which contains producer B. 37

Some of the studies that have employed FDH analysis include Deprins, Simar & Tulkens (1984); Tulkens & Vanden Eeckhaut (1995); Fakin & de Crombrugghe (1997).

32

Chapter Two

Similarly, the rectangular area to the right and below producers C and D corresponds to relatively inefficient production possibilities. If there is no observation in the rectangular area to the left of and above an observed production result, the latter production result is among the relatively efficient production results in the sample of observations. The border of the set of production possibilities, that is, all the production results to the right of and below the relatively efficient observations, is given by the bold line connecting A, C and D in Fig. 2.1. This is the production possibility frontier, or FDH. A free disposal is required to obtain a continuous production possibility frontier. In the absence of that assumption, it cannot be inferred that all output combinations on the line connecting A, C, and D are feasible. In fact, a producer can be relatively efficient even though no producer is inefficient in relation to him or her (i.e., there is no producer in the rectangular area to the right of and below the relatively efficient producer). Such producers are assumed to be on the production possibility frontier. Producers who are efficient by default will be called independently efficient. Examples of independently efficient production results are producers C and D in Fig. 2.1. Producer A is not independently efficient as producer B is inefficient in relation to A. Using the criteria described above, a distinction can be made between relatively efficient production results (production results on the production possibility frontier) and relatively inefficient production results (production results in the production possibility set). We also use the measure of efficiency score that enables us to make a ranking of production results. The calculation of a producer's efficiency score can be illustrated using the example in Fig. 2.1. Producer B is the only relatively inefficient producer in the figure. FDH analysis suggests two alternative ways of measuring the distance of producer B’s production result from the production possibility frontier: from either the input’s side or the output’s side. In terms of input, the distance is given by the line bB, that is, the quotient of input used by producer A over inputs used by producer B— x(A)/x(B). This measure of efficiency is referred to as the input efficiency score. For all observations in the production possibility set, the input efficiency score is smaller than 1. For all observations on the production possibility frontier (producers A, C and D) the efficiency score is 1. The input efficiency score indicates the excess use of input by the inefficient producer and therefore the extent to which this producer allocates his or her resources in an inefficient manner. In terms of output, the efficiency score of producer B is given by the line b'B, that is the output quotient y(B)/y(A). This score indicates the loss of output relative to the most

Efficiency of Social Sector Expenditure in India

33

efficient producer with an equal or lower level of input. As in the case of the input efficiency score, the output efficiency score is smaller than 1 for observations in the interior of the production possibility set (producer B) and equal to 1 for observations on the production possibility frontier (producers A, C and D). In the one-input and one-output case depicted in Fig. 2.1, the formulation of an efficiency score is relatively straightforward. In the case of multiple inputs and outputs, the derivation of an efficiency score is more complicated.38 Non-FDH techniques typically assume a convex PPF.39 Consequently, the status of producer C changes; rather than being a relatively efficient unit on the production possibility frontier as under FDH, producer C is viewed as relatively inefficient, with a production result in the interior of the production possibility set.40 The example in Fig. 2.2 below illustrates that FDH singles out more observations as relatively more efficient than DEA, thereby reducing the informational value of FDH analysis. Another drawback of FDH analysis (as well as DEA), compared to parametric techniques, is that the correction of random factors unrelated to efficiency is not possible and therefore statistical noise remains in the measure of inefficiency. On the other hand, both DEA and parametric techniques impose more restrictions on the production technology than FDH analysis. As noted above, DEA assumes convexity whereas parametric techniques impose a functional form for the production possibility frontier. Where these assumptions inaccurately capture the production processes underlying the observed production results, the efficiency results are affected (Ferrier & Lovell 1990). The above discussion suggests that the choice between different techniques to estimate efficiency is a trade-off between imposing fewer restrictions on the production technology and obtaining relatively unambiguous results (Bauer 1990). In the case of government spending on education and 38

See, for instance, Gupta, Honjo & Verhoeven (1997). For instance, the Data Envelopment Analysis (DEA) technique (Charnes, Cooper & Rhodes 1978) assumes that the production possibility set is convex. With DEA, the area under the straight line connecting producer A and D would become part of the production possibility set (see Fig. 2.2). 40 If the production technology is also assumed to feature constant returns to scale (i.e. if the technology can be described by a Cobb-Douglas production function), the production possibility frontier would be a straight line through the origin. In this case, producer A would be the only producer on the production possibility frontier as it would have the highest observed output-input ratio, that is the highest average productivity. 39

34

Chapter Two

health, there is little a priori justification for making certain assumptions regarding convexity and economies of scale. This argues against the use of parametric techniques and favours the use of the relatively cost-conscious FDH analysis. Tulkens & Eeckhaut (1995) provide a more comprehensive overview of the differences between FDH analysis and these alternative techniques. Fig. 2.1. Free Disposable Hull (FDH) production possibility frontier

Efficiency of Social Sector Expenditure in India

35

Fig. 2.2. Alternate production possibility frontiers—Free Disposable Hull (FDH), Data Envelopment Analysis (DEA) and Constant Returns to Scale (CRS)

III. Results of our FDH analysis The results of our FDH analysis for the education and health sector using data for Indian states, both major and smaller ones, are presented below in Annexure Figs. A1–A10. It can be observed that, for per capita public expenditure on health (2001–2), those states that emerged as independently efficient states from FDH, for major states, are UP, Bihar, Gujarat West Bengal (WB), Maharashtra and Kerala (Fig. A1). Among the smaller states, the independently efficient states are Chhattisgarh, Jharkhand, Uttarakhand, Tripura and Manipur (Fig. A2). Likewise, in Figs. A3 and A4 (for 2004–5 per capita public expenditure), the situation somewhat changes for UP, whereas it remains the same for other independently efficient states. Among smaller states, a changed situation with lower efficiency emerges for Tripura only (Fig. A4). The FDH for public expenditure in 2008–9 for the health sector (Figs. A5 and A6) shows some more states, namely WB

Chapter Two

36

and Tamil Nadu, among the independently efficient states (Fig. A5), and the inclusion and exclusion of Goa and Chhattisgarh, respectively, in the category of such (independently efficient) states (Fig. A6). Table 2.1. Input Efficiency Score (IES) (education) (2008–11) States

PC Public Net IES Literacy IES Exp. enrolment (2008– (2011) (2011) Education primary 9) (2008–9) (2008–9) AP 1195.59 79.12 0.67 67.66 0.85 Assam 1374.02 83.58 0.95 73.18 0.74 Bihar 725.89 53.38 1.00 63.82 1.00 Gujarat 1015.67 59.75 0.79 79.31 1.00 Haryana 1615.77 74.14 0.81 76.64 0.92 Karnataka 1429.04 69.14 0.92 75.60 0.71 Kerala 1661.71 84.71 0.79 93.91 1.00 MP 799.49 97.28 1.00 70.63 1.00 Maharashtra 1487.72 88.93 0.88 82.91 1.00 Odisha 1193.44 69.16 0.67 73.45 0.85 Punjab 1395.89 74.15 0.94 76.68 0.73 Raj 1096.43 76.54 0.73 67.06 0.93 TN 1310.20 119.56 1.00 80.33 0.78 UP 763.40 56.35 1.00 69.72 1.00 WB 943.52 87.17 0.85 77.08 1.00 Smaller States—Input Efficiency Score (IES) (education) (2008–11) Arunachal 3684.77 115.15 1.03 66.95 0.90 Pradesh Chhattisgarh 1211.87 88.30 1.00 71.04 1.00 Goa 4648.96 62.04 0.81 87.40 0.81 HP 3299.52 115.11 1.00 83.78 1.00 JK 1497.35 100.69 1.00 68.74 Jharkhand 1162.75 73.18 1.00 67.63 1.00 Manipur 2054.26 83.20 0.73 79.85 1.00 Meghalaya 2110.56 83.46 0.71 75.48 0.97 Mizoram 3780.70 104.75 1.00 91.58 1.00 Nagaland 2339.54 88.34 0.64 80.11 1.00 Source: Purohit (2014).

Efficiency of Social Sector Expenditure in India

37

Table 2.2. Input Efficiency Score (IES) (health) (2001–5) Infant Public Infant IES Public Exp. PC Surviv Exp. PC Survival al (in `.) (2003) (in `.) (2006) (2004–5) (2001–2) Andhra Pradesh 182 941 0.81 191 944 Assam 176 933 0.83 162 933 Bihar 92 940 1 93 940 Gujarat 147 943 1 198 947 Haryana 163 941 0.90 203 943 Himachal Pradesh 493 951 0.49 630 950 Karnataka 206 948 0.95 233 952 Kerala 240 989 1 287 985 Madhya Pradesh 132 918 0.69 145 926 Maharashtra 196 958 1 204 965 Odisha 134 917 1.09 183 927 Punjab 258 951 0.93 247 956 Rajasthan 182 925 0.81 186 933 Tamil Nadu 202 957 1.18 223 963 Uttar Pradesh 84 924 1 128 929 West Bengal 181 954 1 173 962 Smaller States/UT—Input Efficiency Score (IES) (health) (2001–5) Arunachal Pradesh 627 966 0.55 841 960 Chhattisgarh 121 930 1 146 939 Delhi 426 972 0.81 560 963 Goa 685 984 1 861 985 Jammu & Kashmir 271 956 0.66 512 948 Jharkhand 146 949 1 155 951 Manipur 345 984 1 294 989 Meghalaya 407 943 0.85 430 947 Mizoram 836 984 1 867 975 Pondicherry 841 976 0.99 1014 972 Sikkim 825 967 1.01 1082 967 Tripura 301 968 1 328 964 Uttarakhand 178 959 1 280 957 Nagaland Na Na 639 980 Source: Purohit (2014).

States/UT

IES

0.91 1.07 1.00 0.87 1.00 0.46 0.88 1.00 0.64 1.00 0.95 0.83 0.93 0.91 0.73 1.00 0.35 1.00 0.53 0.34 0.57 1.00 1.00 0.68 0.34 0.29 0.27 0.90 1.00 0.46

Chapter Two

38

Table 2.3. Input Efficiency Score (IES) (health) (2010) States

Public Exp. PC Infant survival rate (2010) (in `.) (2008–9) Andhra Pradesh 410.00 954.00 Assam 471.00 942.00 Bihar 173.00 952.00 Gujarat 270.00 956.00 Haryana 280.00 952.00 Himachal Pradesh 884.00 960.00 Karnataka 419.00 962.00 Kerala 454.00 987.00 Madhya Pradesh 235.00 938.00 Maharashtra 278.00 972.00 Odisha 263.00 939.00 Punjab 360.00 966.00 Rajasthan 287.00 945.00 Tamil Nadu 410.00 976.00 Uttar Pradesh 293.00 939.00 West Bengal 262.00 969.00 Smaller States/UT: Input Efficiency Score (IES) (health) (2010) Arunachal Pradesh 771.00 969.00 Chhattisgarh 378.00 949.00 Delhi (UT) 840.00 970.00 Goa 1149.00 990.00 Jammu & Kashmir 845.00 957.00 Jharkhand 328.00 958.00 Manipur 695.00 986.00 Meghalaya 690.00 945.00 Mizoram 1611.00 963.00 Puducherry 1333.00 978.00 Sikkim 1446.00 970.00 Tripura 740.00 973.00 Uttarakhand 630.00 962.00 Source: Purohit (2014).

IES 1.00 0.96 1.00 1.00 0.99 0.96 0.98 1.00 0.74 1.00 1.06 0.77 0.97 1.00 0.95 1.00 0.90 0.87 0.83 1.00 0.82 1.00 1.00 0.91 0.71 0.86 0.79 0.94 1.00

In the education sector, using literacy (2011) and public expenditure (2008–9), states like Bihar, UP, WB, Gujarat and Tamil Nadu. Maharashtra and Kerala (among the major states) and Jharkhand, Chhattisgarh, Manipur and Himachal Pradesh (among smaller states) emerge as independently efficient states (see Figs. A7 and A8). By and large, a similar observation can be made using net enrolment at the primary level in 2008–9 (Figs. A9 and A10). Using this FDH analysis,

Efficiency of Social Sector Expenditure in India

39

Input Efficiency Scores (IES) are presented in Tables 2.1, 2.2 and 2.3. It can be observed that there is a range of 7-25% for major states and a scope of nearly 10% for smaller states to improve their input efficiencies relative to the nearest independently efficient states in 2011 in the education sector (see Table 2.1 above). In case of the health sector, this range is much wider for some years, like 2004 and 2005 (see Table 2.2 above, last column), and it has been 1%–26% for major states and 6%–29% for smaller states for the year 2010 (see Table 2.3 above, last column). Thus, it is suggestive of a considerably better scope for improvement in the efficiency of public expenditure in health relative to education. Therefore, we focus on the health sector and also apply a parametric approach of Stochastic Frontier Analysis (SFA) to explore efficiency and find the factors that can be separated to suggest ways to improve efficiency of public expenditure in this sector.

IV. Parametric Technique—The Stochastic Frontier Method In the application of parametric techniques, stochastic methods can be used to correct measurement and other random errors while making estimations of the production possibility frontier. In any parametric techniques, a functional form is postulated for the production possibility frontier and then a set of parameters is selected that best fits the sample data.

IV.1 Model Specification and Methodology In our estimation of health system efficiency, our specification is based on a general stochastic frontier model presented as: lnqj = f(ln x) + vj- uj …………………(1) Where: ln qj is the health output (life expectancy or inverse of IMR) produced by a health system “j,” x is a vector of factor inputs represented by per capita health facilities (including per capita availability of hospital beds, per capita primary health centres [or sub centres], per capita doctors, per capita paramedical staff, and per capita skilled attention for birth). vj is the stochastic (white noise ) error term. uj is a one-sided error term representing the technical inefficiency of the health system “j.”

Chapter Two

40

Both vj and uj are assumed to be independently and identically distributed (iid) with variance Vv2 and Vu2, respectively. From the estimated relationship ln q^j = f (ln x) - uj The efficient level of health outcome (with zero technical inefficiency) is defined as: ln q* = f (ln x) This implies ln TEj = ln q^j - ln q* = - uj Hence TEj = e-uj, 060 Number of Schools With Pre-Primary Number of Schools with Common Toilet Number of Schools With Girls Toilet Number of Schools with Drinking Water Facilities Number of Schools With Ramp

179

6,343

257

36

215

116

6,967

1,046

1,220

53

448

64

2,831

877

664

196

16

6

1,759

26,883

11,828

338

2,599

682

42,330

7,769

6,199

392

2,280

598

17,238

30,459

15,033

635

3,476

962

50,565

10,618

8,468

140

1,150

170

20,546

Note: Sec: Secondary; H. Sec: Higher Secondary. Source: DISE (2011).

Every primary school should have a separate girls’ toilet, but only 22.61% (7,769) do, and 78.25% have common toilet facilities. Drinking water facilities in primary schools have increased to 88.67%, while 11.33% of primary schools do not yet have drinking water facilities. Table 10.3. Teacher availability and student enrolment status

Parameters

No Female Teacher Schools (tch>=2) Number of Schools with =100 Number of Schools Established Since 1994 Number of Schools Established Since 2001

Primary Only

Primary with Upper Primary

Primary with upper Primary Sec/H.Sec

Upper Primary Only

Upper Primary with Sec./H.Sec

No Response/ Total

11,914

4,415

66

2,275

212

18,882

16,716

533

19

801

93

18,162

221

265

114

40

185

825

9,795

870

141

355

105

11,266

8,210

442

34

49

66

8,801

Chapter Ten

180 Percentage of Schools with Kitchen Shed (2009–10) Percentage Singleclass Room Schools (2008–9) Percentage Singleteacher Schools (2008–9) Percentage of Schools with SCR >= 60(2008–9) Percentage of Schools with =100(2008–9) Enrolment in Government Schools in 2009–10 Enrolment in Private School Enrolment in Government Schools—Rural Enrolment in Private Schools— Rural Enrolment in Single Teacher Schools Girls Enrolment Regular Teachers Male Regular Teachers Female Para Teachers Male Para Teachers Female SC Teachers Male SC Teachers Female ST Teachers Male ST Teachers female

22.08

34.84

51.52

8.07

20.42

24.74

2,813

123

4

118

27

3,085

4,289

118

21

141

35

4,604

1,370

1,322

77

634

67

3,470

15,746

449

39

675

2,349

19,260

169

175

82

35

22

483

2,295,830

2,677,718

87,222

248,261

187,277

0

84926

121243

163162

107198

16675

0

2,116,140

2,488,442

66,860

231,438

115,977

0

62,956

74,334

45,389

102,479

10,821

0

266,870

23,783 1,357,366

12,520 117,957

13,786 171,865

20,499 101,446

337,458

1,169,775

32,071

28,985

2,403

6,363

1,394

0

20,119 19,713

18,051 19,276

2,558 263

1,390 1,474

858 390

0 0

12,134 9,416

12,951 7,482

185 216

1,038 442

410 179

0 0

4,193 10,639 3,888

3,821 7,834 3,408

184 242 150

249 427 234

98 127 76

0 0 0

2,918,409

Source: DISE (2011).

In order to achieve a 100% enrolment ratio, the government has relaxed the norm for the opening of new primary schools in specific situations:

The Distant Goal of Universalising Elementary Education

181

(1) In KBK (Koraput, Balangir and Kalahandi) districts and Tribal Sub Plan areas, new primary schools will be opened in habitations with at least 25 children in the 6–14 age group provided there is no primary school within one kilometre of such habitations. (2) In all the districts, the distance norm for the opening of new primary and upper primary schools is relaxed in the case of natural barriers like rivers, hilly terrain and dense forests. There are 218 minority and mission-managed primary schools, wherein 599 teachers are receiving grant-in-aid from the government. Besides this, Odia medium schools functioning in the neighbouring states are also being provided with financial assistance from the government (2011).

I.2. Study Description In spite of the concerted efforts made by both central and state governments, many children have not seen the light of education and many others have not been able to complete basic elementary education, i.e. up to the 8th standard. What is the reason behind this? Is there a supply or demand side problem? In this context, the present study seeks to take stock of the impact of Sarva Siksha Abhiyan on the educational attainments of the rural households in Odisha and explore the reasons behind the failure of school-age children to complete the minimum elementary level of education. For this purpose, we undertake a case study of Bhagabatipur village in the Kantabada Panchayat of Bhubaneswar block, Odisha state.

I.3. Objectives The main objectives of the present study are: (1) to assess the progress of Sarva Siksha Abhiyan in the survey village, and (2) to examine the constraints faced by households in providing their children with elementary education. The rest of the chapter is organised as follows: Section II describes the methodology adopted for the study and describes the study area. Section III explains the socio-economic profile of the survey village. Section IV presents the education profile of the adults and progress of SSA in

182

Chapter Ten

Bhagabatipur. Section V provides conclusions and offers a few policy recommendations based on the survey findings.

II. Methods In order to assess the progress of SSA, we conducted a survey of all households in the Bhagabatipur village of the Kantabada Panchayat, which comes under the Bhubaneswar Block of the Khordha district in Odisha, around 25 kilometres north-west of Bhubaneswar. The village was selected given its particular characteristics. Most of the households of the village live in abject poverty, and the social categories the residents of the village belong to are even more complicated to decipher. Although most of the households claim to be Scheduled Tribe they have been denied ST certificates and provided with OBC certificates. Apart from these pathetic socio-economic conditions, the village comes under the Bhubaneswar block, the capital city of Odisha. The backwardness of a village lying so near the capital city of the state seems to be a development paradox and makes a strong case for research. We gathered information from every household on the following indicators through a well-designed and pre-tested questionnaire: their socio-economic conditions; asset holding; income; occupation; education profile of each family member; reasons for adults’ failure to receive education; the current status of the children in the families, if any; education performance in school; each household’s association with the school; and reasons for a child’s failure to continue their education. We conducted the pilot survey on January 29, 2011, after which the necessary modifications were made in the questionnaire and the final survey was conducted during February 2011. After the survey, we entered the data into Microsoft Access software, exporting it to SPSS we did the necessary cleaning. Finally, we analysed the data in SPSS to answer the research questions we set earlier.

II.1. Study area Bhagabatipur village is located at latitude 20°18'45"N and longitude 85°43'3"E (see Fig.10.1 below). The village comes under the Kantabada Panchayat of the Bhubaneswar block in the Khordha district. By road, it is around 25 kilometres northwest of Bhubaneswar railway station. Adjacent to the village are three tourist destinations, namely Deras dam, Jhumuka dam and the famous Chandaka-Dampara sanctuary.

The Distant Goal of Universalising Elementary Education

183

The area records a maximum temperature of around 40ºC in the summer (April to May) and a minimum of 10ºC on winter nights (December to January), and an average rainfall of 1,400 mm. A large part of the land of the village is dry. Nevertheless, the presence of two dams, Deras and Jhumuka, facilitates irrigation. Fig.10.1. Location of survey village

II.2. Village economy The mainstay of the Bhagabatipur village economy is its income from forestry, and a majority of the households depend upon the forest directly or indirectly for their livelihoods. Agriculture plays the second most important role in supplementing the livelihoods of a few households. People depend upon forests, selling timber and charcoal in the market. Along with this, households depend upon the forest for collecting firewood for domestic consumption and non-timber forest products. The forest also serves as grazing ground for domestic animals. Households are involved in the eco-tourism project of the Chandaka-Dampara sanctuary. The nearest dams also attract many tourists throughout the year. Villagers earn some money by renting out cooking accessories and spaces for feasts. They also earn some money from cashew plantations and stone quarries, which are collectively owned by the village. There is one common pond in

Chapter Ten

184

the village, and they earn some money from fishing. The village has two youth clubs named Chakadola Yuvak Sangha and Vivekananda Yubak Sangh, which do not undertake any developmental activities. Women have formed three self-help groups for some economic activities and mobilising savings. However, only one SHG is functioning, and not efficiently. During our focus group discussion with the villagers they pointed out that many voluntary groups and NGOs have come to create awareness about the importance of education, stop the consumption of liquor and empower women. However, no impact is visible in the village. The village has one temple and a meeting pandal. Although there are three ward members in the village, their decisionmaking role for the village is minimal, as this is usually led by the village president. The residents meet at least once a month to discuss important issues relating to the village. They also celebrate various festivals together.

III. Socio-Economic Profile of the Study Area III.1. Demographic profile of the village As per the government voter list there are three wards in Bhagabatipur with around two hundred households. Our objective was to survey all households in the three wards. Due to the absence of some family members, we could survey only 173 households. Of these, around 79% belonged to the scheduled tribe (ST) community, 7% to general classes, 12% to other backward castes (OBC), and the remaining 2% to scheduled castes (SC) (see Table 10.4 below). Table 10.4. Social composition of the households Caste Category General OBC Scheduled Caste Scheduled Tribe Total Source: Primary survey.

Number of Household 12 21 4 136 173

% 6.9 12.1 2.3 78.6 100.0

The Distant Goal of Universalising Elementary Education

185

III.2. Gender and age distribution The surveyed households (173) have a total population of 957. The village demonstrates a healthy sex ratio with 49.6% male and 50.4% female (see Table 10.5 below). Table 10.5. Sex composition in Bhagabatipur Sex Male Female Total

Number 475 482 957

% 49.6 50.4 100.0

Source: Primary survey.

Table 10.6 below shows the gender distribution of the total population in different age groups. Although the overall sex ratio looks almost balanced, this is not so in different age groups. The under-five sex ratio is tilted marginally towards the male population, with around 51.6% of the total children being male. But the proportion of girls in the 5–14 age group is much higher than their male counterparts. Around 60% of the total population in this age group is male and the remaining 40% female. The population aged over 15 demonstrates an almost balanced distribution.1 Table 10.6. Gender distribution in different age groups Age Groups 1–4 5–14 15–30 31–60 61 and above Total

Male 48 (51.6%) 71 (39.9%) 152 (50.8%) 148 (53.6%) 32 (48.5%) 451 (49.5%)

Female 45 (48.4%) 107 (60.1%) 147 (49.2%) 128 (46.4%) 34 (51.5%) 461 (50.5%)

Total 93 178 299 276 66 912

Source: Primary survey.

During our survey, 4.7% of the total population could not report their ages. Leaving this apart, Table 10.7 below shows that a large share of the total population is in the working age group (see the last column). Only 1The

total number of the male and female population is matching in Tables 10.5 and 10.6 because a few people could not report their age, and are unaccounted for in Table 10.6.

Chapter Ten

186

7% of the people are senior citizens and another 30% are below 14 years old. Thus, the total dependent population (below 14 and above 61) constitutes 37% of the total population. The working-age group (15–60) constitutes 63% of the total population. Table 10.7. Age distribution of the population Age Group 1–4 5–14 15–30 31–60 61 and above Total Missing System Total

Frequency 93 178 299 276 66 912 45 957

% 9.7 18.6 31.2 28.8 6.9 95.3 4.7 100.0

Valid % 10.2 19.5 32.8 30.3 7.2 100.0

Source: Primary survey.

III.3. Economic Profile A majority of the households (77.5%) in the village live below the poverty line (BPL) (see Table 10.8 below). Only 10.4% of the households have been able to break out of the vicious circle of poverty. Another 12% of the households belong to the BPL category but have not been able to get a BPL card. Without a BPL card, the households are deprived of a number of benefits like the public distribution system, Indira Awas Yojana. Many times, obtaining a BPL card becomes a political issue. The households that are not on good terms with the ward member or sarpanch cannot get one, while others in the good books of political people easily obtain one, even though they are not entitled to. Table 10.8. Economic profile of the households Economic Profile Above poverty line Below poverty line BPL but do not have card Total Source: Primary survey.

Frequency 18 134 21 173

% 10.4 77.5 12.1 100.0

The Distant Goal of Universalising Elementary Education

187

III.4. Income and poverty A very large proportion of the population (57%) earn no income. The size of such a population is definitely much larger than the unemployed population because some occupations like household chores do not yield any income. Similarly, occupations like farming one’s own land and share-cropping also do not generate income in the case of subsistence farming. A few individuals who work as domestic servants also reported that they receive payments in kind and rarely in cash. Individuals who earn income are classified into four categories. The first comprises those earning an income below `780, who are considered BPL persons, taking into account the requirement of a daily income of `26 in rural areas suggested by the Planning Commission of India. Keeping in view the debate on this measure, we can also call it the destitute line (see Table 10.9 below). The second category includes individuals who earn at least one dollar a day (a dollar is valued at `50) but remain above the destitute line (income between `781 and `2250 per month). The third category includes individuals who earn below two dollars a day but more than one dollar a day (income between `2251 and `3000 per month). The last category includes individuals who earn more than two dollars a day (income above `3001 per month). Table 10.9. Income distribution of individuals aged over 14 Income category No income 1–780 (Destitute line) 781–2250 (Old International poverty line 2,251–3,000 (New international poverty line

3,001 and above (Above international poverty line)

Total Source: Primary survey.

Number of Households 365

% 56.9

Cumulative % 56.9

23

3.6

60.5

105

16.4

76.9

85

13.3

90.2

63

9.8

100.0

641

100.0

Chapter Ten

188

If we take into account the Planning Commission’s estimate, 57% (see the cumulative percentage) of the population come below the destitute line. If we consider the old international poverty line as a one dollar a day income, then 77% of the people are considered to be poor. Further, if we consider the revised international poverty line with two dollars a day, 90% of the people are poor. Therefore, only 10% of the people of Bhagabatipur come above the international poverty line. Table 10.10 below shows the income distribution among the male and female populations aged over 14. The non-monetisation of the works of the female population is the major reason behind its high incidence of poverty. This is reflected in table 10.10 below, which shows that 71.5% of the female population have no income. Similarly, 76% of women fall below the destitute line and 97% of women fall below the revised international poverty line. Table 10.10. Income distribution among the male and female populations aged over 14 Income category

Male

Cumulati ve %

Female

Cumulat ive %

No income

144(43. 4)

43.4

221 (71.5)

71.5

1–780

10(3.0)

46.4

13(4.2)

75.7

781–2,250

60 45 64.5 (18.1) (14.6) 2,251–3,000 65 20 84.1 (19.6) (6.5) 53(16.0 3,001 and above 100 10(3.2) ) 332(100 309 Total ) (100) Note: Values in the parentheses are percentages. Source: Primary survey.

90.3 96.8 100

Total 365 (56.9) 23(3.6 ) 105 (16.4) 85 (13.3) 63(9.8 ) 641 (100)

Cumulat ive % 56.9 60.5 76.9 90.2 100

A common parlance hypothesis is made about the education and income relationship, claiming that higher education positively influences the income level. Therefore, we wanted to know if education has some bearing on the income level of individuals in this village. Table 10.11 shows the association between the income of individuals and their levels of education. We find that a very large proportion of individuals in all education categories (illiterate to tertiary level) have no income. Thus, no pattern is observed between the level of education and income of

The Distant Goal of Universalising Elementary Education

189

individuals. Among individuals who have obtained tertiary-level education, 50% have no income and another 50% earn a monthly income of more than `2251, which keeps them above the old international poverty line (one dollar a day). The reason behind the lack of a clear-cut association between education and income level would be the very low level of the education itself. Very few people have obtained tertiary level education, and primary or secondary levels of education hardly improve the skills of an individual. In rural areas, primary or secondary levels of education make the youth unemployable. After attaining secondary-level education, children do not feel comfortable working in low-paying jobs like agricultural labour, preferring to remain unemployed. Table 10.11. Income and education relationship

Illiterate 160 (54.4) 14 (4.8) 66 (22.4)

Education level Primary Secondary Tertiary 104 (56.2) 91 (64.1) 10 (50.0) 7 (3.8) 2 (1.4) 0 (0.0) 28 (15.1) 11 (7.7) 0 (0.0)

Total 365 (56.9) 23 (3.6) 105 (16.4)

34 (11.6)

28 (15.1)

20 (14.1)

3 (15.0)

85 (13.3)

Above 3,001

20 (6.8)

18 (9.7)

18 (12.7)

7 (35.0)

63 (9.8)

Total

294 (100)

185 (100)

142 (100)

20(100

641(100)

Income category No income 1–780 781–2,250 2,251– 3,000

Note: Values in the parentheses are percentages. Source: Primary survey.

IV. Education Profile and SSA IV.1. Education profile of individuals aged five or above On average, around 41% of the total population aged five or above are illiterate; 38.3% attended primary school; 18.4% obtained secondary-level education; and only 2.4% received tertiary-level education (see Table 10.12 below). A great deal of disparity is observed in the educational attainments of males and females. Out of the total female population in this age group, 47% are illiterate; around 40% attended primary school; 11.5% attended secondary school and only 2% received tertiary-level education. On the other hand, among the total male population aged five or above, 34.5% are illiterate; 37% attended primary school; 25.6% reached secondary school and only 3% have tertiary-level education. It is

Chapter Ten

190

worth pointing out that a greater percentage of the female population is either illiterate or has attended only primary school, where as a higher percentage of the male population has a secondary level of education. A very small percentage of men or women are able to obtain tertiary-level education. Table 10.12. Education profile of males and females aged over five Education level of individual Illiterate Primary Secondary Tertiary Total

Male 139 (34.5) 149 (37.0) 103 (25.6) 012 (03.0) 403 (100.0)

Gender category Female 195 (46.9) 165 (39.7) 048 (11.5) 008 (01.9) 416 (100.0)

Total 334 (40.8) 314 (38.3) 151 (18.4) 020 (02.4) 819 (100.0)

Note: Values in parentheses are percentages. Source: Primary survey.

Table 10.13 below demonstrates generational change in receiving education, which shows the level of education attained by the population in various age groups. A very high rate of illiteracy is recorded among those aged 61 or above. The percentage of illiterates among the younger generation has been consistently declining, but there is a long way to go before the universalisation of primary education among children aged 5– 14. Even after the enforcement of the right to education and implementation of Sarva Siksha Abhiyan, 22.5% of children aged 5–14 are illiterate. A few children aged 14 can also enter class 9, which is categorised under the secondary level of schooling. Therefore, it is observed that 5% of children aged 5–14 attained a secondary level of schooling. A very negligible percentage of senior citizens have obtained either secondary (4.5%) or tertiary-level (1.5%) education. Although there has been a rise in the percentage of the young population with secondaryor tertiary-level education, this is far from satisfactory. In the age group 15–30, only 38.5% of the population have obtained secondary-level education and only 3% have reached tertiary (university) level education.

IV.2. Reasons behind adults’ inability to get education Table 10.14 below presents the factors responsible for the illiteracy of individuals aged 15 and above. A large proportion (27.89%) of adults reported that poverty was the main obstacle for their education.

The Distant Goal of Universalising Elementary Education

191

Unawareness of the importance of education was also responsible for 16.67% of illiteracy. Fear of education and the need to do household chores also served as barriers for 11.56% and 11.9% of the adults, respectively. The unavailability of schools in the locality was also a barrier for 6.80% of adults. Table 10.13. Education level of individuals aged five and above across various age groups Education level of individual 5–14 15–30 Illiterate 40 (22.5) 89 (29.8) Primary 129 (72.5) 86 (28.8) Secondary 9 (5.1) 115 (38.5) Tertiary 0 (0.0) 9 (3.0) Total 178 (100) 299 (100) Note: Values in parentheses are percentages. Source: Primary survey.

Age Group 61 and above 31–60 153 (55.4) 52 (78.8) 89 (32.2) 10 (15.2) 24 (8.7) 3 (4.5) 10 (3.6) 1 (1.5) 276 (100) 66 (100)

Total 334 (40.8) 314 (38.3) 151 (18.4) 20 (2.4) 819 (100)

Table 10.14. Reasons for adults’ inability to access education Stated Reasons Poverty Unavailability of schools Cultural background Household chores Unawareness of parents Physical disability Lack of transportation Child marriage Fear of education School environment Other reasons Total

Number of Individuals 82 20 05 35 49 06 10 07 34 03 43 294

% 27.89 6.80 1.70 11.90 16.67 2.04 3.40 2.38 11.56 1.02 14.63 100.00

Source: Primary survey

IV.3. Education profile of the children Very few children enter college-level education, and a large proportion is at the lower primary level. As we go up the ladder of education, the number of children receiving education goes down. Although the demographic profile will have significant implications on the enrolment

Chapter Ten

192

profile of children in various classes, we can make some inferences on the dropout of children from Table 10.15 below. The fall in the number of children at higher school levels partly reflects a high dropout rate. This is evident from the fact that no-one from the village is going to college. This also became clear during discussions with the villagers. Table 10.15. Distribution of children currently in school Class First Second Third Fourth Fifth Sixth Seventh Eighth Ninth Tenth Total

Number of Children 32 18 17 14 13 12 09 14 07 11 147

Valid % 21.8 12.2 11.6 9.5 8.8 8.2 6.1 9.5 4.8 7.5 100.0

Cumulative % 21.8 34.0 45.6 55.1 63.9 72.1 78.2 87.8 92.5 100.0

Source: Primary survey.

IV.4. Dropout and reasons for it One of the important objectives of SSA was to reduce the dropout of children from primary schools to zero and thus ensure a 100% retention rate. Nevertheless, in our survey village we found that 22.5% of children aged 5–14 are still illiterate. We asked the parents of illiterate children for the reasons behind this, 40.32% of who opined that poverty still acts as a major obstacle to their children’s education (see Table 10.16 below). In poor families, children support the family in several ways. First of all, female children assist their mothers in household chores. A few households pointed out that when the mother goes to collect firewood or do some other work, they have to stay at home to look after the younger children or do household chores. Meanwhile, boys have to go to the fields with their fathers to help in the agricultural activities. When the child has to assist the family in eking out its basic livelihood, free education provided by the government seldom motivates poor parents to send their wards to school. Fear of education also remains a major factor for children dropping out of school, and 35.48% opined that due to this they stopped attending. Unawareness of the parents about the importance of education

The Distant Goal of Universalising Elementary Education

193

and lack of transportation facilities are the reasons for the dropout of 14.52% and 9.68% of children, respectively. Of 62 children who have dropped out of school, 40 are aged 5–14 and the rest are under 16. Table 10.17 below shows the present status of the dropped-out children. Of the 62, 16.13% help in domestic work, 9.68% are engaged in agriculture, 4.84% are collecting firewood, 8.06% are working as labourers and 17.74% are involved in other activities, while 43.55% do not do any work. Table 10.16. Reasons for children’s school dropout Stated Reasons Poverty Unawareness/ignorance of parents Lack of transport facility Fear of Education Total

Number of Children 25 06 09 22 62

% 40.32 9.68 14.52 35.48 100

Source: Primary survey.

Table 10.17. Present status of children who have dropped out of school Present Occupation Helping in agriculture Helping in domestic work Collecting firewood Not doing anything Labour Other Total

Number of Children 6 10 03 27 05 11 62

% 09.68 16.13 04.84 43.55 08.06 17.74 100

Source: Primary survey.

Due to the deterioration of educational instruction in the schools and increasing parental expectations, the demand for private tuition in the country is growing rapidly, and its need is even greater for the firstgeneration2 educated children. In our survey village, only 29.25% of the school-going children attend private tuition (see Table 10.18 below). Compared with any developed region, this percentage is much lower. The low percentage of children having private tuition in Bhagabatipur village does not necessarily reflect a better quality of education; the reason might 2

Sen (2009) used this concept for school-going children from families where nobody had previously been in education.

194

Chapter Ten

be something different, such as the economic poverty of the households. In a village where households struggle to provide two square meals a day throughout the year, private tuition would be beyond their imaginations. Moreover, the importance of education has not been realised in this village. Since most of the households belong to the ST community, the government provides them with free education by setting up ashram schools. Nevertheless, households have not been able to fully reap the benefits of these. Who assists the children in their home work if they do not have private tuition? In 19% of families it was reported that no homework is given to their children (see Table 10.19 below). In another 30% of cases, nobody helps the children with their homework. For 16.33% of children, fathers help the children with some homework. Similarly, mothers, brothers and sisters also help the children with homework in 5.44%, 7.48% and 1.36% of cases, respectively. The lower share of mothers’ assistance with homework indicates the low level of education of women in the village. Similarly, the households who report that no homework is given to their children inform us that parents are either not aware of the problems faced by their school-going children or are not bothered about them. Table 10.18. Students receiving private tuition Private Tuition Yes No Total

Number of students 043 104 147

% 29.25 70.75 100

Source: Primary survey.

All households with school-going children reported that they receive free dress materials and books from the schools. Similarly, schools also provide midday meals. Out of 81 households sending their children to school, 11 reported that their child does not eat a midday meal at school (see Table 10.20 below). The reason for this is quite obvious—the quality of the food is not good (another 39 households opine that the quality of the food is average).

The Distant Goal of Universalising Elementary Education

195

Table 10.19. Who assists the child with their homework? Assistance from Father Mother Brother Sister Relatives School Teacher Tutor Nobody No homework Total

Frequency 24 08 11 02 05 13 12 44 28 147

% 16.33 05.44 07.48 01.36 03.40 08.84 08.16 29.93 19.05 100

Source: Primary survey.

To assess the functioning of the schools, we asked a few questions as to whether there is corporal punishment, whether teachers regularly attend, and whether the households attend parent-teacher association (PTA) meetings. The results are presented in table 10.21 below, showing that 71.6% of households opine that there is no corporal punishment in the school. Around 75% report that teachers come to school regularly. Only 60.49% of the households say that they attend PTA meetings, and such a low participation is a matter of serious concern. This could be due to the lack of time available for the parents or their sheer negligence or ignorance. Given the high illiteracy levels of the parents, it is quite possible that it is the latter, and this could be reflected in the fact that a large proportion of the male population drinks alcohol, even in the daytime. Table 10.20. Households’ perceptions of the midday meal Eat in school Do not eat 51 11 Household perceptions of food quality Good Average 31 39 Source: Primary survey.

Not applicable 19 Bad 11

Chapter Ten

196

Table 10.21. Households’ opinions on and participation in the functioning of schools Households’ opinion Corporal punishment Regularity of teacher attendance Attending PTA meetings

Yes 23 (28.4%) 61 (75.31%)

No 58 (71.60%) 20 (24.69%)

32 (39.51%)

49 (60.49%)

Source: Primary survey.

IV.5. Supply side of elementary education In Bhagabatipur an Ashram School was set up in 1984. Currently, 388 students are studying in this school. The school provides residential facilities; and310 students are staying in the hostel, of which 200 are boys and 110 girls. All come from different tribal districts of Odisha. From Bhagabatipur village, only 15 students are staying at the hostel. The school has two buildings at two locations; one is located in the village where only children from Bhagabatipur study. For higher classes, children have to walk about one kilometre to reach the main building of the Ashram School, where residential education is provided. There are only four teachers for these 388 students. To address this shortage, the village education committee (VEC) has appointed one teacher in the village school. Since the school provides residential education, and seeks to provide vocational training, it needs at least 12 teachers, according to the school headmistress Kumari Baijayanti Mishra. Previously, there was a Hindi teacher, an agriculture teacher and a tailoring teacher, but they were transferred and their posts have remained vacant. The headmistress also pointed out that there are seven sanctioned posts, of which three are vacant. Apart from teaching, the teachers have to look after the boarders and are supposed to stay on campus. However, due to a lack of proper housing facilities, only two teachers stay on campus, the others commuting from Bhubaneswar. Therefore, it is difficult for two teachers to look after all the students and handle administrative affairs; teaching therefore gets the lowest priority, and as a result the quality of education is affected. Although there are a number of students who are talented sportspersons, due to the lack of sports teachers they get no specialised training for participation in competitions at higher levels. The lack of infrastructure also affects the teaching and learning. There are not enough hostel rooms. Therefore, boys are admitted into the hostel from class four and girls from class one. Due to this demand, classrooms

The Distant Goal of Universalising Elementary Education

197

are also used as hostel rooms. During our visit to the school we found that students kept their baggage near the walls, and after the class they used the same room as a dormitory. The dining hall is also used as a classroom. There is only one toilet each for 110 female students and 200 males and no bathroom, meaning that girls have to take a bath in the open. Due to some internal problems, a VEC was not formed last year. Although a school managing committee existed, most members played a passive role. In order to comply with government rules, 50% of the committee members are women, but they seldom play any role except as signatories—the chairperson, vice-chairperson, ward members and teacher stake the decisions.

V. Conclusions and Policy Recommendations V.1. Summary of the findings The objective of the study was to assess the progress of SSA in Odisha and identify the constraints faced by the households in providing their children with education up to at least class 8. For this purpose, we undertook a survey of 173 households in Bhagabatipur village. At the outset we gathered information on the socio-economic condition of each household. Then we collected information on the education of adult members of the households and the reasons behind their inability to attain at least primary education and, if possible, higher education. Finally, we assessed the progress of SSA in Bhagabatipur and the reasons behind the high dropout rate. We explored both supply and demand constraints. In Bhagabatipur, 22.5% of children aged 6–14 are illiterate. This says a lot about the failure of SSA and RTE in schooling all children. Furthermore, 40.32% of illiterate children’s parents opined that poverty still acts as a major barrier to their education. In poor families, children support their families in several ways. First of all, female children assist their mothers in household chores. A few households pointed out that when a mother goes to collect firewood or does some other work, the girl has to be at home to look after younger children or do household chores. In the case of boys, they have to go to the fields with their fathers to help them in their agricultural activities. When the child has to assist the family in eking out a basic livelihood, the free education provided by the government does not make schooling attractive for the parents. Fear of education also remains a major factor for school dropouts, and 35.48 % of the children opined that due to fear of education they did not continue their education. A lack of awareness of parents about the importance of

198

Chapter Ten

education and lack of transportation facilities are also reasons for the dropout of 14.52% and 9.68% of children, respectively. Out of 62 dropped-out children, 16.13% help in domestic work, 9.68% are engaged in agriculture, 4.84% collect firewood, 8.06% work as labourers and 17.74% are involved in other activities, while 43.55% do not do any work. Around 75% of the households opine that teachers attend school regularly. Only 60.49 % of households say that they attend PTA meetings. The low participation of households in such meetings is a matter for serious concern and could be due to a lack of available time or sheer negligence or ignorance. Given the high levels of illiteracy, it is quite possibly the latter, and could stem from the fact that a large proportion of the male population drinks alcohol, even in the daytime. On the supply side, we found inadequacies of teachers and classrooms, and that hostel rooms serve as the major barrier in providing quality education. Similarly, a lack of bathrooms and insufficient toilets also raise the issue of the dignity of children. It is observed that tribal students perform well in sports, but due to the lack of sports teachers their potentials are not realised.

V.2. Policy recommendations Based on our research findings we can make the following policy recommendations: (1) All vacant teaching and non-teaching posts must be filled (2) Keeping in view the large number of students, more class and hostel rooms must be constructed (3) Serious efforts must be undertaken to create awareness among rural households regarding the importance of education. (4) Since rural poor households depend on their children for income generation, the government incentivizes the poor households to send their children to school. (5) The government should undertake an alcohol awareness campaign to help tribal households save money for productive investments. (6) Tribal households should be provided with land records for their homestead plots and agricultural lands.

V.3. Conclusions The overall literacy rate of Odisha is close to the national average. Nevertheless, if the state is to emerge as developed, it has to increase the

The Distant Goal of Universalising Elementary Education

199

literacy rate and minimise the dropout rate substantially. The matter is serious in the cases of SC/ST households. Although both central and state governments have introduced several measures, these have not yielded satisfactory results. Recently, the education department has strengthened the monitoring mechanism. However, it is observed that due to a shortage of regular teachers and other non-teaching staff, working teachers are overburdened as they are saddled with a number of non-teaching tasks. Thus, the quality of education is severely affected. As per the 2001census, SC and ST populations comprise 16.53% and 22.13% of the total population, respectively. The disadvantages they face are clearly indicated by the lower literacy rates. As per the 2001 census, the overall literacy rate of the state is 63.08%, whereas among the SC and ST communities the rates are 55.53% and 37.37%, respectively. Given the special disadvantages faced by the SC/ST population, and their huge combined share in the total population (around 40%), the government needs to initiate special measures to improve their current statuses.

References District Information System on Education. 2011. Odisha State Report Card 2009–10. http://dise.in/src.htm. Government of Odisha. 2011. Status of Elementary Education in the State. Department of School and Mass Education, http://www.odisha.gov.in/schooleducation. Sen, A. 2009. Behind the Differential Reach of Primary Education. The Hindu. December 9, http://www.thehindu.com/opinion/op-ed/behindthe-differential-reach-of-primary-education/article 67453.ece.

CHAPTER ELEVEN GENDER DISPARITY IN EMPLOYMENT IN ODISHA: SOME EVIDENCE FROM NSS DATA SMRUTIREKHA MOHANTY

I. Introduction The history of gender studies is replete with incidents of discrimination and evidences that show women being perceived as second to men (Duraisamy & Duraisamy 1999; Sabir & Aftab 2007; Wright & Ermisch 1991). Gender-discriminatory practices have been reported to be more distinct in developing countries than in developed ones (Volart 2004). There is a huge productivity loss in the economy due to the difficulties in capturing women’s varied and very complex activities. Various socioeconomic problems like discrimination based on caste and poverty combined with societal attitude towards women add to their vulnerable condition. One of the important indicators for evaluating the status of women in society is their economic wellbeing. In the absence of any direct measuring rod or channel to assess the socio-economic status of women, a study of the employment structure will be of some help. The majority of women workers in India belong to the lowest economic strata where they offer themselves to the labour market for their subsistence. Due to the preexisting social structure and stigma, women workers are subjected to discrimination from the very first step of hiring until they get their wages. The prevalence of wage discrimination against women workers has been confirmed by different studies time and again (Gardeazabal & Ugidos 2005; Jarrell & Stanley 2004; Paul & Paul 2012; Sengupta & Das 2014). Thus, it is necessary to study the structure of the female workforce, including the factors and processes that lead women to work. It is also pertinent to highlight the discriminatory practices against women workers

Gender Disparity in Employment in Odisha

201

in the labour market to have a comprehensive understanding of their socioeconomic statuses.

II. Objectives and Rationale behind the Study The Indian labour market witnessed a decline in the number of women before the 1980s; however, this trend was reversed during the postliberalization period. During the 1990s, a clear rise in the employment of women is observed, and is corroborated by a sharp decline in the share of domestic work they did. An observation of the pattern of women employment in India reveals that most are principally engaged in the lowpaying informal sectors. Female employment in India is conditional, and various socio-economic factors affect it. Odisha is one of the major states in India and bears the special characteristics of mass poverty and distinct regional inequalities. The human development indicators in Odisha vary widely across the districts and social groups. Gender disparities are marked across districts with skewed sex ratios. All these unfavourable factors interact with each other, making the regional disparity more pronounced and awful. In this context, it will be interesting to study the state of affairs in the labour market with respect to female workers and their conditions in the workplace in comparison with those of men. It also becomes necessary to investigate the dynamics of the participation of women in the workforce in the context of various socio-economic and household characteristics. The objective of this paper is to bring forth the determinants of the participation of women in the workforce and to analyse the pattern of gender disparity in employment in Odisha.

III. Data and Methods This study primarily draws upon reports on the status of employment and unemployment in India published by the National Sample Survey Organisation in four different periods: 1993–4 (50th round), 1999–2000 (55th round), 2004–5 (61st round) and 2009–10 (66th round). The unit records of the above-mentioned rounds were collected and used for this paper wherever the published reports failed to provide a deeper understanding of the situation. While the information from these rounds was used to ascertain the trends, patterns and structures of various labour market variables in Odisha, the unit records of the latest round for the year 2009–10 were rigorously analysed to carry out a regression analysis for the identification of factors determining the status of the participation of women in the workforce. In the period 2009–10, about 12,997 and 4,260

202

Chapter Eleven

people of rural and urban areas, respectively, of Odisha were surveyed. The sample size of women for the binary non-linear regression model in this survey is 8,493.

IV. Literature Review The dynamics of female employment in India, as pointed out by previous research, are quite varied and interesting. Presented below is a short thematic survey of the literature on the pattern and structure of female employment in India.

IV.1. Trends and patterns of female employment Studies have highlighted the differential impacts of location and education on women’s participation in work. The crude rate of participation of women is much higher among rural women than among urban women. Although the rate among rural men is higher than that among urban men, this rural-urban difference is much higher in the case of women. Likewise, the rate of work participation is higher among illiterate women than among literate women, falling in the transition from illiteracy to literacy and not climbing again until the level of education rises to matriculation and above (Nath 1968).

IV.2. Structure of female employment Women are better represented in the traditional occupations and those in which the typical enterprise is the household or small-scale industries than in more modern ones and those with large enterprises (Nath 1968). In the agricultural sector, women are employed during the peak season for sowing, cotton picking and harvesting, as and when required. The existing structure of female employment is more applicable to low-paying unskilled and semi-skilled jobs (Kak 1994). Considering the nature and extent of employment as indicators of the status of women in a capitalist economy, Kak shows that large sections of women in India have been relegated to marginal and subsistence-oriented activities in rural areas. The agricultural sector in India is over-populated by women workers whereas there is a completely opposite situation in most developed countries, where agriculture is primarily a male activity and the proportion of workers in a non-agricultural sector is much higher for women than for men (Nath 1968).

Gender Disparity in Employment in Odisha

203

IV.3. Wage disparity Traditionally, women in less-developed countries were confined to the four walls of their homes, but in the recent decades more and more women have come to be part of the labour market. This change from mainly inactive to active agents depending on the agricultural season indicates that women are increasingly employed as a part of the unpaid family labour force. Where women work for wages, these are barely above the subsistence level and are much less than their male counterparts get. The wages that women get for casual labour and all agricultural and nonagricultural operations are lower than for men, and this difference has increased with mechanized and non-manual operations. The wage difference between male and female workers persists and has become larger in recent decades, being generally more pronounced in the agricultural sector where the rate of wage increase for male agricultural workers is higher (Kak 1994). In some sectors, women workers are paid less than men for the same operations, but this gap has narrowed. There is not only wage discrimination but also limited secure employment, restricted to peak seasons and certain categories of operation for women workers (Nayyar 1987).

IV.4. Causes of the casualization of the female workforce Contrary to the common perception, international data reveal that the highest wages for women are not in the highly industrialized economies but in the least developed with a per capita income of less than $100. This can be explained by the predominance of household agricultural and nonagricultural enterprises in which productive work can be conveniently combined with family responsibilities (Nath 1986). In agriculture, women are pushed into casual and low-paying work, while in the emerging industrial sector they are largely a part of the unskilled and low-wage sectors. The introduction of machines takes jobs from women and pushes them further into the unskilled labour pool. Changes in production conditions and the differential impact of development policies, given the pre-existing economic and social stratification, create unfavourable conditions for the participation of women in the labour force outside the domestic arena. The setting up of the modern industrial and agricultural sectors based on capital-intensive technology has compelled women to move into the low-wage sectors that form the bulk of unskilled labour. A decline in the number of jobs in the self-employed category and the unavailability of jobs in the organized sector have left women with no

Chapter Eleven

204

option but to enter the labour market as casual workers. The participation of women in the economy of a country is largely a function of the socioeconomic conditions interlinked with the stages of development. A large section of the population belonging to the oppressed class and with lower consumption brackets offers itself for jobs outside the domestic activities. Over the last four decades, the domination of the casual labour force in female work has increased at a much faster rate than in male work (Kak 1994). According to Nayyar (1987) “Female participation rates are highly correlated to poverty and landlessness in rural India. It has been observed that, in areas where incomes had gone up, consequent on the Green Revolution, women tended to withdraw from the labor market due to improved income effect.”

V. Gender Disparity in Odisha V.1. Trends in rates of female work participation and unemployment A society’s progress is assessed in terms of whether women get equal opportunities in every sphere of life. One of the very important spheres is the economic independence of women. In this study, the aspect of economic independence of women in their empowerment has been examined by using an employment variable (captured by WorkerPopulation Ratio (WPR), which is a direct measure of income and therefore a crude method for assessing economic wellbeing.

WPR in %

Fig. 11.1. Trend of WPR of men and women in Odisha by sector (1993–2009) 70 60 50 40 30 20 10 0 1993-94 1999-00 2004-05 2009-10 Rural Male

Rural Female

Source: NSS Reports on Employment and Unemployment Situation in India, various rounds.

Gender Disparity in Employment in Odisha

205

The high level of employment (WPR) among men, both in rural and urban Odisha and all-India, indicates the general level of employment opportunity in the economy. It is as high as 58% in both the rural and urban sectors of Odisha for men (compared to 55% and 54% in rural and urban India, respectively). However, it is also evident from the data that women are not able to avail themselves of employment opportunities in the state, because in 2009–10 24% of rural women and 12% of urban women were in the workforce of the state (these figures are two percentage points lower than the all-India levels). The trend of female employment reveals a discernible picture of the period. As can be seen from the above figures, the rate at which women participate has been declining in both the rural and urban areas of Odisha. Although until 2004–5 the WPR of women has marginally increased and mostly remained stable at a low level, the subsequent decline has been alarming and sharper among rural women. However, in contrast to the trend of female employment, there has been a gradual expansion in the employment opportunities for men, as exhibited by the sustained rise in the WPR of men since 1999–2000, mostly among urban men (see Fig.11.1 above). Table 11.1. Unemployment rates in India and Odisha by sex and sector (1993–2009) Years (NSS rounds) 1993–4 1999–00 2004–5 2009–10

Odisha Rural

Urban

India Rural

Male

Female

Male

Female

Male

Female

Male

Female

1 2.4 3.1 3.1

0.2 1.1 8.3 2.7

3.6 7 9 4

1 5.3 26.6 5.4

0.8 1.7 1.6 1.6

0.3 1 1.8 1.6

2.2 4.5 3.8 2.8

1 5.7 6.9 5.7

Urban

Source: NSS reports on employment and unemployment situations in India, various rounds.

The rate of unemployment among women has remained at a constant higher level after 2004–5 compared to males. It should be noted that in 2004–5 the unemployment rate among women in urban Odisha recorded an all-time high of 27%, which eventually declined to 5% in 2009–10.

V.2. Structure of the female workforce After discussing the availability of employment opportunities in the state, it is important to study the structure and conditions of women

Chapter Eleven

206

workers, because these are important parameters in assessing the conditions in which they work. Table 11.2. The nature of the employment of men and women in Odisha (2009–10) Status of employment OAW Employer Helper Self Employed Regular Casual

Male 40.6 0.8 10.9 52.2 9.7 38.0

Female 11.2 0.6 42.1 53.9 4.1 42.0

Source: Computed from the unit records of NSS 66th round (2009–10). Note: OAW=own account workers.

The nature of the employment structure in Odisha shows the disadvantaged position of women workers. By definition, those who have a regular source of income are considered better off than those who are in contractual employment. The regular job holders are assured of their future earnings and most get social security benefits. The casual workers are never sure of their futures and are not organized. Most of the economically poor and vulnerable groups are engaged in these types of jobs. The self-employed category is ambiguous, consisting of both wellpaid entrepreneurs and businessmen at one end and low-paid rickshaw pullers and street vendors at the other. Table 11.2 above signifies that in Odisha about 54% of women are self-employed, which is two percentage points higher than men. However, the distribution of women within the sub-categories of self-employed displays a totally different picture compared to men. In all, 41% of men are own account workers, which is considered a better class of self-employment with respect to income and power, compared to only 11% of women. Furthermore, 42% of female workers are self-employed as unpaid household helpers, compared to 11% of men. These helpers are mainly engaged in repetitive and specified jobs. Regular employment among women is merely 4%, which is half that of their male counterparts (see Table 11.2 above).

V.3. Education and nature of employment Stigmatized social attitudes towards women working outside the home outweigh the benefits of education as an instrument of female empowerment.

Gender Disparity in Employment in Odisha

207

Table 11.3 below reveals that even if women in Odisha have received education up to secondary level and above, a major chunk are engaged in unpaid household work as helpers, thus undermining their productivity. About 33% of women with higher education are engaged in unpaid family work compared to only 17% of men. About 8% of women with higher education are engaged in insecure casual jobs in Odisha, and this may be one of the outcomes of the labour-market discrimination against women by employers, denying them secure and regular jobs (see Table 11.3 below).

V.4. Gender disparity in the labour market A direct method of confirming the role of social beliefs and practices in subordinating women to men involves comparing the monetary return both genders get for doing the same job. This system discriminates against women, wherein women doing regular jobs are paid differently for the same type of work and for an equal number of hours as their male counterparts. Women who are in regular salaried jobs are paid less than their male counterparts by 94% and 55% in the rural and urban areas of Odisha, respectively. In contrast to the above scenario of the regular job sector, the gender wage gap is less pronounced among those who are casually employed. Female casual workers are paid 37% less than male casual workers in both rural and urban Odisha (see Table 11.4 below). Table 11.3. General education and nature of employment of men and women in Odisha (2009–10) Gender / education

Men Not Literate Primary Middle Secondary and above

Women Not Literate Primary Middle Secondary and above

Categories of self-employment OAW Employer Helper

Total self employed

Regular

Casual

39.8 40.7 43.1 38.4

0.5 1.1 0.8 0.6

3.7 8.6 16.1 17.2

44.1 50.4 59.9 56.2

2.7 4.2 8.3 28.2

53.2 45.3 31.7 15.6

10.2 11.4 16.4 11.0

0.3 0.4 1.7 1.5

39.3 47.5 50.6 33.4

49.8 59.3 68.7 45.9

0.9 1.7 5.5 46.4

49.3 39.0 25.9 7.7

Source: Unit records of NSS 66th round (2009–10).

208

Chapter Eleven

Table 11.4. Average daily wages of men and women workers aged 15– 59 in Odisha by Sector (2009–10) Type of employment Regular Other than public work

Rural Male 293.87 81 83.42

Female 151.72 59.06 0

Urban Male 358.89 100.26 -

Female 238.48 72.8 -

Casually employed in public works Source: NSS Report on Employment and Unemployment Situation in India (2009–10).

A simple disparity index has been constructed to show the wage differentiation between men and women. This disparity index (D) is defined as the ratio of wages of men to those of women. D=Average daily wage of a male worker/ Average daily wage of a female worker If D=1; there is no wage disparity If D>1: wage disparity exists against female workers. For both types of jobs, the wage disparity index carried values greater than one, signifying the existence of wage disparity in the labour market of Odisha (see Table 11.5 below). Table 11.5. Values of D (Wage disparity Index) in Odisha by sector (2009–10) Type of employment Regular Casual

Rural 1.9 1.4

Urban 1.5 1.4

Source: Author’s calculation.

Women working in the agricultural sector (excluding crop production) and those in non-agricultural sectors also face indirect discrimination in the form of denial of a written job contract. In Odisha, about 80% of women workers are engaged in intermittent and unstable jobs that guarantee no formal contract. This figure is 2% higher than that of men. About 80% of women workers are not entitled to any paid leave for major inconveniences, whereas 27% of men are eligible for paid leave. Furthermore, 83% of women workers who are employed, the job being their major source of income, are denied any social security for their

Gender Disparity in Employment in Odisha

209

future. This figure is 10% higher than their male counterparts (see Table 11.6 below). Table 11.6. Various decent work indicators for workers in Odisha (2009–10) Decent work indicators Job Contract No written contract Contract for < 1year Contract for < 3years Contract for >3 years Paid leave Eligible Not eligible Social security Available Not available

Men

Women

78.3 2.0 0.4 19.3

80.0 4.2 1.9 13.9

27.3 72.7

20.0 80.0

27.2 72.8

17.6 82.4

Source: Unit records of NSS 66th round(2009–10).

V.5. Income and female employment The literature suggests that in underdeveloped countries it is insulting and demeaning for women to work outside of their homes, and women belonging to the upper classes are mostly confined within these four walls. This notion is reflected in many sociological studies. Working outside the home in traditional societies is considered an indicator of deprivation and poverty. This present study confirms that the effect of income has worked well among women in Odisha. The graphs exhibit a strong negative correlation between income (captured by MPCE) and participation in work among women in both rural and urban Odisha. As income increases the WPR among women falls, and this fall is sharper for urban women in Odisha. However, there is no such relationship between income and participation for men in work (see Figs. 11.2 to 11.5 below).

Chapter Eleven

210

WPR

Fig. 11.2. MPCE class-wise WPR of men in rural Odisha (2009–10) 66.0 64.0 62.0 60.0 58.0 56.0 54.0 52.0 50.0 48.0

MPCE Classes

Fig. 11.3. MPCE class-wise WPR of women in rural Odisha (2009–10) 35.0 30.0 WPR

25.0 20.0 15.0 10.0 5.0 0.0

MPCE Classes

Gender Disparity in Employment in Odisha

WPR

Fig. 11.4. MPCE class-wise WPR of men in urban Odisha (2009-10) 64.0 62.0 60.0 58.0 56.0 54.0 52.0 50.0 48.0 46.0

MPCE Classes

WPR

Fig. 11.5. MPCE class-wise WPR of women in urban Odisha (2009-10) 18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0

MPCE Classes Source: Unit records of NSS 66th round(2009–10).

211

212

Chapter Eleven

VI. Estimation and Results VI.1. Specification of the model The estimation method applied in this paper is a Logit regression analysis, allowing us to establish the non-linear relationship between dependent and independent variables. This method has the property of following the normal cumulative density function, which is useful for many practical purposes. In our model, the dependent variable is categorical and the regressors are of a mixed nature, i.e. a mixture of categorical and continuous variables, mainly covering some socioeconomic and demographic indicators corresponding to each individual observation. The dependent variable here is defined as the probability of a woman participating in the workforce, which is a dummy variable. The quantitative independent variables are household size, number of members in the household who are in non-agricultural occupations, number of members in the household who are in regular employment, and the monthly per capita consumption expenditure (MPCE) of the household. The qualitative independent variables (dummy variables) in this model are sector (the place of residence), marital status, caste (social group), religion and general education.

VI.2. Results The logit regression model estimation (see Table 11.7 below) reveals that most of the variables taken into consideration are highly significant in determining the dependent variable. All independent variables do carry the expected and desirable signs, thus maintaining the usual and logical relationship in the model. Setting all the explanatory variables at their mean levels, the model estimation shows that there is an inverse relationship between household size and the probability of a woman being employed. As the household size increases, the probability of a woman being employed decreases by 0.79 units, and this estimation is highly significant. The number of members in non-agricultural employment positively impacts the probability of a woman’s employment status. If the number of members in a household engaged in non-agricultural activities increases by 1 unit, the probability of a woman from that household working increases by 1.6 units. The higher participation in non-agricultural activities reflects the movement of a society from a traditional to a modern one, which

Gender Disparity in Employment in Odisha

213

facilitates and encourages women’s work engagement. This result is also highly significant. Setting all the explanatory variables at their mean levels, a woman in urban Odisha is 0.5 times less likely to be employed compared to a woman in rural Odisha. The significance of this causal relationship is also very high. In Odisha, those who are never married, separated, divorced or widowed are less likely to be employed compared to those who are married. Those never married are 0.24 times less likely to be employed compared to the base category. Religion, as an explanatory variable, has no significant effect on the probability of a woman’s work status. However, caste has an important bearing on the dependent variable. Table 11.7. Logit estimates for Odisha (2009–10) Independent variable Household size Number of non-agricultural workers in the household Number of regularly employed members in the household MPCE Sector Rural® Urban Marital Status Married ® Never married Others Social Group Others® Scheduled Tribe (ST) Scheduled Caste(SC) Other Backward Class(OBC) Religion Hindu® Islam Christian General Education Secondary and above® Not Literate Literate up to primary Literate up to middle Constant

Coefficient -.2633973*** .498595***

Odds ratio .7917492*** 1.654058***

.0470424

.9462281

-.0004853***

.9995148***

-0.5778***

0.5030***

-1.4278*** -0.0869

0.2490*** -0.9075

1.5267*** 0.8384*** 0.7283***

5.3409*** 2.5749*** 2.1906***

0.0734 0.2571

1.1255 1.3001

0.2206** -0.133 -0.2027* -.5104581***

1.5419** 1.0237 0.9158*

214

Chapter Eleven

LR chi2(15) 1542.39 Prob>chi2 0.0000 Pseudo R2 0.1822 Log likelihood -3462.5906 No. of observations 8493 Note: dependent variable: workforce status of a woman; 0= not in the workforce, 1= in the workforce, ®: Reference category. *** Significant at 1% level, ** Significant at 5% level, * Significant at 10% level.

Women in the ST, SC and OBC categories are respectively 5.3, 2.5 and 2.1 times more likely to be employed compared to the reference category of women in the general caste. All the estimates of caste dummy are highly significant. The education of women is also an important factor in determining the workforce status. Women who have attained education up to middle level are 0.9 times less likely to be employed than illiterate women, who are 1.5 times more likely to be employed compared to the reference category of higher educated women. One of the reasons for the higher participation of illiterate women in the labour markets maybe their chronic poverty and vulnerability. Setting all the explanatory variables at their mean levels, it is estimated that as income increases the probability of a woman participating in the workforce declines by 0.9 times (see Table 11.7 above). Due to improved income, women withdraw or abstain from participating in the workforce of Odisha.

VII. Conclusion This study substantiates the fact that in Odisha religion has no effect on female participation in the workforce. Even education bears a lesser effect on a woman’s work participation status. The significant variables describing a woman’s work status are largely dependent on social attitude. Caste, economic condition of the household and progress in family attitude (captured by the variable of number of members in the household who are in non-agricultural employment) are the main determinants of the participation of women in work. While social progression in a family encourages its girls to engage in economic activities, improved economic conditions, on the other hand, restrict a woman’s employability in Odisha. The confinement of women within the four walls of the home has remained a status symbol in Odisha, which until now has had a strong hold over society. Poverty-ridden households send their girls and women to work from necessity. Along with low job profiles among most of the women, the discriminatory practices aggravate their vulnerable condition in society. In order to make women equal to men in the economic sphere

Gender Disparity in Employment in Odisha

215

of life in a state like Odisha, the first necessary conditions are bringing about reforms in the broader social attitudes of people with respect to female empowerment and getting rid of the evils of caste discrimination.

References Abraham, V. 2009. Employment Growth in Rural India: Distress Driven? Economic and Political Weekly. 44: 97–104. Aggarwal, S. C. 2004. Child Labour and Household Characteristics in Selected States: Estimates from NSS 55th Round. Economic and Political Weekly. 39: 173–85. Arjan, D.H. & A. Dubey. 2005. Poverty, Disparities, or the Development of Underdevelopment in Orissa. Economic and Political Weekly. 40: 2321–29. Datt, G. & M. Ravallion. 2002. Is India's Economic Growth Leaving the Poor Behind? The Journal of Economic Perspectives. 16: 89–108. Dayal, E. 1989. Rural Poverty in India: a Regional Analysis. Journal of Rural Studies. 5: 87–98. Duraisamy, M & P. Duraisamy. 1999. Gender Bias in the Scientific and Technical Labor Market: A Competitive Study of Tamil Nadu and Kerala. Indian Economic Review. 34: 149-69. Gardeazabal, J. & A. Ugidos. 2005. Gender Wage Discrimination at Quantiles. Journal of Population Economics. 18:165-79. Ghose, J. 2005. .Integration of Gender Perspectives in Macroeconomics. Commission on the Status of Women. Forty-Ninth Session, United Nations. Government of India, National Sample Survey Organisation. October 1996. Key Results on Employment and Unemployment, 50th Round 1993–4, Report No: 406. —. December 2000. Key Results on Employment and Unemployment, 55th Round 1999-2000. Report No: 45555/10/1. —. September 2006. Employment and Unemployment Situation in India, 61st Round 2004–5. Report No: 51561/10/1. —. June 2011. Key Indicators of Employment and Unemployment in India, 66st Round 2009–10. Report No: NSS KI 66/10. Gupta, N. D. & A. Dubey. 2006. Fertility and the Household's Economic Status: A Natural Experiment Using Indian Micro Data. Journal of Development Studies. 42: 110–38.

216

Chapter Eleven

Jarrell, S. B. & T. D. Stanley. 2004. Declining Bias and Gender Wage Discrimination? A Meta-Regression Analysis. Journal of Human Resources. 39: 828-38. Jeemol U. 2001. Gender and Informality in Labour Market in South Asia. Economic and Political Weekly. 36: 2360–77. Kak, S. 1994. Rural Women and Labor Force Participation. Social Scientist. 22: 35–59. Lieten, G. K. 2002. Child Labour and Poverty: The Poverty of Analysis. The Indian Journal of Labour Economics. 45: 451–64. Madheswaran, S. & P. Attewell. 2007. Caste Discrimination in the Indian Urban Labor Market: Evidence from the National Sample Survey of India. Economic and Political Weekly. 42: 4146–53. Nath, K. 1968. Women in the Working Force in India. Economic and Political Weekly. 3: 1205–13. Nayyar, R. 1987. Female Participation Rates in Rural India. Economic and Political weekly. 22: 2207–16. Panda, P. K. 1997. Female Headship, Poverty and Child Welfare: A Study of Rural Orissa. Economic and Political Weekly. 32: WS73–82. Parikh, A. & K. Sen. 2006. Probit with Heteroscedasticity: an Application to Indian Poverty Analysis. Applied Economics Letters. 13: 699–707. Paul, S. & S. B. Paul. 2012. Trade Reforms and Gender Wage gap in India. http://paa2013.princeton.edu/papers/131268. Sabir, M. & Z. Aftab. 2007. Dynamism in the Gender Wage gap: Evidence from Pakistan. The Pakistan Development Review. 46: 865-82. Sengupta, A. & P. Das. 2014. Gender Wage Discrimination across Social and Religious Groups in India: Estimates with Unit level Data. Economic and Political Weekly. XLIX:71-76. Visaria, P. 1981. Poverty and Unemployment in India: An Analysis of Recent Evidence. World Development. 9: 277–300. Volart, B. E. 2004. Gender Discrimination and Growth: Theory and Evidence from India. http://eprints.lse.ac.uk/6641/1/Gender_Discrimination_and_Growth_T heory_and_Evidence_from_India.pdf Wright, R.E. & J. F. Ermisch. 1991. Gender Discrimination in the British Labour Market: A Reassessment. The Economic Journal. 101:508-22.

CHAPTER TWELVE PARTICIPATION OF THE CENTRAL GOVERNMENT IN RURAL DEVELOPMENT OF INDIA: AN ANALYSIS OF THE UNION BUDGET (1981–82 TO 2008–09) ANIMESH KUMAR

I. Background The Indian government’s involvement in the rural sector is crucial for many reasons. One of these is the fact that it runs several poverty alleviation and public works programmes with the objective of providing food security and employment to the poor. These programmes also tend to reduce inequality amongst people as they do not discriminate on the basis of religion, caste, sex or age. Public investment in rural infrastructure, agricultural research, rural health and education indirectly results in agricultural and non-agricultural growth along with the reduction of poverty. Therefore, one way to eradicate poverty is to raise public investment in rural areas and agriculture. It is difficult for the market to provide services in rural areas because of their low profitability and lack of participation in market activities. Even if the market exists, a majority of the rural population is denied access because of the high cost associated with such services. Post liberalization, there has been greater reliance on market forces, while government interventions have been decreasing over time. There have also been concerns that pro-poor spending may be reduced under the fiscal adjustment policy implemented under the reforms. This makes a strong case for analysing government expenditure in the rural sector. The planned and effective allocation of a country’s resources amongst its different sectors is necessary for inclusive growth in the economy. Budgetary resources allocated by the central and state governments are an

218

Chapter Twelve

important source of finance for rural development. A study of the expenditure strategies reveals changes at various levels and the direction of the flow of public funds in the economy. This reflects the priorities of the government as well as the factors that contribute to the changing priorities over time. Management or administrative aspects are also very important for the efficient delivery of the programmes, as allocation of funds alone will not guarantee a significant impact on the living conditions of rural people. The nature of expenditure, i.e., whether it is productive or not, is also very important in influencing the conditions of rural people. The union budget is the main source of data for this study and is considered for two reasons. Firstly, budgets are crucial policy documents that reveal the social and economic priorities of the government. Although a significant proportion of the expenditure is carried out by state governments, the centre plays an important role as it has considerable influence over the state governments in a federal system like India. Secondly, volume II of the expenditure budget provides separate data for different programmes initiated for the development of the rural sector. An analysis of “rural development” and “social services” can also be done by using the data on functional or budget classification, but this classification does not provide data for urban and rural components separately. Data for revised estimates of the period from 1981–2 to 2008–9 are considered. To observe the nature of the flow of public resources to rural areas through budgetary expenditure, only those heads of expenditure exclusively for the development of rural areas are considered. There are many other programmes that are important to both urban and rural sectors, but it is difficult to separate the rural component from the total expenditure under these heads. Investments on education, electricity and remote sensing technology are examples of such expenditures.

II. Trends in the Rural Expenditure of the Central Government India entered into a programme of fiscal adjustment in 1991–2, which continues to guide its economic policy. On the one hand, it was apprehended that the new policies may affect the public expenditure negatively, but on the other it was claimed that one of the aims of the process of economic reform was to withdraw the state from some of its economic activities in order to increase social sector expenditures. To analyse the differences in trends in the pre- and post-reform periods, one needs to look at expenditure during the two periods separately.

Participation of the Central Government in Rural Development

219

In this study, the sum of all allocations to the Ministry of Rural Development,1 Government of India, is referred to as Rural Development Expenditure.2 Figs. 12.1 and 12.2 reflect the trend in the share of Rural Development Expenditure in GDP over time. The graphs are drawn on the basis of the value of total expenditure obtained from the Ministry of Rural Development for the period 1981–2 to 2008–9.3 Fig. 12.1. Trend in the share of rural development expenditure in the pre-reform period (% of GDP)

0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00

rural devt. expenditure Linear (rural devt. expenditure ) Source: Constructed from data obtained from the union budget (Government of India).

Fig. 12.1.shows an overall increase in the share of rural development expenditure for this period. It is found that the share has increased continuously from 0.24% of GDP in 1981–2 to 0.70 % of GDP in 1989– 1

Agriculture Marketing was transferred to the Ministry of Agriculture from 1999– 2000. The allocation for agricultural marketing for the year 1999–2000 onwards is therefore taken from the Department of Agriculture and Cooperation, Ministry of Agriculture. 2 In annual budgets, rural development expenditure is taken as the sum of special programmes for rural development, rural employment, land reforms and other rural development programmes. Rural development expenditure in this study includes the allocations for the above four categories along with other components. 3 See Annexures 1 and 2.

Chapter Twelve

220

90, with minor fluctuations. During this period, only at one point in 1988– 9, did the share decrease from the previous year? After 1989–90, the share of rural development expenditure exhibits a cyclical pattern until 2007–8, when the share became 0.87% of GDP. This means that the share tripled in the first ten years, but during the next seventeen years there was only a minor increase. The share of less than 1% is not a healthy sign given its important role in the development of the rural sector. A sudden increase in the share is observed in 2008–9 due to the implementation of the National Rural Employment Guarantee Act (MNREGA). Overall, the share of rural development expenditure in the budget fell in the post-reform period. Fig. 12.2. Trend in the share of rural development expenditure in the post-reform period (% of GDP)

1.50 1.00 0.50 0.00

rural devt. expenditure Linear (rural devt. expenditure ) Source: Constructed from data obtained from the union budget (Government of India).

III. Rate of Growth of the Central Government Expenditure in Rural Development Compound Annual Growth Rates (CAGR)4 of the expenditure are calculated for all major components of rural sector expenditure at constant 4

The expenditure values of the initial and final years are used to calculate the compound annual growth rate. The formula used to calculate it is: CAGR= ((Vf/Vi)1/f-i-1)*100; where Vf=Value in the final year; Vi=Value in the initial year; f= final year, i=initial year.

Participation of the Central Government in Rural Development

221

prices (1999–2000) for the period 1981–2 to 2008–9. The rate of growth analysis is useful as it gives immediate information about the period in simple numbers. To capture the decadal change in the expenditure pattern, the overall period is divided into three sub-periods: 1981–2 to 1989–90, 1990–1 to 1999–2000, and 2000–1 to 2008–9. Table 12.1 below shows the rate of growth of the central government expenditure over the period 1981–2 to 2008–9. Table 12.1. Compound annual growth rate of rural development expenditure of the central government for the 1980s, 1990s and 2000s (%)

RURAL DEVELOPMENT

GROWTH RATE AT CONSTANT PRICES (1999– 2000) 1990– 1 TO 1999– 2000 -1.21

2001–2

1981–2

2000) 1990–1

TO

TO

TO

2008–9

1989– 90 21.51

1999– 2000 -2.85

2001– 2 TO 2008– 9 11.43

12.85

-9.53 5.26 -5.77 56.46

-6.99 2.67

44.88 18.89

-8.53 0.97

42.65 17.39

1.54

-11.14 3.39 -7.45 53.68

2.09

0.40

0.26

25.49

-1.34

31.93

23.26

-2.97

30.26

8.24

13.88 17.83

6.45

12.44 16.34

2.04

3.72

0.22

2.00

20.42

4.56 10.59

4.57 9.65 18.88 33.11

18.29

2.83 8.76

3.25 8.27 17.37 31.43

0.31

-0.91

-1.35

-2.16

6.07 5.71

-38.22 7.56

4.31 3.96

-39.00 6.20

ACTIVITIES

LAND REFORMS AGRICULTURAL MARKETING IRDP AND OTHERS NATIONAL GRID SPECIAL DEVELOPMENT PROGRAMMES RURAL EMPLOYMENT DRINKING WATER AND SANITATION HOUSING WASTELAND DEVELOPMENT ROAD AND BRIDGES TOTAL RURAL DEVT. RURAL HEALTH AGRICULTURE AND ALLIED SERVICES IRRIGATION AND FLOOD CONTROL

GDP

PER CAPITA GROWTH RATE AT CONSTANT PRICES (1999–

1981– 82 TO 1989– 90 23.71

5.35

3.48

** Some data are missing in the above table due to the unavailability of the expenditure amount, either for the initial or the final year under the corresponding head. Source: Constructed from data obtained from the union budget (Government of India).

222

Chapter Twelve

Table12.1 above highlights some important facts. For almost all heads, the rates of growth of expenditure during the 1980s were more than that of the 1990s. The decrease in the growth rates during the 1990s were mainly because of the fiscal crisis that led to a fiscal adjustment policy from 1992 onwards. The rate of growth increased during the 2000s but could not reach the level of the 1980s. The total rural development expenditure grew at a rate of 20.42% in the 1980s. This large increase can be attributed to the low base-year value, i.e., low spending in 1981–2. The rate decreased to 4.56% in the subsequent decade. Again, during the 2000s the rate of growth of the rural development expenditure increased to 18.88%, which was lower than that of the 1980s. The negative rate of growth of some of the components is one of the major concerns. A keen observation of the rate of growth reveals that expenditure on land reforms, agricultural marketing and rural employment grew at negative rates in the 1990s. The rate of decline of growth was even more when we look at the real per capita expenditure. In the 1990s, the real per capita rate of growth is negative even for agriculture and allied services, which is disturbing. Employment and agriculture are very important for the rural sector, affecting most of the rural population. Thus, these components should not be ignored by any government. A negative rate of growth is also observed for agricultural marketing and the national grid programme during the 1980s. This is because the subject was gradually being transferred to the state governments. In the case of rural health expenditure, the rate of growth during the 2000s is high when compared to the 1990s. Although an increasing rate of growth was recorded, it did not increase throughout the decade but only from 2005 when the National Rural Health Mission (NRHM) was implemented. One of the major shifts during the 2000s was observed in the expenditure on Irrigation and Flood Control. The real per capita rate of growth of Irrigation and Flood Control turned out to be -39% during the 2000s. The expenditure was not cut gradually but came down suddenly from `25,007 crore in 2005–6 to a mere `462 crore in 2006–7, and has only decreased further since. The resources were diverted to the Bharat Nirman programme, of which irrigation is an important component. Phase I of the Bharat Nirman programme was implemented in the period 2005–6 to 2008–9, promising to create 10 million hectares of additional irrigation capacity by 2009, of which 6.5 million hectares were brought under assured irrigation until 2009.

Participation of the Central Government in Rural Development

223

The pattern of expenditure in terms of aggregate does not show the trends in the expenditure of the main component within it. An attempt is made in the following sections to analyse the expenditure under rural infrastructure, rural employment and the rural social sector separately.

IV. Rural Employment Expenditure Rural employment programmes are important as they help to directly reduce poverty within the country. In agriculture, female labourers get lower wages than male labourers. Similarly, there is also discrimination on the basis of caste. In this context, public works programmes are important as they do not discriminate on such bases. Public works programmes have also been long recognized as effective policy instruments to ensure food security in rural areas. The rural labour market is unlikely to ensure the best outcome in terms of wages and income if it is left to the market alone, and hence the government must intervene to improve its conditions. Fig. 12.3 below shows the trend in the share of expenditure on rural employment programmes. Fig. 12.3. Share of rural employment expenditure in GDP (%)

Rural Employment Percent of GDP

0.80 0.60 0.40 0.20 1981-82 1983-84 1985-86 1987-88 1989-90 1991-92 1993-94 1995-96 1997-98 1999-00 2001-02 2003-04 2005-06 2007-08

0.00

Rural Employment Source: Constructed from data obtained from the union budget (Government of India).

Chapter Twelve

224

It is observed that the share of rural employment expenditure in GDP fluctuated greatly over the period. While the initial years experienced a pattern of increasing consistently, the share of expenditure in the final years changed a lot. The share of expenditure increased from 0.10% in 1981–2 to 0.42% in 1987–8. A trend of continuous decrease in the share of rural employment expenditure has been observed during 1994–5 to 2000–1. The difference in the share of expenditure between the two decades is reflected in the decadal rate of growth, as shown in Table 12.1 above. The rate of growth of rural employment expenditure of the central government reduced from 25.49% in the 1980s to -1.34% in the 1990s. This had serious implications, as public rural works programmes play a major role in the alleviation of poverty and the provision of the rural infrastructure. After 2000, the share increased with some fluctuations but consistently remained below the mark it attained in 1989–90. This has however changed in the last couple of years with a sharp increase in the share due to MNREGA. Other programmes providing productive employment opportunities to the rural poor were the Integrated Rural Development Programme (IRDP) and the Swarnjayanti Gram Swarozgar Yojana (SGSY). These programmes aimed to provide income-generating assets to the rural poor through the provision of cheap bank credit. Fig. 12.4 below shows the trend in the share of IRDP expenditure as a percentage of GDP. Fig.12.4. Share of IRDP and SGSY expenditure in GDP (%)

0.15 0.10 0.05 0.00 1981-82 1983-84 1985-86 1987-88 1989-90 1991-92 1993-94 1995-96 1997-98 1999-00 2001-02 2003-04 2005-06 2007-08

Percent of GDP

IRDP and SGSY

IRDP and others Source: Constructed from data obtained from the union budget (Government of India).

Participation of the Central Government in Rural Development

225

Unlike other components, the share of expenditure of IRDP and SGSY continuously decreased during 1981–2 to 1999–2000, but from 2000– 1onwards there was a slight increase. This increase was mainly because of the start of SGSY, a major on-going programme for the self-employment of the rural poor. The programme started from April 1, 1999 after the reviewing and restructuring of the erstwhile IRDP and allied programmes, namely Training of Rural Youth for Self Employment (TRYSEM), Development of Women and Children in Rural Areas (DWCRA) and the Million Wells Scheme (MWS). The earlier programmes are no longer in operation following the launch of the SGSY.

V. Rural Investment and Infrastructure The role of the state in building agricultural infrastructure (e.g. technology evolution and transfer, irrigation, electricity, inputs supply, marketing) and general infrastructure (e.g. roads and communication) is important for revitalizing the agricultural and rural sectors. Poverty eradication in India’s backward regions is not possible without a massive increase in public investment in all forms of rural infrastructure. Empirical studies have highlighted the positive impact of infrastructure on performance in terms of growth. Rural infrastructure in particular showed an improvement in agricultural productivity and the reduction of rural poverty (Binswanger et al. 1989; Fan, Hazell & Thorat 2000). A strong positive correlation between rural poverty and deficiency of infrastructure is a well-established phenomenon (NCAER 2006). The rural sector, with its low purchasing power, is not a profitable market for investors. It is difficult to impose user charge in return for the use of rural infrastructure facilities because of its non-excludable and nonrival nature. Unlike inter-city highways, a toll cannot be imposed on village roads. The sum of allocations to agriculture marketing, the national grid, special development programmes, drinking water and sanitation, rural housing, roads and bridges, irrigation and flood control is taken to examine the trend in rural infrastructure. Fig. 12.5 below reflects the trend in the share of rural infrastructure expenditure as a share of GDP. After exhibiting a downward trend in the initial years, the share started increasing from 1995–6. The decline during the initial years was mainly because growth in spending was not matched with growth in GDP. The emphasis of the government seemed to be more on industrial development rather than the development of rural infrastructure. In the initial Five Year Plans, it was assumed that agriculture was subject to

Chapter Twelve

226

diminishing returns whereas industries operated at increasing returns. So, industrialization would consume surplus labourers who were otherwise under-employed in agriculture to be more productively employed in industries. It was believed that the “food problem” could be taken care of through changes in agrarian relations, the creation of “infrastructure” through community development programmes and the greater provision of public irrigation facilities. Fig. 12.5. Share of rural infrastructure expenditure in GDP (%)

1.5 1 0.5 0 1985-86 1987-88 1989-90 1991-92 1993-94 1995-96 1997-98 1999-00 2001-02 2003-04 2005-06 2007-08

Percent of GDP

Rural infrastructure

Rural infrastructure Source: Constructed from data obtained from the union budget (Government of India).

The increase in the following years was mainly due to spending on rural housing, roads and bridges. The rural housing scheme started in 1992–3 with the allocation of `5 crore for the year, which was increased substantially in subsequent years. During 1996–7, `1194 crore was allocated for rural housing. Starting from 2000–1, large allocations for roads and bridges were also made with the initiation of the Pradhan Mantri Gram Sadak Yajna (PMGSY). The steep decrease in 2006–7 was due to the decrease in allocation for Irrigation and Flood control, as the resources were diverted to the Bharat Nirman programme. Rural labourers have little access to healthcare, safe drinking water, sanitation and housing. The provision of drinking water and sanitation is an important part of rural infrastructure, which is necessary to create capacity in agriculture. Its share of expenditure remained at around 0.1% of the GDP since it started in 1985–6 until 2004–5, after which it increased sharply. The CAGR for drinking water and sanitation was

Participation of the Central Government in Rural Development

227

8.24% in the 1990s and it increased to 13.88% in the 2000s, indicating that the centre increased its provision of these facilities in rural areas. The trend in the share of expenditure on agriculture and allied services as a percentage of GDP is shown in Fig. 12.6 below. Fig. 12.6 shows a trend of decreasing order for the period. The share continuously decreased from 0.85% in 1987–8 to 0.2% in 2008–9. It is found that the CAGR was 0.31% in the 1990s and -0.91% in the 2000s, indicating that the agricultural sector expenditure was not taken seriously by the central government in that period. This observation is consistent with the findings of Gross Capital Formation in Agriculture (GCFA), which explains that the share of public investment in agriculture continuously fell over the period. A study by Gulati & Bathla (2001) concluded that during the period under consideration, the household sector was mainly responsible for any growth in the GCFA. Since poor farmers do not have the money to invest in infrastructure, like personal irrigation facilities, agriculture and machinery, most of the investment must have been carried out by rich farmers. This had serious implications on the inequality in accessing basic infrastructure in agriculture. Public investment in agriculture is one of the basic factors for increasing production. The government must ensure that the rural poor have access to agricultural infrastructure, which can only be provided by greater public participation in agricultural capital formation. The formation of public capital is also necessary because it induces private investment. It is high time we move beyond the Mahalanobian concept of treating agriculture as a bargain sector. Fig. 12.6. Share of agriculture and allied services expenditure in GDP (%)

Agriculture and Allied Services 1.00 0.00

Agriculture and Allied Services Source: Constructed from data obtained in union budget (Government of India).

228

Chapter Twelve

VI. Trend in Expenditure on Rural Social Services Even after sixty years of independence, rural Indians have no guarantee of state-sponsored education or health. The basic needs of health, education, food and social security may only be addressed through government intervention. Most employment in the rural areas takes place in the unorganized sector, which is characterized by low earnings, poor working conditions and a lack of social security. A study carried out by Sen (1998) found that there was a remarkable increase of the proportion of labourers working as casual labourers in the rural workforce, indicating greater insecurity of contracts as well as uncertainty of finding employment. This provides a rationale for introducing social security measures by the government. An increase in security and income of the rural labourers through government programmes may reduce malnutrition in rural households. Old people and widows are more vulnerable and face greater financial insecurity. It is the primary responsibility of the government to provide financial security to these disadvantaged groups. Various studies have shown that the level of public expenditure on the social sector is significantly associated with improvements in human development (Chakraborty 2003; Anand & Ravallion 1993). The public services requirement is even greater in the rural sector because of the lack of private players and the low purchasing capacity of rural people. Further, the emphasis on reducing budget deficits may result in the relative reduction of expenditure on the social sector. In the next section, the central government’s interventions in rural social services are analysed. The sum of the allocations to land reform, social security and welfare, wasteland development and rural health is taken together as rural social services expenditure. The allocation for rural social services has shown an increasing trend from 1981–2 with slight fluctuations, except in 2005–6 when a sudden increase in the expenditure is observed. This was because of the rural health component which registered high growth after remaining steady at around 0.05% for a long period (1981–2 to 2004–5). This high increase in health allocation can be attributed to the National Rural Health Mission (NRHM) launched on April 12, 2005, with the objective to provide accessible, affordable and accountable quality health services to even the poorest households in the remotest rural regions.

Participation of the Central Government in Rural Development

229

Fig. 12.7.Trend in the share of rural social services expenditure in GDP (%)

2007-08

2005-06

2003-04

2001-02

1999-00

1997-98

1995-96

1993-94

1991-92

1989-90

1987-88

1985-86

1983-84

0.3 0.25 0.2 0.15 0.1 0.05 0 1981-82

Percent of GDP

Rural Social services

Rural Social services Source: Constructed from data obtained from the union budget (Government of India).

VII. Conclusion The study reveals that the expenditure of the central government on the rural sector underwent considerable change during the period 1981–2 to 2008–9. During the initial period, the rural sector was ignored, and the share of expenditure allocated to uplift the rural sector was very small. However, in recent years emphasis has been laid on the provision of rural employment, infrastructure and health facilities with the initiation of the MNREGA, Bharat Nirman and NRHM programmes. These allocations are made keeping in mind the requirements of the Indian rural sector. Looking at the CAGR, it was found that for almost all heads the rate of growth of expenditure in the 1980s was more than that observed in the 1990s. The decrease in the rate of growth in the 1990s was mainly because of the fiscal crisis, which led to a fiscal adjustment policy from 1992 onwards. The rate increased in the 2000s but was unable to reach the level of the1980s.The negative rate of growth of some of the components is one of the major concerns. Slow growth in allocation to employment and agriculture is another concern as it is very important for the rural sector. These components should not be ignored by any rational government as they affect most of the rural population. Policies have to be framed for

230

Chapter Twelve

both unskilled and skilled workers. Direct employment programmes such as wage and self-employment schemes have to be effectively implemented. This paper also argues that the agricultural sector was not given due attention by the central government. In the initial years, the government prioritized industrial development more than the development of rural infrastructure. The lesser public participation in the rural sector resulted in a low planned fund transfer to the agricultural sector. This may have adversely affected the economy in general and the rural masses in particular. It also points to the fact that the Indian rural agriculture sector is currently in dire need of greater government attention.

References Anand, S. & M. Ravallion. 1993. Human Development in Poor Countries: On the Role of Private Incomes and Public Services. Journal of Economic Perspectives. 71: 133–50. Binswanger, H. S. R. Khandker. & M. Rosenzweig. 1989. How Infrastructure and Financial Institutions Effect Agriculture Output and Investment an India. Policy Planning and Research Working Paper No. 163. Washington D.C. : World Bank. Chakraborty, L. 2003. Public Expenditure and Human Development: An Empirical Investigation. Paper presented at the WIDER World Institute for Development Economics Research Conference on Inequality, Poverty and Human Wellbeing. Helsinki. Finland. May 30–31. Chakravarty, S. 1986. Development Planning: The Indian Experience. New Delhi: Oxford University Press. Chandrasekhar, C. P. & J. Ghosh. 2002. The Market that Failed: A Decade of Neo-liberal Reforms in India. New Delhi: Left Word Books. Dev, S. Mahendra. 1998. Government Interventions and Social Security for Rural Labour. In Empowering Rural Labour in India: Market, State and Mobilization. edited by R. Radhakrishna & Alakh N. Sharma. New Delhi: Institute of Human Development. Devarajan, S. V. Swaroop. & H. Zou. 1996. The Composition of Public Expenditure and Economic Growth. Journal of Monetary Economics. 372: 313–44. Evans, P. & G. Karras. 1994. Are Government Activities Productive? Evidence from a Panel of U.S. States. The Review of Economics and Statistics. 761: 1–11. Fan, S. P. Hazell & S. Thorat. 2000. Government Spending, Growth and Poverty in Rural India. American Journal of Agricultural Economics. 824: 1038–51.

Participation of the Central Government in Rural Development

231

Government of India. All issues from 1981–2 to 2009–10. Annual Financial Statement, Budget Documents. New Delhi: Ministry of Finance. Gulati, A. & S. Bathla. 2001. Capital Formation in Indian Agriculture: Revisiting the Debate. Economic and Political Weekly. 3620: 1697– 1708. Mooij, J. & S. M. Dev. 2004. Social Sector Priorities: An Analysis of the Budgets and Expenditures in India in the 1990s. Development Policy Review. 221: 97–120. NCAER. 2006. India Rural Infrastructure Report. New Delhi: National Council for Applied Economic Research. Patnaik, U. 2007. New Data on the Arrested Development of Capitalism in Indian Agriculture. Social Scientist. 357–8:4–23. Radhakrishna, R. & A. N. Sharma. 1998. Introduction: Towards Empowering Rural Labour. In Empowering Rural Labour in India: Market, State and Mobilization. edited by R. Radhakrishna & Alakh N. Sharma. New Delhi: Institute of Human Development. Rajaraman, I. 2003. Interstate Variations in Utilisation of the Rural Infrastructure Development Fund. Working Paper No E/235/2003. New Delhi: Institute of Economic Growth. Ramakumar, R. 2008. Levels and Composition of Public Social and Economic Expenditures in India: 1950–1 to 2005–6. Social Scientist. 369–10: 48–94. Sen, A. 1998. Rural Labour Markets and Poverty. In Empowering Rural Labour in India: Market, State and Mobilization. edited by R. Radhakrishna & Alakh N. Sharma. New Delhi: Institute of Human Development.

Land reforms

6.64 3.95 7.35 22.17 11.85 10.79 9.72 28.17 36.43 47.99 26.84 40.51 54.85 39.15

Year

1981–2 1982–3 1983–4 1984–5 1985–6 1986–7 1987–8 1988–9 1989–90 1990–1 1991–2 1992–3 1993–4 1994–55

38.85 36.15 19.60 29.69 24.25 27.57 16.92 16.42 17.43 13.34 16.65 8.51 17.29 5.53

Agricultural marketing

574.65 708.62 789.53 769.58 707.87 839.81 812.96 874.78 866.09 749.51 657.85 637.87 952.94 879.13

IRDP and others

13.11 10.07 11.14 17.18 19.22 11.99 9.58 8.84 8.15 7.60 9.25 3.42

National grid

Social security welfare and

4.37 3.02 3.71 5.15 65.90 120.63 124.85 132.27 156.91 142.81 103.33 102.97 160.80 191.46

Special devt. Programme s

794.85 732.17 1082.14 2198.30 3022.63 3450.10 3653.94 3179.61 4888.28 4209.32 3375.82 4326.13 6034.51 6579.06

Rural employment

952.62 962.63 1067.11 1086.46 1000.93 886.22 1187.53 815.61 1189.60 1224.34

Drinking water and sanitation

An MPLAD Fund of `790 crore is also allocated to the Ministry of Rural Development for this year only. The amount is subtracted to make it comparable.

5

Chapter Twelve

Annexure 1 Allocation of planned funds (revised estimates) under different heads for 1981–2 to 2008–9 at constant prices (1999–2000) (crore)

232

50.94 47.26 44.33 35.83 43.00 91.96 141.05 126.81 131.36 155.23 205.63 204.35 192.92 241.93

4.27 7.74 8.14 7.53 6.95 6.46 28.18 61.59 76.19 94.72 101.22 127.76 148.72 122.62

825.13 705.73 579.06 676.05 950.00 358.33 451.37 594.18 630.53 747.09 717.18 820.45 1173.40 1430.33

Source: Budget documents, Government of India, various years.

1995–6 1996–7 1997–8 1998–9 1999–2000 2000–1 2001–2 2002–3 2003–4 2004–5 2005–6 2006–7 2007–8 2008–9

709.26 658.05 549.88 664.68 710.00 710.36 622.04

188.79 217.63 325.72 231.86 172.00 231.56 239.04 243.20 274.10 287.01 264.02 262.05 234.68 261.74

Participation of the Central Government in Rural Development

6424.62 4645.82 4747.97 4595.66 3729.00 2710.14 3973.00 8606.48 8442.05 5319.25 9325.04 9777.07 10299.86 24876.97

1508.79 1381.90 1573.34 1743.75 1807.00 2033.76 1984.15 2037.95 2408.26 2739.32 3793.78 4026.30 5403.44 5753.83

233

Chapter Twelve

Year 1981–2 1982–3 1983–4 1984–5 1985–6 1986–7 1987–8 1988–9 1989–90 1990–1 1991–2 1992–3 1993–4 1994–5 1995–6 1996–7 1997–8

8.50 15.45 31.66 634.47 1428.56 1283.53

Rural housing

Wasteland development 196.65 192.48 159.26 161.83 169.78 194.81 191.74 255.85 231.05 202.31 186.83 172.06 312.08 323.68 367.53 269.20 265.68

Road and bridges

Total rural devt. 1629.12 1686.47 2072.73 3203.90 4974.12 5618.34 5886.82 5582.40 7205.27 6259.10 5564.10 6115.58 8737.50 9274.02 10713.80 9361.90 9377.64 Agriculture and allied services

5849.53 6641.55 7508.26 7334.86 7227.67 7163.75 7123.45 7163.77 6586.03 7528.98 6553.58 7159.56 6653.58

Rural health 144.29 134.46 140.14 146.94 394.00 455.48 490.04 486.50 434.47 386.62 512.90 511.91 445.41 451.30 418.84 515.98

8944.09 9656.62 9161.55 9073.19 8468.36 8360.87 7828.21 7994.68 8296.27 8590.07 9342.91 9540.50 11115.48

Irrigation and flood control

Annexure 2 Allocation of planned funds (revised estimates) under different heads for 1981–2 to 2008–9 at constant Prices (1999–2000) (` crore)

234

238.56

281.00

682.81 658.25 733.66 700.59 716.37 907.00 872.87 821.13 976.53

1591.09

1659.00

1443.00 1707.03 1406.18 1497.50 2164.06 1972.61 1994.20 2633.63 5360.54

3136.40 2981.39 2895.70 2729.67 2993.03 4724.61 5587.91 6301.60 6556.08

11404.79 12785.50 16705.74 16890.25 15216.08 22011.09 23672.98 27209.39 45580.57

9357.94

9785.01

Source: Budget documents, Government of India, various years.

1998–9 1999– 2000 2000–1 2001–2 2002–3 2003–4 2004–5 2005–6 2006–7 2007–8 2008–9 829.00 894.71 1505.50 1367.40 1429.51 5384.55 6085.91 7805.66 8170.44

1075.00

628.33

7338.01 7756.04 6933.55 7685.43 9100.34 10711.84 5615.55 6188.61 6819.98

7365.00

7994.89

Participation of the Central Government in Rural Development

13102.28 13684.04 10837.35 11296.95 15792.58 19930.88 350.97 328.84 278.21

14209.99

11231.07

235

CHAPTER THIRTEEN THE DETERMINANTS OF CHILD LABOUR IN INDIA: A PROBIT ANALYSIS CHANDAN KUMAR MOHANTY

I. Introduction The International Labour Organisation (ILO) published a report in 2002 stating that there were 211 million “economically active” children aged 5–14 around the world. In the same year, an additional 141 million children aged 15–17 were also working. It was estimated that 306 million children aged 5–17 were employed in 2008. From 2004 to 2008, the number of people employed aged 5–14 declined from 196 million to 176 million, a decrease of 20 million. This may be due to the growing concern about child labour along with the adoption of stringent policies to ban it. Over the same period, the number of employed children aged 15–17 rose by two million from 127 million to 129 million. The data reveals that boys continue to be more exposed to work than girls, with a 4.5% higher incidence rate. The estimates given by the ILO or any other international organization on child labour are subject to differences depending on the definition of work and a child, and the way data is collected (Basu, 1998). These statistics raise a series of key questions—why do children work? What forces them to work? Why does society permit it? Are there any economic consequences in terms of losing human capital because of child labour? When we move to the Indian context, a bleak picture of child labour emerges, which draws our attention to what the policymakers should do to make economic development inclusive. As per the population census of India, the number of economically active children aged 5–14 increased from 11.3 million in 1991 to 12.7 million in 2001. The statistics provided above indicate that there is a large pool of human resources which remains unutilized in the growth process of the economy. The endogenous growth model emphasizes that human resource

The Determinants of Child Labour in India: A Probit Analysis

237

is a key factor in economic development. Following this growth model, it has been well documented in the literature that investment in children’s education will assure economic development. Thus, every child should attend school and get formal education. But in India, like in many other developing countries, a number of children work instead of attending school. The primary reason for children not attending school is the dismal poverty in their families. However, there are other socio-economic factors which accentuate child labour. The goal of this paper is two-fold. First, the incidence of child labour in India is calculated with respect to gender, state and sector. Second, a regression model is fitted in order to study the causes of child labour in India by using the NSS unit level data (66th round). We find that most of the individual variables in our regression model, like age and gender of the child, poverty level of the household and the education of the head of the household explain the participation probability of children in the labour market quite well. This paper also finds that India has experienced a decrease in the incidence of child labour for both boys and girls, and that the decrease is sharper in the case of girls irrespective of the sectors (rural or urban). The urban India performs better in terms of decrease in the incidence of child labour. A majority of states are successful in reducing the incidence of child labour, whereas states like Arunachal Pradesh, Chandigarh, Assam, Gujarat, Pondicherry and Delhi fail in this regard. Although there is a huge body of literature on child labour, one of the contributions of the paper is the detailed analysis of household income to determine the working status of a child. This study is mostly related to the paper by Mukkherjee & Das (2008) in which they ask a specific question on the role of parental education in the schooling of children and the decision to send them to work. The rest of the paper is organized as follows. Section II presents a brief review of the literature on child labour and its determinants. Section III describes the sources of data and methodology used for empirical analysis. Section IV highlights the incidence of child labour in India calculated from the secondary data. Section V presents the econometric model and the methods. Section VI reports the results, and section VII concludes the paper.

238

Chapter Thirteen

II. Literature Review II.1. Causal factors of child labour The problem of child labour is multidimensional originating from a complex combination of determinants which can be broadly classified into two categories: (a) those related to the decision-making of the household in sending the child to work (child labour supply); and (b) factors from the demand point of view, such as labour market structure, technology requirements and globalization (child labour demand). The following subsections proceed with the discussion on child labour with reference to the key factors that cause child labour. II.1.1.

Supply-side factors of child labour

(a) Household income The most evident determinant of child labour is the level of income of the household. Literature confirms that there exists a strong negative correlation between child labour and the household’s total income. Cigno, Rosati & Tzannatos (2002) note that there seems to be an apparent “income paradox,” by which income redistribution to households has less impact on child labour reduction. This implies that the structural factors, such as the global income inequality of a country, do not allow for a substantial modification of the position of the poor. This phenomenon subsequently minimizes the possible effects of income redistribution. Therefore, the institutional structure perpetuates the social inequalities causing child labour. (b) Household structure The composition of the household is a decisive factor in making the decision of sending the child to school or forcing them to work. These decisions are a combination of substitution and income effects. Cigno, Rosati & Tzannatos (2002), in their work on India, Morocco and Vietnam, observe that to correctly identify these two factors, it is necessary to classify household members into three groups: adults, children of schoolgoing age, and pre-school children. The expected relation is that the greater the number of adults working, the lower the probability of a working child (substitution effect). But at the same time, if the number of pre-school children is large, the probability of a child of the age of schoolgoing working goes up. The reason behind this is that pre-school children are very delicate and need a lot of care. This implies that an increase in their number is essentially equivalent to not only a reduction of total

The Determinants of Child Labour in India: A Probit Analysis

239

income (income effect) but also to an increase in domestic work. These two factors may force a child of school-going age to work part or fulltime. (c) Socio-economic profile of parents The empirical results show that parental education and employment conditions are two of the most important determinants in sending a child to work or to study (Psacharopuolos 1997; Gustafsson-Wright & Pyne 2002). In general, children of better-educated parents are more likely to attend school and are less likely to work part or full time than children of less-educated parents. It can be noted that differences between mother and father in terms of the level to which they are educated also have a vital impact (Cigno, Rosati & Tzannatos 2002). In some cases, the mother’s education tends to affect the welfare of the children more positively. Basu (1993) provides evidence based on the study of suburbs of New Delhi that mothers, who are bound to participate in the labour market, depend on their children to substitute them at home. (d) Child labour and education The literature on child labour and education predicts that the decision to send a child to a primary school and the quality of education of that school are positively related (Lavy 1996; Cigno, Rosati & Tzannatos 2000). With the quality of school being high, the same level of education is reached in less time (Cigno & Rosati 2000). Evidence on the link between school quality and child labour is provided by Schultz (1997) and Dreze & Kingdom (2000). The relationship between education and child labour is mostly based on the theory on human capital, which states that working and attending school are not substitutes for but complementary to each other (Basu & Van 1998; Anker 2000). Some of the studies based on the relationship between the quality of education and child labour are by Bonnet (1993), Schultz (1997) and Dreze & Kingdom (2000). The quality of education determines the perception that parents have of the process of education as a way to improve quality of life. There is also a large body of literature establishing a negative correlation between child labour and attending school (Patrinos & Psacharopoulos 1995; 1997; Boozer and Suri 2001). (e) Rural life The share of the population working in the agricultural sector may be positively associated with child labour because the variable is generally considered to be an indicator of poor human capital. Moreover, the characteristics of rural life play a catalytic role in creating child labour.

240

Chapter Thirteen

Children belonging to households which farm a large plot of land are more likely to work because their services are more valuable. Cigno, Rosati & Tzannatos (2000) and Rosenzweig & Everson (1977) found a positive relationship between farm size and child work. II.1.2.

Demand-side factors of child labour

(a) Labour market structure and the demand of child labour It can be clearly stated that the categorization of the labour market into formal and informal employment is a key determinant of child labour. This directly impacts the level of salary and the type of work that a child can access or indirectly impacts the working conditions of a household member. In developing countries, the principal characteristic of the labour markets is a high proportion of activity in the informal sector and the existence of an unskilled labour force. This may lead to a large market space where children can be made to participate as a labour force. Likewise, company strategies such as avoiding social legislations, cost reductions and sub-contracting can create pressure on the demand for child labour. Beegle, Dehejia & Gatti (2009) find that child labour is associated with a significant increase in wages years later; moreover, the increase in wages is sufficiently large to offset the cost of displacing education. Controlling for completed schooling, Emerson & Souza (2011) empirically establish that child labour has a negative effect on adult earnings. (b) Child labour and technological change Technological change influences both demand for and supply of child labour. If new technologies like mechanization or use of fertilizers and improved strains are introduced to agriculture, the demand for unskilled labour in general and child labour in particular is likely to fall. At the same time, the return of children to school is likely to increase as a reflection of a higher demand for skilled labour. This will automatically reduce the supply of child labour. The combined impact of reduced willingness to supply child labour and employ unskilled workers is likely to lead to a decrease in the use of children for productive activities (Cigno, Rosati & Tzannatos 2002).

III. The Data and Empirical Strategy This paper uses the data set of the National Sample Survey (NSS) published by the Ministry of Statistics and Programme Implementation

The Determinants of Child Labour in India: A Probit Analysis

241

(MOSPI), Government of India. Three rounds of the Employment and Unemployment Schedule of NSS such as the 50th (1993–4), 61st (2004– 5), and 66th rounds (2009–10) are used and the analysis is based on the unit level study. Children aged 5–14, who are in the labour force according to usual principal and subsidiary status, are considered child labour. However, no distinction is made between child employment and child labour. For 2009–10, the sample sizes of children for rural and urban India are 56,286 and 31,435 respectively. A cross tabulation of household characteristics and spatial distribution of child labour over the period from 1993–4 to 2009–10 is presented. These statistics shed light on some key issues which may be of importance for the policies for fighting child labour. We have also calculated the Labour Force Participation Rate (LFPR) and the Student Population Ratio (SPR). The LFPR is defined as the ratio between the populations of children aged 5–14 in the labour force to the total population of children in the same age group. The SPR is defined as the ratio between children aged 5–14 attending school to the total population of children in the same age group. In the analysis of this paper, household income is divided into six categories following Sengupta (2008), where we try to see the impact of different income levels on the probability of a child participating in the labour market. In order to show the impact of various micro factors at the household level on the probability of a child’s participation in the labour force, a Probit regression model is fitted. The socio-economic and demographic variables used in the regression model are constructed from the unit level analysis of the data. The definitions of independent and dependent variables are presented in section V.

IV. Incidence of Child Labour in India IV.1. National estimates India has been successful in dealing with child labour, as its incidence has declined from 6.3% in 1993–4 to 2% in 2009–10, along with a sharp rise in the student population ratio from 68.5% to 87.1% during the same period. The data indicate the annual decrease in the incidence of child labour to be very sharp (4.2%), and that the absolute number of child labourers in our country is 4.5 million in 2009–10, as per the NSS estimates. The sector-wise study of the child labour force enables us to make more distinct observations on its composition. The Labour Force Participation Rates (LFPR) of male and female children in 1993–4 in rural India were not different. Fig.13.1 below shows that the male child’s LFPR

Chapter Thirteen

242

in rural India has declined sharply and that there is also a steady rise in the Student Population Ratio (SPR). However, the decline in LFPR and the rise in SPR of female children in rural India are sharper than those of the male children. The condition of girls in rural India has improved impressively in terms of two things: a decrease in the incidence of child labour for girls and a rise in SPR for girls during 1993–4 to 2009–10 (see Fig. 13.2 below). Likewise, there is a sharp decline in the incidence of child labour among female children in urban areas and a sharp increase in the student population ratio as compared to urban male children engaged in child labour during 1993–4 to 2009–10 (see Fig. 13.3 and 13.4 below). Fig. 13.1. Male child LFPR and SPR in rural India (2009–10) Male Child LFPR and SPR in Rural India, 2009-10 8 7 6 5 LFPR4

100 80 60

SPR

40

3 2 1 0

20

SPR

0 1993-94

2004-05

LFPR

2009-10

Year

Fig. 13.2. Female child LFPR and SPR in rural India (2009–10) Female Child LFPR and SPR in Rural India, 2009-10 90 80 70 60 50 SPR 40 30 20 10 0

8 7 6 LFPR5 4 3 2 1 0 1993-94

2004-05 Year

2009-10

LFPR SPR

The Determinants of Child Labour in India: A Probit Analysis

243

Fig. 13.3. Male child LFPR and SPR in urban India (2009–10) Male Child LFPR and SPR in Urban India, 2009-10 4 3.5 3 2.5 LFPR 2 1.5 1 0.5 0

92 90 88

SPR

86 84 82 1993-94

2004-05

LFPR SPR

2009-10

Year

Fig. 13.3. Female child LFPR and SPR in urban India (2009–10) Female Child LFPR and SPR in Urban India, 2009-10 3

92 90 88 86 84 SPR 82 80 78 LFPR 76 SPR 74

2.5 2 LFPR 1.5 1 0.5 0 1993-94

2004-05

2009-10

Year

Source: NSS unit records of the 50th, 61st and 66th rounds.

IV.2. Spatial difference in the incidence of child labour and the student population ratio The incidence of child labour was the highest in Andhra Pradesh during 1993–4 with 15% of its child population in the labour force. The other states with high incidences of child labour in 1993–4 were Karnataka (11.3%), Rajasthan (12.1%) and Himachal Pradesh (13.5%). The student population ratios of these high child labour incidence states were also

244

Chapter Thirteen

quite low, standing at 66% in Andhra Pradesh in 1993–4. Until 2009–10, the incidence of child labour had declined drastically across all states except Chandigarh. The rate of decrease in the incidence of child labour is higher for most of the Union Territories of India, i.e. Lakshadweep, Andaman and Nicobar, and Daman and Diu. The interesting figure is that of the rate of decrease in the incidence of child labour in the underdeveloped states like Andhra Pradesh and Himachal Pradesh, which has been quite significant compared to that in developed states like Delhi, where the decrease in the incidence of child labour is only 1.3%. In fact, there are some states where the incidence of child labour has increased over the period from 1993–4 to 2009–10. These states are Chandigarh, Arunachal Pradesh, Assam, Gujarat, Dadra and Nagar Haveli. There has been a gradual improvement in the student population ratio in India during this period; for instance, Lakshadweep experienced 100% school attendance for children aged 5–15 in 2009–10. The states like Odisha, Madhya Pradesh, Andhra Pradesh, Rajasthan, Uttar Pradesh and Bihar have managed to increase their student population ratios by more than 2% per annum, moving ahead of the developed states. During this period, Delhi experienced a slight decline in its student population by 0.2% per annum (see Table 13.1 below).

IV.3. Household characteristics of children engaged in child labour The analysis of child labour rests on the socio-economic background of the children engaged in it. On the basis of the evidence from developing countries, Aggarwal (2004) suggests that larger households tend to have fewer school-going children and the relationship is influenced by family culture, level of social expenditure by the state, and the level of socioeconomic development. There is evidence that poor households tend to send children to work in order to supplement family income, which is uncertain due to natural calamities, failed harvests, prolonged illness and many other reasons. It is also predicted that, if the system of education remains undeveloped and irrelevant, many parents will not send their children to school and prefer child labour as an option.

The Determinants of Child Labour in India: A Probit Analysis

245

Table 13.1. Incidence of child labour and student population ratio (1993–4 to 2009–10) States

Incidence of Child Labour 2009– 1993–4 2004–5 10 4.7 2.8 1.5 13.5 2.7 0.7 2.5 2.2 1.1 1.0 0.6 2.2 2.7 1.4 2.5 1.8 1.5 0.8 0.4 0.6 12.1 4.9 2.6 4.7 4.0 3.5 3.3 1.5 1.2 0.7 1.3 0.6 1.9 2.0 2.4 1.5 3.2 0.9 0.7 0.6 0.2 2.6 1.0 0.1 1.6 1.9 0.1 2.8 4.6 1.8 2.8 1.9 3.0 4.9 3.7 3.4 2.5 1.1 7.0 5.0 1.7 4.6 0.2 6.7 2.9 1.1 3.5 2.5 3.6 0.5 0.5 0.0

Student Population Ratio 2009– 1993–4 2004–5 10 81.2 88.1 93.5 87.0 95.0 96.5 80.2 89.0 91.8 88.3 93.9 89.0 87.6 90.8 77.2 87.2 91.9 88.5 90.3 86.3 58.5 78.0 82.0 60.6 77.5 84.7 53.3 65.2 74.6 89.7 94.1 98.1 64.0 72.1 71.5 90.5 93.3 97.6 90.2 93.2 92.1 86.6 95.3 95.3 85.6 88.2 92.0 75.5 86.9 91.9 75.6 87.1 87.6 67.9 82.9 88.5 76.7 77.0 64.0 80.2 92.1 81.0 88.9 61.2 78.4 86.1 74.7 85.6 81.2 94.7 98.0 92.5

J& K Himachal Pradesh Punjab Chandigarh Uttaranchal Haryana Delhi Rajasthan Uttar Pradesh Bihar Sikkim Arunachal Pradesh Nagaland Manipur Mizoram Tripura Meghalaya Assam West Bengal Jharkhand Orissa Chhattisgarh Madhya Pradesh Gujarat Daman & Diu Dadra & Nagar 1.6 1.8 2.3 61.9 84.2 68.3 Haveli Maharashtra 5.1 3.6 1.4 82.4 89.1 94.7 Andhra Pradesh 14.9 6.7 1.8 65.9 87.6 94.4 Karnataka 11.3 4.7 2.2 73.3 88.3 92.9 Goa 1.5 3.4 0.0 89.1 94.6 99.6 Lakshadweep 0.3 0.1 0.0 92.6 93.7 100.0 Kerala 0.9 0.3 0.1 93.4 97.6 97.6 Tamil Nadu 8.7 1.6 0.2 82.4 96.1 98.5 Puducherry 1.0 0.2 0.3 93.8 97.9 98.7 Andaman & 6.6 1.0 0.2 88.0 97.4 98.4 Nicober India 6.3 3.4 2.0 68.5 82.1 87.1 Source: NSS unit records of the 50th, 61st and 66th rounds, EUS. All figures are in percentages.

246

Chapter Thirteen

With respect to the type of household, the households which are selfemployed in agriculture send more children to the labour market as compared to others in rural India, and it is the households in urban areas that are engaged in casual labour from which the supply of child labour is the highest. The next highest category in this respect comprises households engaged in manual labour, those engaged in agricultural labour in rural India and self-employed households in urban India. The incidence of child labour is more pronounced among boys than girls across all types of households in both rural and urban India. With respect to the castebased distribution of child labour in both rural and urban India, the ST households supply a huge number of children to the labour market. The proportion of ST children in the labour force with respect to their total population stands at 3.4%, which is higher by 1.5% than that of nonST/SC/OBC in rural India. The pattern is the same in urban India. The ST households send more children to work because they are economically and socially backward classes and are abjectly poverty-ridden. In rural India, the maximum concentration of child labour is in the small households with 1 to 3 members (3.1%) engaged in labour, which is higher by 1.9% than that of large households with 11 and more members. If households are poor and small in size, they tend to supply more child labour so as to compensate for the fewer earning members. In both rural and urban India, the maximum proportion of child labour belongs to the Islamic religion (3.8% in rural and 2.1% in urban). The pattern is the same for both boys and girls in rural and urban India. The incidence of child labour among the extremely poor and poor households is high in both rural and urban India. However, the proportion of children in the labour force from marginal and vulnerable households is also significant, at least in rural India (see Table 13.2 below).

V. Econometric Models and Estimation Methods V.1. Specification of the model What forces a child to work? To answer this question, we consider estimating a Probit regression model. There are many factors that may influence the probability of a child being in the labour force. Following the voluminous literature on child labour, we have tried to incorporate some control variables in our regression model, but to be sure, our list of variables is likely to remain incomplete. The Probit regression model possesses the property of following the normal cumulative density function which is useful for many practical purposes. The estimation

The Determinants of Child Labour in India: A Probit Analysis

247

method applied to the Probit model is the Maximum Likelihood Estimation (MLE). All the regressions are run in STATA 11. The key advantage of the Probit model used in our study is that it automatically checks the identification in STATA. If the model is under-identified, it will drop some variables and observations for the successful estimation. In our model, the regressand is a categorical variable and the regressors are of mixed nature, i.e. a mixture of categorical and continuous variables mainly covering some socio-economic and demographic indicators. The Probit model for the rural is as follows. Table 13.2. Household characteristics of child workers in rural and urban India (2009–10) Household type Self-employed in NA (Selfemployed) Agl. Labour (Regular) Manual Labour (Casual labour)

Rural India Boys Girls

Total

Urban India Boys Girls

Total

2.4

1.4

1.9

1.8

0.8

1.3

2.5 2.2 3.2 1.3 2.6

2.1 1.8 2.8 0.3 2.0

2.3 2.1 3.0 0.8 2.3

1.1 2.5

0.3 1.4

0.7 2.0

0.4 1.6

0.2 0.7

0.3 1.2

3.4 2.2 1.8 1.1 1.6

1.7 1.1 0.7 0.4 0.7

2.6 1.7 1.3 0.8 1.2

1.3 3.2 1.6 1.1 1.6

0.6 1.0 0.5 0.4 0.7

1.0 2.1 1.1 0.8 1.2

2.3 1.5 2.6 0.4 1.6

0.8 0.6 1.2 0.2 0.7

1.7 1.1 1.9 0.3 1.2

3.7 3.1 2.6

1.5 0.9 0.5

2.6 2.1 1.6

Self-employed in Agl Others (Others) Total Social group ST 3.4 3.4 3.4 SC 2.5 1.8 2.1 OBC 2.4 2.2 2.3 Others 2.4 1.3 1.9 Total 2.5 2.0 2.3 Religion Hindu 2.3 1.9 2.1 Islam 4.6 2.9 3.8 Christian 1.2 0.4 0.8 Others 2.4 1.9 2.2 Total 2.6 2.0 2.3 Household Size (number of members in the household) 1–3 3.5 2.4 3.1 4–7 2.6 1.8 2.2 8–10 2.6 3.0 2.8 11+ 1.1 1.2 1.2 Total 2.6 2.0 2.3 Poverty Status Extremely Poor 3.6 1.8 2.7 Poor 3.0 2.3 2.7 Marginal 2.4 2.1 2.2

248

Chapter Thirteen

Vulnerable 2.1 2.2 2.1 0.9 0.6 0.8 Middle Class 1.6 0.9 1.3 0.1 0.3 0.2 High Income Class 0.4 0.6 0.5 0.6 0.0 0.3 Total 2.6 2.0 2.3 1.6 0.7 1.2 Source: NSS unit records of the 66th round, EUS. All figures in percentages.

”ሺ݄݈݈ܾܿ݅݀ܽ‫ ݎݑ݋‬ൌ ͳሻ ൌ ߔሺߚ଴ ൅  ߚଵ ݄݈ܿ݅݀ܽ݃݁ ൅ ߚଶ ݄݈ܿ݅݀ܽ݃݁ ଶ ൅ ߚଷ ݄‫ ݁ݖ݅ݏ݈݀݋݄݁ݏݑ݋‬൅ ߚସ ݄‫ ݁ݖ݅ݏ݈݀݋݄݁ݏݑ݋‬ଶ ൅ ߚହ ‫݀݁ݏݏ݁ݏݏ݋݌݈݀݊ܽܽݐ݅݌ܽܿݎ݁݌‬൅ߚ଺ ܽ݃݁‫݈݀݋݄݁ݏݑ݋݄݄݁ݐ݂݋݄݄݀ܽ݁݁ݐ݂݋‬ ൅ ߚ଻ ‫ ݐݏ‬൅ ߚ଼ ‫ ܿݏ‬൅ ߚଽ ‫ ܾܿ݋‬൅ ߚଵ଴ ݅‫ ݈݉ܽݏ‬൅ ߚଵଵ ݄ܿ‫ݕݐ݅݊ܽ݅ݐݏ݅ݎ‬ ൅ ߚଵଶ ‫ ݕݐ݅ݎ݋݊݅݉ݎ݄݁ݐ݋‬൅ ߚଵଷ ‫ ݈݄݄݀݅ܿ݁ݐ݂݋݁ݏ‬൅ ߚଵସ ݈݈݅݅‫݁ݐܽݎ݁ݐ‬ ൅ ߚଵହ ‫ ݕݎܽ݉݅ݎ݌݋ݐ݌ݑ‬൅ ߚଵ଺ ‫ ݈݁݀݀݅݉݋ݐ݌ݑ‬൅ ߚଵ଻ ‫ݕݎ݀݊݋ܿ݁ݏ݋ݐ݌ݑ‬ ൅ ߚଵ଼ ‫ ݄݄݄ݔ݁ݏ‬൅ ߚଵଽ ‫ ܽ݊݁ݏ‬൅ ߚଶ଴ ݈ܾܽ݃ܽ ൅ ߚଶଵ ݈ܾ݉ܽ݊ܽ ൅ ߚଶଶ ‫݈݃ܽ݁ݏ‬ ൅ ߚଶଷ ݁‫ ݎ݋݋݌ݔ‬൅ ߚଶସ ‫ ݎ݋݋݌‬൅ ߚଶହ ݉ܽ‫ ݈ܽ݊݅݃ݎ‬൅ ߚଶ଺ ‫ݎ݈݊ݑݒ‬ሻ The model for the urban is almost the same as that of the rural with little variation in the independent variables. For urban India, we have four categories in the type of household category, unlike the five categories in rural India. These four categories are self-employed (se), regular (reg), casual (cas) and others, where others is the reference category.

V.2. Description of variables Dependent Variable: D_childlab= Dummy for the child to be in the labour force. Variable takes the value 1 if the child is in the labour force; 0 otherwise Independent Variables: Continuous Variables: Age=Age of the child age2= Square of the age of the child hh_size=Household Size hh_size2=Square of the household size pc_lnd_posd= Per capita L and Possessed age_hhh= Age of the head of the household Dummy Variables: d_ST= The variable takes the value 1 if the child belongs to ST; otherwise 0. d_SC= The variable takes the value 1 if the child belongs to SC; otherwise 0. d_OBC= The variable takes the value 1 if the child belongs to OBC; otherwise 0.

d_Islam= The variable takes the value 1 if the child belongs to Islam; otherwise 0. d_Christianity= The variable takes the value 1 if the child belongs to Christianity; otherwise 0.

The Determinants of Child Labour in India: A Probit Analysis

249

d_oth_minority= The variable takes the value 1 if the child belongs to Other minority religion; otherwise 0. d_sexchild= The variable takes the value 1 if the child is a boy; otherwise 0.

d_illiterate= The variable takes the value 1 if the child belongs to a household whose head is illiterate; otherwise 0. d_uptoprim= The variable takes the value 1 if the child belongs to a household whose head has education up to primary; otherwise 0. d_uptomiddle= The variable takes the value 1 if the child belongs to a household whose head has education up to middle; otherwise 0. d_uptosecondary= The variable takes the value 1 if the child belongs to a household whose head has education up to secondary; otherwise 0.

d_sexhhh= The variable takes the value 1 if the child belongs to household whose head is a female; otherwise 0. d_sena= The variable takes the value 1 if the child belongs to household whose type is Self employed in Non agriculture; otherwise 0. d_agllab= The variable takes the value 1 if the child belongs to household whose type is Agricultural labour; otherwise 0. d_manlab= The variable takes the value 1 if the child belongs to household whose type is Manual labour; otherwise 0. d_seagl= The variable takes the value 1 if the child belongs to household whose type is Self employed in agricultural; otherwise 0. d_se= The variable takes the value 1 if the child belongs to household whose type is Self Employed; otherwise 0. d_reg= The variable takes the value 1 if the child belongs to household whose type is Regular; otherwise 0. d_cas= The variable takes the value 1 if the child belongs to household whose type is Casual; otherwise 0. d_expoor=The variable takes the value 1 if the child belongs to household which is extremely poor; otherwise 0. d_poor=The variable takes the value 1 if the child belongs to household which is poor; otherwise 0. d_marginal= The variable takes the value 1 if the child belongs to household which is marginal; otherwise 0. d_vulnr= The variable takes the value 1 if the child belongs to household which is vulnerable; otherwise 0.

a a a a a a a a a a a a

VI. Results The tables showing the results focus on the marginal coefficients for interpretation which are obtained after Probit regression by using the mfx command in STATA.

250

Chapter Thirteen

Table 13.3. Probit estimates for rural India (2009–10) Variables

Probit Coeff.

S.E.

Age age_sq hh_size hh_size_sq pc_lnd_posd age_hhh d_sena d_agllab d_manlab d_seagl d_expoor d_poor d_marginal d_vulnr d_sexchild d_illiterate d_uptoprim d_uptomiddle d_uptosecondary d_sexhhh d_ST d_SC d_OBC d_islam d_christianity d_oth_minority Constant No of Obs. Pseudo R2 Log likelihood Ȥ2 Prob.> Ȥ 2

.1495** .0033 .0503** -.0028** -.1196* -.0073*** .3627*** .2543*** .2733*** .4959*** .1974*** .1902*** .0908 -.0461 .1383*** .6730*** .3711*** .2766*** .2343*** .1893*** .4209*** .2313*** .1425*** .2155*** -.3833*** .0454 -5.2141*** 56286 0.1945 -3944.378 1904.50 0.0000

.0630 .0028 .0227 .0012 .0635 .0015 .0669 .0747 .0702 .0668 .0668 .0611 .0602 .0567 .0500 .0796 .0807 .0846 .0907 .0500 .0540 .0527 .0433 .0455 .0817 .0686 .3643

Marginal Coeff. .0020** .0000 .0007** -.0000** -.0016* -.0001*** .0064*** .0046*** .0048*** .0092*** .0033*** .0030*** .0013 -.0006 .0019*** .0139*** .0066*** .0049*** .0041*** .0032*** .0086*** .0039*** .0020*** .0037*** -.0035*** .0006

S.E. .0008 .0000 .0003 .0000 .0008 .0000 .0015 .0017 .0015 .0016 .0013 .0011 .0009 .0007 .0004 .0024 .0018 .0018 .0019 .0016 .0016 .0011 .0006 .0009 .0005 .0010

Note: *** significant at 1% level, ** significant at 5% level, * significant at 10% level.

Table13.3 above presents the estimates of a Probit regression for rural India, revealing that most of the variables taken into consideration are highly significant. All independent variables do carry the expected and

The Determinants of Child Labour in India: A Probit Analysis

251

desirable signs, thus maintaining the usual and logical relationship between dependent and independent variables. Setting all the explanatory variables at their mean level, the model shows that the age of the child is a good explanatory variable with a 5% level of significance. As the age of a child in the sample increases, his or her chances of participating in the labour force also increase by 0.2%. The size of the household to which a child belongs determines the working status of the child. The results show that there is a direct relationship between the number of members in the household and the probability of a child being in the labour force. As the number of members in the household increases, the probability of the child’s being in the labour force increases by 0.07%, and this is significant at the 5% level. But the variable “hh_size_sq” has a negative effect on the likelihood of a child participating in the labour force. The per capita land possessed by the household is a significant variable in explaining the probability of the child being a labourer. An increase by a unit in the per capita land possessed by the household decreases the probability of a child in that household being a labourer by 0.0016 units. Education of the household head becomes a very strong factor in determining the child’s future. The reference category for the variable of education of the head of the household is being educated to the level of higher secondary and above. A child belonging to a household whose head is illiterate is 1.3% more likely to participate in the labour force compared to a child who belongs to a household whose head is educated to higher secondary and above. Likewise, if a child belongs to a household whose head has education to primary level, the probability of that child being a labourer is higher than the reference category by 0.6%. The probability of a child working also goes up by 0.5% and 0.4% if the head of the household has education to middle level and secondary level, respectively. The gender of the household head is also a very strong covariate in explaining the dependent variable. A child who belongs to a household whose head is female is 0.3% more likely to be working compared to that of a child belonging to a male-headed household. Household type is an indicator of the economic well-being of the household. The dummies of all the household types in this model proved to be highly significant. The reference category for household type is the surplus category of “others.” A child who belongs to a household whose major source of income comes from being self-employed in nonagricultural occupations is 0.6% more likely to be a labourer compared to a child who belongs to other household types. Likewise, a child from the household of an agricultural labourer is 0.4% more likely to participate in the labour force. A child from a household engaged in manual labour has

252

Chapter Thirteen

0.4% more chance of being in the labour force compared to other household types, and this becomes 0.9% more for those who are selfemployed in agriculture. The status of poverty of a household is an important variable affecting the status of a child in the labour force. There are four types of status dummies to measure poverty. The poverty status generally used is of two types only: a household is above the poverty line or below the poverty line. But in this paper, the status of poverty is further divided into different small categories depending on how much below or above the poverty line the households are. Children from the middle and high income classes are the reference category. The estimates show that a child from an extremely poor or poor household has a 0.3% better chance of participating in the labour force compared to a child from middle or high-income groups. These dummy variables are significant at the 1% level of significance. So far as the gender of the child is concerned, we find that a boy is 0.19% more likely to participate in the labour force compared to a girl, which is consistent with the cross-tabulation of the data. The social group and religion of the child are strong factors in determining their status in work. The reference category for social group is the “others” category, which often contains general caste people. It is found that an ST child has a 0.8% higher chance, an SC child a 0.3% higher chance and an OBC child a 0.2% higher chance of participating in the labour force than a child of the general caste. The study also confirms that a Muslim child is 0.3% more likely to be a labourer compared to a Hindu child, whereas a Christian child is exactly 0.3% less likely than a Hindu child to be a labourer. Table 13.4 reports the results for urban India. Like rural India, we find the age of the children in the sample significant at the 10% level in explaining the participation probability of a child in the labour force. So far as the education of the head of the household is concerned in urban India, we find that a child belonging to a household whose head is illiterate is 0.9% more likely to participate in the labour force compared to a child from a household whose head is educated to the higher secondary level and above. The probability that a child will work goes up by 0.3% compared to the reference category if he or she is from a household whose head is educated to the primary level, and the probability is higher by 0.1% if a child is from a household whose head is educated to the middle level. The gender of the household head is also a very strong variable in explaining the dependent variable in urban India. The reference category for the type of household in urban India is the surplus category of “others.” A child who belongs to a household whose major source of

The Determinants of Child Labour in India: A Probit Analysis

253

income is from self-employment is 0.1% more likely to be a labourer when compared to a child who belongs to the “other” household type. Likewise, a child from a household engaged in casual labour is 0.1% more likely to participate in the labour force. When we consider the status of the poverty of a household in urban India, the estimates show that a child from an extremely low income group has a 0.4% higher chance of being in the labour force compared to a child from a middle or high income group. Similarly, for the poor income class, the chance is higher by 0.2%, and for the marginal income class, the figure becomes 0.3% higher. For urban India, a boy is 0.13% more likely to participate in the labour force compared to a girl child. Here, it is noted that a boy aged 5–14 is less prone to participate in the labour market in urban India than a boy in rural India. The results show that a ST Child has a 0.1% higher probability of participating in the labour force than a child of the general caste. Our study confirms that a Muslim child is more likely to be a labourer than a Hindu child in urban India. In the model for urban India, the household size and gender of the head of the household do not have any significant effect on the probability of a child being in the labour force. Table 13.4. Probit estimates for urban India (2009–10) Variables hh_size hh_size_sq Age age_sq pc_lnd_posd age_hhh d_se d_reg d_cas d_expoor d_poor d_marginal d_vulnr d_sexchild d_illiterate d_uptoprim d_uptomiddle d_uptosecondary d_sexhhh d_ST d_SC d_OBC

Probit Coeff. 0.0293 -0.0028 0.2328* 0.0014 -0.1019 -0.0070*** 0.3343** 0.1781 0.2556* 0.6145** 0.4476* 0.4886* 0.1954 0.3317*** 0.8844*** 0.5515*** 0.2341* 0.0920 -0.0110 0.1909* 0.0932 -0.0847

S.E. 0.0420 0.0025 0.1333 0.0059 0.3024 0.0027 0.1371 0.1423 0.1404 0.2520 0.2511 0.2506 0.2479 0.0516 0.1105 0.1108 0.1201 0.1342 0.0834 0.1080 0.0771 0.0620

Marginal Coeff. 0.0001 0.0000 0.0009* 0.0000 -0.0004 0.0000*** 0.0015** 0.0008 0.0013* 0.0049** 0.0029* 0.0035* 0.0009 0.0013*** 0.0092*** 0.0039*** 0.0012* 0.0004 0.0000 0.0010* 0.0004 -0.0003

S.E. 0.0002 0.0000 0.0004 0.0000 0.0012 0.0000 0.0007 0.0007 0.0010 0.0036 0.0026 0.0030 0.0014 0.0003 0.0025 0.0014 0.0008 0.0007 0.0003 0.0007 0.0004 0.0003

254

Chapter Thirteen

d_islam 0.1197* d_christianity -0.0129 d_oth_minority -0.0978 Constant -6.2588*** No of Obs 31435 Pseudo R2 0.2596 Log likelihood -1429.8227 Ȥ2 1002.80 Prob.> Ȥ 2 0.0000 Note: *** significant at 1% level, 10% level.

0.0633 0.1408 0.1450 0.8014

0.0005* -0.0001 -0.0003

0.0003 0.0006 0.0005

** significant at 5% level, * significant at

VII. Conclusion Using data from the NSS (2009-10) at the individual level, we examine the factors that affect the participation probability of a child in both urban and rural labour market in India. We also analyse the trend of incidence of child labour utilizing three rounds of the NSS unit level data (1993–4, 2004–5 & 2009–10). This paper finds that India has experienced a decrease in the incidence of child labour for both boys and girls, and that the decrease is sharper in the case of girls irrespective of the sectors (rural or urban). The comparison of urban and rural India demonstrates that the decrease in the incidence of child labour in urban India is more than that in rural India. A study of the trends of child labour incidence (LFPR) for different states over the period 1993–4 to 2009–10 brings forth some interesting observations. The trajectory of LFPR takes different shapes such as overall increasing, overall decreasing, U-shaped or inverted U-shaped. It is found that the incidence of child labour is on a continuous increase in Arunachal Pradesh. A majority of the states follow an inverted U-shaped pattern which is indicative of decent performance by these states in tackling child labour problem. There is, however, concern for the states like Chandigarh, Assam, Gujarat, Pondicherry and Delhi in fighting child labour problem. This study adds to the literature by raising questions as to why states behave differently with respect to the trajectory of the incidence of child labour. However, this paper has not attempted to answer such delicate questions. So far as the household profile is concerned, we find that households which are self-employed in agriculture and engaged in casual labour send more children to the labour force in rural and urban India, respectively. ST households and Muslim households lead in contributing more child labour both in rural and urban India. As expected, the study

The Determinants of Child Labour in India: A Probit Analysis

255

finds that the incidence of child labour is the highest in extremely poor households. Whether the children will engage themselves in the labour market is a composite outcome of different socio-economic household variables. Although many of the individual variables in our regression model, like age and gender, explain our dependent variable quite well, the most effective policy variables recognized are status of poverty of the household and the education of the head of the household. Education of the head of the household enhances the earning capacity of the household, and the educated elders evaluate the return to schooling of their children in the long run. Hence, it is expected that educated parents are less likely to send their children to the labour force than their counterparts. Our results have some important policy implications. In particular, government should focus on imparting education to the uneducated parents, generating employment opportunities and ensuring proper implementation of poverty alleviation programmes for the poor. Moreover, initiating reforms in broader societal attitudes against the practice of child labour will be helpful in reducing the incidence of child labour. Formulating laws alone to ban child labour will not suffice, unless the root causes of child labour are tackled through reforms in the socio-economic structure of the country.

References Anker, R. 2000. The Economics of Child Labor: A Framework for Measurement. International Labor Review. 139:257–80. Basu, K. 1993. Lectures in Industrial Organization Theory. Oxford: Blackwell Publishers. —. 1999. Child Labor: Cause, Consequence, and Cure with Remarks on International Labor Standards. Journal of Economic Literature. 37:1083–1119. —. 2003. International Labour Standards and Child Labour. In Child Labour and the Right to Education in South Asia: Needs Vs Rights?. edited by Naila Kabeer, Geetha B Nambissan and Ramya Subrahmanian. New Delhi: Sage Publications. Basu, K. & P. H. Van. 1998. The Economics of Child Labor. American Economic Review. 88:412–27. Beegle, K. R. Dehejia. & R. Gatti. 2009. Why Should We Care About Child Labor? The Education, Labor Market, and Health Consequences of Child Labor. Journal of Human Resources. 44:871–89.

256

Chapter Thirteen

Bhargava, P. H. 2003. The Elimination of Child Labour: Whose Responsibility. New Delhi: Sage Publications. Bhatty, K 1998. Education Deprivation in India: A Survey of Field Investigations. Economic and Political Weekly. 33:1731–40, 1858–69. Bonnet, M. 1993. Child Labor in Africa. International Labor Review. 132:371–89. Boozer, M.A. & T. K. Suri. 2001. Child Labor and Schooling Decisions in Ghana. Yale University: Mimeo. Burra, N. 1995. Born to Work: Child Labour in India. New Delhi: Oxford University Press. Cavalieri, C. 2002. The Impact of Child Labor on Educational Performance: An Evaluation of Brazil. Manuscript. Chaudhri, D. P. 1996. Dynamic Profile of Child Labour in India 1951– 1991. New Delhi: ILO. Cigno, A. 1991. Economics of the Family. Oxford and New York: Clarendon Press and Oxford University Press. Cigno, A. & F. C. Rosati. 2000. Why do Indian Children Work, and is it Bad for Them?. Discussion Paper No. 115, Bonn: IZA. Cigno, A., F. C. Rosati & Z. Tzannatos. 2001. Child Labor, Nutrition and Education in Rural India: An Economic Analysis of Parental Choice and Policy Options. Social Protection Discussion Paper Series No. 131. Washington D.C.: World Bank. —. 2002. Child Labour Handbook. Social Protection Discussion Paper Series No. 206. Washington D.C. : World Bank. Das, S. & D. Mukkherjee. 2007. Role of Women in Schooling and Child Labour Decision: The Case of Urban Boys in India. Social Indicator Research. 82:463–486. Deaton, A. 2003. Prices and Poverty in India. Economic and Political Weekly. 38: 362–68. Dreze, J. & G. Kingdom. 2000. School Participation in Rural India. NEUDC Conference. Emerson, P. M. & A. P. Souza. 2011. Is Child Labor Harmful? The Impact of Working Earlier in Life on Adult Earnings. Economic Development and Cultural Change. 59:345–385. Eswaran, M. 2002. The Empowerment of Women, Fertility, and Child Mortality. Journal of Population Economics. 15:433–54. Folbre, N. 1986. Hearts and Spades: Paradigms of Household Economics. World Development. 14:245–55. Gustafsson-Wright, E. & H. H. Pyne. 2002. Gender Dimensions of Child Labor and Street Children in Brazil. World Bank Policy Research Working Paper No. 2897.

The Determinants of Child Labour in India: A Probit Analysis

257

Hoddinott, J. & L. Haddad. 1995. Does Female Income Share Influence Household Expenditures?. Evidence from Cote d’Ivoire. Oxford Bulletin of Economics and Statistics. 57:77–96. Lavy, V. 1996. School Supply Constraints and Children's Educational Outcomes in Rural Ghana. Journal of Development Economics. 51:291–314. Misra, L. D. 2000. Child Labour in India. New Delhi: Oxford University Press. Mukkherjee, D. & S. Das. 2008. Role of Parental Education in Schooling and Child Labour Decision: Urban India in the Last Decade. Social Indicator Research. 89:305–22. Patrinos, H. A. & G. Psacharopoulos. 1995. Educational Performance and Child Labor in Paraguay. International Journal of Educational Development .15:47–60. Patrinos, H. Anthony & G Psacharopoulos. 1997. Family Size, Schooling and Child Labor in Peru –An Empirical Analysis. Journal of Population Economics. 10:387–405. Psacharopuolos, G. 1997. Child Labor versus Educational Attainment: Some Evidence from Latin America. Journal of Population Economics. 10:337–86. Ray, R. & G. Lancaster. 2003. Does Child Labour Affect School Attendance and School Performance?. Multi Country Evidence on SIMPOC Data, Manuscript. Rosenzweig, M. R. & R. Everson. 1977. Fertility, Schooling, and the Economic Contribution of Children in Rural India: an Econometric Analysis. Econometrica. 45:1065–79. Sengupta, A., K. P. Kannan & G. Raveendran. 2008. India’s Common People: Who are they, How many are they and How do they Live?. Economic and Political Weekly. 43:49–63. Schultz, T. Paul. 1997. Demand for Children in Low Income Countries. In Handbook of Population and Family Economics. edited by Mark R. Rosenzweig and Oded Stark. 1A:349–430. Amsterdam: Elsevier. Throat, S. 1999. Poverty, Caste and Child Labour in India: The Plight of Dalit and Adivasi Children. In Against Child Labour: India and International Dimensions and Strategies. edited by K. Voll. New Delhi: Berlin Studies in International Politics, Mosaic Books. Weiner, M. 1991. The Child and the State in India: Child Labour and Education Policy in Comparative Perspective. Princeton: Princeton University Press.