Economics of Child Labour [1st ed.] 978-981-13-8198-0;978-981-13-8199-7

Table of contents :
Front Matter ....Pages i-ix
Introduction (Biswajit Chatterjee, Runa Ray)....Pages 1-6
Incidence of Child Labour—Global and Indian Estimates (Biswajit Chatterjee, Runa Ray)....Pages 7-28
Types and Causes of Child Labour (Biswajit Chatterjee, Runa Ray)....Pages 29-39
Survey of Literature (Biswajit Chatterjee, Runa Ray)....Pages 41-68
Impact of Trade Versus Non-trade Policy on Child Labour in a Two Sector General Equilibrium Framework (Biswajit Chatterjee, Runa Ray)....Pages 69-85
Effectiveness of Trade and Non-trade Policies on the Incidence of Child Labour—A Three-Sector General Equilibrium Framework (Biswajit Chatterjee, Runa Ray)....Pages 87-150
Conclusion (Biswajit Chatterjee, Runa Ray)....Pages 151-156

Citation preview

Biswajit Chatterjee · Runa Ray

Economics of Child Labour

Economics of Child Labour

Biswajit Chatterjee Runa Ray •

Economics of Child Labour

123

Biswajit Chatterjee Department of Economics Jadavpur University Kolkata, West Bengal, India

Runa Ray Department of Economics Vidyasagar College Kolkata, West Bengal, India

ISBN 978-981-13-8198-0 ISBN 978-981-13-8199-7 https://doi.org/10.1007/978-981-13-8199-7

(eBook)

© Springer Nature Singapore Pte Ltd. 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

Economics of Child Labour is an outcome of joint research by the authors over the years on incidence of child labour in developing countries, with particular focus on effectiveness of trade and non-trade policies to combat this menace, which has narrowed down on the world scale in recent years, yet remained fairly high particularly in developing countries in Africa and Asia. The study develops a general equilibrium framework to view the effects of various trade and non-trade policies on the incidence of child labour. Placed in the background of profile of child labour estimates in different regions of the globe in recent years, and also in India, the study makes a detailed critical review of both theoretical and empirical literature on child labour estimates and policies to mitigate the incidence of child labour. It then moves to a detailed analysis of multisectoral general equilibrium models to reflect on the possible effects of growth-promoting non-trade policies on child labour employment vis-à-vis direct trade policies aimed at reduction of the incidence of child labour. Since the ultimate instrument for elimination of child labour is alleviation of poverty, we have also investigated how education subsidy policy is effective in enhancing school enrolment rate. However, the impact on child labour supply is not always favorable. The most effective way to draw children out of damaging work is to encourage school attendance. Introduction of compulsory education will be an effective instrument for curtailing child labour problem when targeted subsidies are given to families for sending their children to school. Though education by itself will not eradicate child labour, it will reduce and discourage it, provided appropriate targeted education subsidy could be implemented. Social protection measures, which are non-trade policies, when appropriately designed and targeted, may have a diminutive impact on the incidence of child labour. Various policy alternatives have been suggested to combat the incidence of child labour in developing countries. One policy is of course administrative or legislative, i.e., ban on the use of child labour. Such a prescription of total ban on child labour use may not be effective, as there is imperfect monitoring, and loss of income from working children may actually compel parents to send their children to work, rather than to school; and since child labour is cheaper compared to adult labour, the producers would also be interested in increasing the demand for child labour. The v

vi

Preface

other alternative suggested by the WTO in imposing labour standards, is to put up tariff wall against exports of developing countries, as these products are often child labour intensive in nature. Trade policy alternatives, if implemented, have the possible consequences of generating various forms of domestic distortions, depending upon the structures of the economies, which often adversely affect national welfare, and may not always result in the diminution of the incidence of child labour in such developing economies. The other alternative is non-trade policy package, which includes reorientation of growth strategies towards non-child labour, using production sectors, policy on education subsidy to encourage sending children to school, and not towards work etc. Our studies in the present monograph analyse the implications of non-trade policies vis-à-vis trade policies towards child labour in developing economies in terms of alternative structures of production and labour use in general equilibrium framework. It is expected that this monograph will encourage future direction of research on child labour and its mitigation strategies in developing countries. Kolkata, India

Biswajit Chatterjee Runa Ray

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 6

. . . . . .

. . . . . .

. . . . . .

7 7 8 16 27 27

3 Types and Causes of Child Labour . . . . . . . . . . 3.1 Types of Child Labour . . . . . . . . . . . . . . . . . 3.2 Different Causes for Child Labour . . . . . . . . . 3.3 Quantitative Derivation of Supply Function of 3.4 A Summing-Up . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

............. ............. ............. Child Labour* . . ............. .............

. . . . . .

. . . . . .

29 30 33 34 39 39

4 Survey of Literature . . . . . . . . . 4.1 Introduction . . . . . . . . . . . . 4.2 General Issues . . . . . . . . . . 4.3 Empirical Literature . . . . . . 4.4 Theoretical Literature* . . . . 4.5 General Equilibrium Models 4.6 A Step Forward . . . . . . . . . References . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

41 41 42 44 52 58 63 65

2 Incidence of Child Labour—Global and Indian Estimates 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Global Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Estimates of Child Labour in India . . . . . . . . . . . . . . . 2.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . .

. . . . . . . .

. . . . . .

. . . . . . . .

. . . . . .

. . . . . . . .

. . . . . .

. . . . . . . .

. . . . . .

. . . . . . . .

. . . . . . . .

5 Impact of Trade Versus Non-trade Policy on Child Labour in a Two Sector General Equilibrium Framework . . . . . . . . . . . . . . 5.1 General Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Child Labour in a Two-Sector General Equilibrium Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

69 70 72

vii

viii

Contents

5.3 Comparative Static Exercises . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Imposition of Stringent Trade Restriction by Rest of the World on the Exported Product of the Small Open Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Alternative Specification of Utility Function and Effect on Child Labour Supply . . . . . . . . . . . . . 5.3.3 Alternative Policy Option for Child Labour Problem—Effect of Protectionism in Import Competing Sector . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.....

75

.....

75

.....

77

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

80 83 84

6 Effectiveness of Trade and Non-trade Policies on the Incidence of Child Labour—A Three-Sector General Equilibrium Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Child Labour in a Three-Sector General Equilibrium Framework Without Any Adult Unemployment . . . . . . . . . . . . . . . . . . . . . 6.2 Comparative Static Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Comparison Between Alternative Policies . . . . . . . . . . . . . . . . 6.4 Child Labour in a Three-Sector General Equilibrium Framework in the Presence of Urban Adult Unemployment . . . . . . . . . . . . 6.4.1 The Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.2 Effect of Imposition of Stringent Trade Restriction by Rest of the World on Exported Product of the Small Open Economy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4.3 Alternative Trade Policy Option—Effect of Protectionism in Non-traded Intermediary Producing Sector . . . . . . . . 6.4.4 Effect of Protectionism in Import Competing Sector . . . 6.4.5 Effects of Protectionism on National Welfare . . . . . . . . 6.4.6 Impact on Trade Balance for Various Trade Policies . . . 6.4.7 Comparison of Alternative Policies . . . . . . . . . . . . . . . . 6.5 Education Subsidy, Adult Unemployment and The Incidence of Child Labour—A Three-Sector General Equilibrium Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.2 The Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.5.3 Comparative Static Exercises . . . . . . . . . . . . . . . . . . . . 6.6 Comparison of Alternative Policies . . . . . . . . . . . . . . . . . . . . . Appendix 6.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Appendix 6.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

..

87

.. 89 .. 92 . . 110 . . 111 . . 111

. . 114 . . . . .

. . . . .

119 124 128 131 135

. . . . . . . .

. . . . . . . .

136 136 137 139 145 147 148 149

7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

About the Authors

Biswajit Chatterjee is a Retired Professor at the Department of Economics, Jadavpur University. He is the former President of The Indian Econometric Society and the Indian Society of Labour Economics, and is the President of the Bengal Economic Association. He specialises in development economics, international trade, and the Indian economy. Dr. Runa Ray is an Associate Professor of Economics at Vidyasagar College Kolkata, and has contributed research in the field of international trade and development economics.

ix

Chapter 1

Introduction

Abstract This chapter reviews the concepts used to describe the phenomenon of child labour across the globe, mostly in developing nations of Africa, Asia and the Pacific, and takes stock of the issues of sanctions to be used against child labour use, following the WTO directives on labour standards. The alternative forms of trade and non-trade policies may be relevant in tackling the issue of child labour use, which often rations children’s access to education and schooling.

The term ‘Child Labour’ is often used as a synonym for ‘employed child’ or ‘working child’. An exact definition of child labour is very difficult in view of the variety of types of work they are engaged in and considerable overlaps between the categories defined in the literature. One may define the phenomenon as that segment of the child population which participates in work, either paid or unpaid. They may be working with the parents inside family occupations or may be working in occupations as separate entities outside the ambit of family. Any work done by children that interferes with their full physical and mental development, denies them the opportunities for education and learning and the needed recreation are called child labour (Stein and Davis 1940). On the basis of this definition, we may distinguish between two types of child labour: (1) those who assist their parents or family members in their works, and (2) those who work outside their family for remuneration, either in real terms Notes: This chapter draws from and reuses with permission the materials contained in the following papers of the authors: (a) Biswajit Chatterjee & Runa Ray (2016), “Impact of Trade Restriction on Child Labour Supply and the Role of Parents’ Utility Function: A Two Sector General Equilibrium Analysis” in Malabika Roy and Saikat Sinha Roy (eds), International Trade and International Finance, Springer 2016. (b) Runa Ray & Biswajit Chatterjee (2010c), “Education Subsidy, Adult Unemployment and the Incidence of Child Labour in an Open Economy: A Three Sector General Equilibrium Analysis”, The Indian Economic Journal, Vol 57, No 4, Jan–March 2010, pages 37–59. Also the materials from the following reports are used with permissions in different parts of this chapter: (c) ILO (2017), Global Estimates of Child Labour: Results and Trends, 2012–2016, Geneva, 2017. (d) UNICEF (2016), State of World’s Children, 2016—A fair Chance for Every Child, UNICEF, New York. (e) UNICEF (1977), State of World’s Children, 1977, UNICEF, New York. © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_1

1

2

1 Introduction

or in cash, which help them supplement their family income. We can distinguish alternative forms of child labourers visible in the following way: Child Labourers Wage Labourers

Those who get cash wages plus

Those who get wages only in cash

Self-employed Children

Self-employed Independently

Self- employed in family

enterprise

Available estimates indicate that about millions of children around the world are engaged in child labour, and they are condemned to a life of poverty and misery affecting their childhood, their health and education. They are classified as child labourers when they are either too young to work or are involved in hazardous activities that may seriously affect their physical, mental, social and educational development. Although the prevalence of child labour is highest in sub-Saharan Africa followed by the Asia and the Pacific region, in the so-called advanced countries in Europe, America and Saudi Arabia also, the incidence of child labour, though small in relative terms, is non-null. Within Asia, the range of child labour use varies between zero per cent in Japan and Hong Kong to over 55% in Bhutan and 45% in Nepal. India, Bangladesh, Pakistan and Thailand, respectively, account for 14%, 30%, 17% and 16% of child labour in the world. The other worst forms of child labour include slavery and slavery-like practices like forced and bonded labour, child soldiering, sexual exploitation or are used by adults in illicit activities including drug trafficking. Child labour is the combined result of many factors, such as poverty, lack of decent work opportunities for adults and adolescents, migration and imperfect credit markets. The continued persistence of child labour has severe negative shortand long-term consequences for the fulfilment of children’s rights guaranteed by the United Nations Convention on the Rights of the Child (CRC)—including denial of education and frequent exposure to violence (UNICEF 2016). According to the ILO Report (2017), different nomenclatures are used in official communications to depict categories of child labour use. One prominent category mentioned is Children in employment which refers to those working in any form of market production and certain types of non-market production (e.g. the production of agricultural produce for own use). This group includes children in forms of work engaged in both the formal and informal economy; inside and outside family settings; for pay or profit (in cash or in kind, part time or full time); and domestic work outside the child’s own household for an employer (paid or unpaid). The next category mentioned is Children in child labour, which is a narrower category than children in employment, as it excludes children in employment who are in permitted light work and those above the minimum age whose work is not classified as the worst form of child labour, or, in particular, as ‘hazardous work’. The category Children

1 Introduction

3

in the worst forms of child labour comprise: (a) all forms of slavery or practices similar to slavery, such as the sale and trafficking of children, debt bondage and serfdom, and forced or compulsory labour, including forced recruitment of children for use in armed conflict; (b) the use of a child for prostitution, for the pornographic performances and for illicit activities, like drug trafficking, etc., and (c) work, which by its nature or environment, may affect the health, safety and morals of children. The last category is often described as belonging to Children in hazardous work, which may include ‘night work and long hours of work, exposure to physical, psychological or sexual abuse; work underground at dangerous heights or in confined spaces; and work in an unhealthy environment, etc. Since reliable national data on the worst forms of child labour other than hazardous work, such as children in bonded and forced labour or in commercial sexual exploitation, are difficult to obtain, hazardous work by children is often treated as a proxy category for the worst forms of child labour and further, children in hazardous work account for the overwhelming majority of those in the ‘worst forms of child labour’ (ILO 2017). There is no doubt that the above categories of child labour as are referred in official estimates and reports are overlapping in nature and clear-cut boundaries between each category seem to be impossible in view of the arbitrariness involved. There are also Children performing household Chores as referring to those performing domestic and personal services for consumption within their own households, and constitute a ‘non-economic’ form of production and are excluded from consideration in the UN System of National Accounts (UNSNA). The real problem of children at work is that in this case the children are definitely excluded from school attendance. The magnitude of child labour in the world is high on the global scale, and there are estimates churned by reports of international bodies like the ILO or the UNICEF. The problem of the incidence of child labour is an important issue today not only for economic analysts, but also for government, social groups and concerned citizens of every nation. That the practice of using child labour is undesirable and needs to be abolished, and instead children should be enrolled in educational or vocational programs, so that they could possess skills for a better future, are in general agreed upon in policy discussions, but with imperfect institutions and inadequate resources (as in the case in most developing nations), the eradication of the incidence of child labour is not taking place and educational opportunities for these working children are not expanding in the desired directions and speed. One may explain the prevalence of child labour on grounds of economic necessities of the poor households in developing countries in terms of a demand— supply framework. In all the three major sub-sectors of such an underdeveloped economy—agriculture, manufacturing and services, there are demands for child labour, although the exact incidence varies across sectors. When child labourers are employed in the agriculture sector, they help in agricultural operations like sowing, weeding, harvesting and threshing, and such child workers are also employed in sectors allied to fishing. There exist a concentration of child labour in cottage and household industries within the secondary sector in many developing nations. A number of factors explain the concentration of children in these sectors. First, such industries can easily evade legal restrictions. Second, it is believed that the human

4

1 Introduction

body is soft, flexible and pliable in childhood and therefore, the children learn with ease how to twist, bend, stretch or assume different gestures as required by a particular job. This is the ‘nimble finger’ argument. Moreover, children are more amenable to discipline and control—they can be coaxed, admonished, pulled up and punished for defaults than the adult labourers. Children are not organized and thus are paid a minimum wage to work for long hours. More importantly, since the productivity per man in several such industries, mentioned above, is low, these enterprises cannot afford to hire adult workers, who are often organized and would demand higher wages. Thus, the principal motive of employment of child labour is to minimize costs to have a comparative advantage in the world market and thus to maximize profits, as many of these small-scale enterprises produce for the world market. In all other sectors, the concentration of child workers is much below than that of general workers. The concentration is the least in the case of transport, storage and communication, banking, insurance (mostly services sector activities) followed by construction, mining. In urban areas, children are engaged in more diversified activities such as loading–unloading, hawking, looking after parked vehicles, rag-picking and shoe-shining. They are also employed as helpers in teashops and auto-repair shops or are engaged as domestic servants (Chatterjee and Ray 2016). Trade sanctions, import tariffs and product labelling (for example, the Rugmark initiative in the carpet industry) has been proposed and in some instances implemented to reduce the extent of child labour. However, such sanctions may not have their intended effect (see Bhagawati 1995; Dixit 2000; Jafarey and Lahiri 2002; Maskus 1997; Ranjan 1999, 2001). Edmonds and Pavcnik (2002) find empirical support for the concern that trade sanctions may indeed raise child labour in the rice sector in Vietnam using data from 1993 to 1998. On the supply side, child labour is often viewed as being driven by the needs of basic survival in developing nations, where children need to work to sustain themselves or their families. Basu and Van (1998) recognize this and establish that even though adults dislike child labour; they may have to endure it for survival. On the other hand, if the endowment/income level of the family is much high, then these same households will not supply child labour. In other words, there are at least two reasonable equilibria, one where children work and the other where they do not. Therefore, children work in developing nations not because of a lack of potential consciousness, but rather because of sheer economic necessity. Baland and Robinson (2000) also find altruistic parents enduring child labour because of poverty (the case of zero bequests) or imperfect capital markets. Higher family incomes, improved educational opportunities (Basu 1999; Maskus 1997; Ray 2002) and a general increase in the living standards of the poor through liberalized trade and improved access to world markets (as in Bhagawati 1995; Dixit 2000) are expected to reduce child labour. The neoclassical theory of child labour supply explains that the household or family supplies child labour for jobs in order to maximize its current income from the employment of the child labour in preference to the income expected from their employment in future after the schooling of its children. In fact, many of the households in a developing country decide for maximizing the present income rather

1 Introduction

5

than for the future income, which involves higher opportunity cost of schooling and investment of time and income going in for the future income. Moreover, the low life expectancy of children in developing countries is another reason why a household decides to maximize the present income rather than the future one. The issue of imposing trade restrictions on products using child labour is related to the agitation for minimal international ‘labour standards’ and ‘social clauses’ as a prerequisite for trade. Use of child labour in production is seen as one way of violating labour standards; and imposing sanctions on the import of goods produced by child labour is one such social clause. The argument in favour of the clause is that use of child labour gives the developing countries a trading advantage in labour-intensive goods and this advantage is illegitimate because child labour is socially unacceptable (vide https://www.blackwellpublishing.com/content/BPL_ Images/Journal_Samples/DPR0950-6764~20~3/174.PDF). Following WTO’s adoption of labour standards, the question that has been debated in the academic and policy circles is whether the use of trade sanctions on the products that use child labour can be effective in reducing the incidence of child labour or not. In the context of the discussion on globalization and poverty, the question of child labour use in different production activities particularly in the export sector of many developing countries, has been raised. It is argued that globalization may increase the employment and earnings opportunities available to poor households in developing countries, through promotion of economic growth (Edmonds 2005). But as the developing countries integrate into the world economy and increasingly rely on export markets to sell their products in a globalized market environment, rich countries raise the issue of unfair dumping by the developing nations as they often use cheap child labour in their exportable sector, and can use the threat of trade sanctions to adopt policies that attempt to curtail child labour. In December 2001, EU trade commissioner Pascal Lamy announced that EU foreign ministers have approved a preferential tariff scheme for countries that adhere to ILO labour standards including child labour. Both the Clinton and Bush administrations have voiced a commitment to a similar policy, and in 1999 the U. S. imposed quotas on Cambodia’s garment industry because of poor working conditions and use of child labour (vide Eric V. Edmonds: www.dartmouth.edu/~eedmonds/nzzessay.pdf) Thus, in the context of globalization process in the world economy, the high-income countries sought to punish poor nations for high levels of child labour, and practice veiled protectionism. This tendency has been strengthened after the formation of WTO. Since the developing world experiences high prevalence of child labour, the important issue is whether they should rely on trade policy instruments like tariffs or ban on exports of products that are child labour-intensive or take recourse to non-trade policy interventions like reorganization of production in favour of sectors that use more adult labourers, and initiate economic expansion through domestic capital accumulation or use policies like education subsidies to attract children to schooling and discourage them from engaging in work. We assess the relative impacts of alternative trade and non-trade policies with the help of general equilibrium framework,

6

1 Introduction

suitably modified to capture the ramifications of production conditions in developing countries. These are discussed in detail in Chaps. 5 and 6 of the present book.

References Baland, J. M., & Robinson, J. A. (2000). Is child labour inefficient. Journal of Political Economy, 108(4), 663–679. Basu, K. (1999). Child labour: Cause, consequence, and cure, with remarks on international labour standards. Journal of Economic Literature, 37(3), 1083–1119. Basu, K., & Van, P. H. (1998). The economics of child labour. American Economic Review, 88(3), 412–427. Bhagwati, J. (1995). Trade liberalization and ‘fair trade’ demands: Addressing the environment and labour standards issues. World Economy, 18, 745–759. Chatterjee, B., & Ray, R. (2016). Impact of trade restriction on child labour supply and the role of parents’ utility function: A two sector general equilibrium analysis. In M. Roy & S. S. Roy (Eds.), International trade and international finance. Springer. Dixit, A. (2000). Comment on “A transaction cost politics analysis of international child labour standards. In A. V. Deardonff & R. Stern (Eds.), Social dimensions of U.S. trade policies. Edmonds, E. (2005). Child labour in the global economy. Journal of Economic Perspectives, 19(1), 199–220. Edmonds, E., & Pavcnik, N. (2002). Does globalization increase child labour? Evidence from Vietnam. NBER working paper, no. 8760. ILO. (2017). Global estimates of child labour: results and trends, 2012–2016. Geneva. Jafarey, S., & Lahiri, S. (2002). Will trade sanctions reduce child labour? The role of credit markets. Journal of Development Economics, 68(1), 137–156. Maskus, K. E. (1997). Should core labour standards be imposed through international trade policy? World Bank, Policy research working paper, no. 1817. Ranjan, P. (1999). An economic analysis of child labour. Economic Letters, 64(1), 99–105. Ranjan, P. (2001). Credit constraints and the phenomenon of child labour. Journal of Development Economics, 64(1), 81–102. Ray, R. (2002). The determinants of child labour and child schooling in Ghana. Journal of African Economics, 11(4), 561–590. Stein, E., & Davis, J. (Eds.). (1940). Labour problem in America. F. Forraer and Rinchar. UNICEF. (2016). State of world’s children, 2016—A fair chance for every child. New York: UNICEF.

Chapter 2

Incidence of Child Labour—Global and Indian Estimates

Abstract The estimates of the incidence of child labour, as are available from the reports of international bodies like the ILO and the UNICEF, indicate the widespread prevalence and large numbers in different parts of the globe. Although there is a declining trend in the incidence of Child labour in recent years, yet the experience over the past 16-year period has not been one of quickening, or even steady progress, so that to achieve the Target 8.7 of the Sustainable Development Goals of the United Nations to ending child labour in all its forms by 2025, we need the adoption of policies by different national governments to speed up the decline in the incidence of Child labour in various activities. The census data for India regarding the incidence of child labour and its spread across the states of India, also reveal a similar picture.

2.1 Introduction Child labour in various forms is considered a curse on modern development as it deprives children of enjoying their childhood and robs them of the freedom to go to school and participate in a happy healthy and knowledgeable life. Its prevalence across the globe also reflects the poor distribution of developmental gains across generations, space and time so that endemic poverty and unemployment have constrained many economies and their people. Child labourers constitute a group of working children who are either too young to work or are engaged in hazardous activities which are potentially harmful to their physical, social, psychological or educational development. The guiding international conventions on this issue are the International Labour Organization (ILO) Convention No. 138 on the minimum age for admission to employment and work, ILO Convention No. 182 on the worst forms of child labour, and the United Nations Convention on the Rights of the Child, form the basis for child labour legislations enacted by signatories of the above ConvenNotes: This chapter draws heavily from the ILO Report ‘Global Estimates of Child Labour: Results and Trends, 2012–2016’, Published by International Labour Office (ILO), Geneva, 2017. Also see UNICEF (2013), Impact of Unpaid Household Practices on the Measurement of Child Labour, MICS Methodological Papers no 2, October, 2013. Census reports, Government of India are used in sections on Estimates of Child Labour in India. © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_2

7

8

2 Incidence of Child Labour—Global and Indian Estimates

tions The ILO (2017) report, the fifth edition of the ILO’s quadrennial report series on global estimates of child labour, charts the progress so far in reducing the incidence of child in different parts of the globe. The report, and the global estimation exercise that underpins it forms part of a broader inter-agency effort to measure and monitor progress towards the Target 8.7 of the Sustainable Development Goals (SDGs) of the United Nations. In response to the persistent and emerging challenges of development, the global community adopted the 2030 Sustainable Development Goals (SDGs) under the ‘2030 Agenda’, whereby the Target 8.7 called on the global community to end child labour in all its forms by 2025 Elimination of endemic nature of child labour in the present world would require both economic and social reform as well as the active cooperation of all stakeholders including governments, enterprises, workers’ and employers’ organizations, international organizations, and civil society at large. According to the UNICEF (2013), above-mentioned ILO Conventions, however, exclude activities that are not economic in nature, i.e. those that fall outside the production boundary, as defined by the United Nations System of National Accounts (SNA). Such activities include unpaid household services (UHS)—more commonly known as household chores—like cleaning the house, looking after siblings, washing dishes and shopping. The UNICEF-supported Multiple Indicator Cluster Surveys (MICS) define child labour according to the Convention on the Rights of the Child and other relevant international conventions. The definition of child labour includes 5- to 11-year olds engaged in economic activities, 12- to 14-year olds who work (in economic activities) for 14 or more hours per week, as well as children who carry out hazardous UHS. The UNICEF’S MICS use the threshold for hazardous UHS as 28 h per week, i.e. children engaged in such hazardous activities for more than 28 h per week are categorized as child labourers.

2.2 Global Estimates The 2016 estimates contained in ILO (2017) report, use data from a total of 105 national household surveys covering more than 70% of the world population of children aged 5–17 years, covering all regions including the OECD countries and the People’s Republic of China. The distribution of national data sets by type of source include: 24 national data sets derived from Child Labour Surveys (CLS) implemented by the ILO in collaboration with national bureaus of statistics, 17 national data sets from the Multiple Indicator Cluster Surveys (MICS), implemented with the assistance of UNICEF, 17 data sets from Demographic and Health Surveys (DHS), 17 data sets from national Labour Force Surveys (LFS) or other national household surveys; and finally, 30 data sets limited to children aged 15–17 years derived from national LFS conducted under Eurostat regulations. The estimation procedure consists of extrapolating the sample of harmonized national data sets to regional and global values by weighting each country according to its relative share of the total number of children in the region. The weighting factors are calculated for

2.2 Global Estimates

9

each sex and age group separately, and are calibrated to conform to the benchmark UN population estimates and projections for 2016. According to the estimates released by the ILO (2017) report on child labour, globally in 2016, there were 152 million children—64 million girls and 88 million boys—are in child labour, accounting for almost one-tenth of all children worldwide. Of these, nearly 73 million children in absolute terms were in hazardous work that directly endangered their health, safety and moral development. Children in employment, a broader measure comprising both child labour and permitted forms of employment involving children of legal working age, number 218 million. Second, the report indicated that the incidence of child labour declined during the period from 2012 to 2016, continuing a trend seen since the publication of the ILO’s first global estimates of child labour in 2000. The 16-year period since 2000 saw a net reduction of 94 million in children in child labour wherein the number of children in hazardous work fell by more than half over the same period. But progress slowed during 2012–2016. A narrower focus on the most recent 4-year period indicates a significant slowing down of progress. The reduction in the number of children in child labour amounted to 16 million for the 2012–2016 period, just one-third of the 47 million reduction recorded from 2008 to 2012. Expressed in relative terms, the share of children in child labour fell by only one percentage point from 2012 to 2016 compared to three percentage points in the previous 4-year period. The decline in hazardous work slowed in a similar fashion. Regarding the regional spread of child labour use across the globe, the said ILO report (2017) indicates that the Africa region and the Asia and the Pacific region together hosted nine out of every ten children in child labour. Africa ranks highest both in the percentage of children in child labour—one-fifth—and the absolute number of children in child labour—72 million, followed by Asia and the Pacific region which ranks second highest in both these measures—7% of all children, 62 million in absolute terms, are in child labour in this region. The remaining child labour population is divided among the Americas (11 million), Europe and Central Asia (6 million), and the Arab States (1 million). This means that to achieve the 2025 target set by the SDG of United nations, it is not sufficient to step up efforts in the African region only, one has to look for initiatives to curb the incidence of child labour also in other regions of Asia, Europe, America and even in oil-rich Saudi Arabia. In fact, the 2016 estimates suggest that sub-Saharan Africa, the regional grouping for which we have comparable estimates for 2012, witnessed a rise in child labour during the 2012–2016 period, in contrast to the other major regions, due primarily to broader economic and demographic forces, including armed conflict, drug trafficking, etc., which tend to thwart a governmental efforts against child labour use. This calls for the need for prioritizing war against endemic child labour so that governments, workers’ and employers’ organizations, and civil society organs could all play a critical role in this context. The agricultural sector of the world accounts for 71% of all in child labour, and for 108 million children in absolute terms—they are primarily engaged in subsistence and commercial farming and livestock herding. Most child labour takes place within the family unit. More than two-thirds of all children in child labour work

10

2 Incidence of Child Labour—Global and Indian Estimates

as contributing family labourers, while paid employment and own-account workers make up 27% and 4%, respectively, of those in child labour. It is important to note that most children in child labour are not in an employment relationship with a third-party employer, but rather work on family farms and in family enterprises. According to the 2016 Global Estimates of Modern Slavery, there were about 4.3 million children aged below 18 years in forced labour, representing 18% of the 24.8 million total forced labour victims worldwide. This estimate includes 1.0 million children in commercial sexual exploitation, 3.0 million children in forced labour for other forms of labour exploitation and 300,000 children in forced labour imposed by state authorities. Children aged 5–11 years form the largest share of those in child labour and also form a substantial share of those in hazardous work—48% of all those in child labour are in the 5–11 years age bracket, 28% are aged between 12 and 14 years, and 25% fall into the 15–17 years age range. About one-fourth of all children in the hazardous workgroup—19 million in absolute terms—are aged between 5 and 11 years. It should be recalled that 15–17-year olds are above the minimum working age and therefore are not counted as child labourers because they are too young; rather, they are included in child labour because their work is or may be physically or psychologically injurious to their health and well-being. They are also more likely than other employed 15–17-year olds to have dropped out of school prematurely. Boys appear to face a greater risk of child labour than girls. The difference in child labour incidence is less than one percentage point for 5–11-year olds, rising to three percentage points for 12–14-year olds and to five percentage points for 15–17-year olds. It may be noted that the decline in child labour among girls was only half that among boys during the 2012–2016 period, meaning that the gender gap in child labour has narrowed. For children engaged in household chores, girls far outnumber the boys—and this feature is observed across all regions. The real damage of the prevalence of child labour or children at work in such a large number is their exclusion from educational access. The estimates indicate that a very large number of children in child labour are completely deprived of education, around 32% or 36 million of such children in child labour, are out of school. Those who are able to attend school in addition to working, are also ‘penalized in the sense the time and energy required by work interfere with children’s ability to derive educational benefit from their time in the classroom and to find time outside the classroom for independent study’ (ILO 2017). Table 2.1 below summarizes the global estimates of child labour. There has been uneven progress against the child labour use across the globe. The first 4-year interval, 2000–2004, saw substantial strides forward, leading to the optimistic conclusion that the end of child labour was perhaps ‘within reach’. But this optimism weaned considerably by the results of the next, 2004–2008 interval, which pointed to a marked slowing down of the progress. The penultimate 4-year interval, 2008–2012, showed the largest decline up to that point in the numbers of child labour, even though the period coincided with a deep global economic recession. The progress once more slowed during 2012–2016, pushing the timeline for ending child labour further into the future. Thus, the experience over the 16-year period has

2.2 Global Estimates

11

Table 2.1 Global estimates of child labour, 2016

2016

2012

2016

2012

2016

2,64,427

218,019

167,956

151,622

85,344

72525

Prevalence (%)

16.7

13.8

10.6

9.6

5.4

4.6

Number (000s)

144,066

130,364

120,453

114,472

37,841

35,376

Prevalence (%)

11.8

10.6

9.9

9.3

3.1

2.9

Number (000s)

120,362

87,655

47,503

37,149

47,503

37,149

Prevalence (%)

33.0

24.9

13.0

10.5

13.0

10.5

Number (000s)

148,327

123,190

99,766

87,521

55,048

44,774

Prevalence (%)

18.1

15.0

12.2

10.7

6.7

5.5

Number (000s)

116,100

94,829

68,190

64,100

30,296

27,751

Prevalence (%)

15.2

12.4

8.9

8.4

4.0

3.6

Number (000s)

–

99,417

–

72,113

–

Prevalence (%)

–

27.1

–

19.6

–

Number (000s)

–

17,725

–

10,735

–

Prevalence (%)

–

8.8

–

5.3

–

Number (000s)

129,358

90,236

77,723

62,077

33,860

28,469

Prevalence (%)

15.5

10.7

9.3

7.4

4.1

3.4

Europe and Central Asia

Number (000s)

–

8773

–

5534

–

5349

Prevalence (%)

–

6.5

–

4.1

–

4.0

Arab States

Number (000s)

–

1868

–

1162

–

616

Prevalence (%)

–

4.6

–

2.9

–

1.5

5–14 years

Male

Female

Region (5–17 years)

Children in hazardous work

2012

15–17 years

Sex (5–17 years)

Children in child labour

Number (000s)

World (5–17 years)

Age range

Children in employment

Africa

America

Asia and the Pacific

Source ILO (2017) Global Estimates of Child Labour: Results and Trends, 2012–2016, Geneva 2017

12

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.2 Global estimates of child population for different age groups, 2000–2012 Age group (Years)

Child population

5–17 Number (‘000)

1,531,400 1,566,300 1,586,288 1,585,566 2.28

1.28

−0.05

5–14 Number (‘000)

1,199,400 1,206,500 1,216,854 1,221,071 0.59

0.86

0.35

15–17 Number (‘000)

332,000

−13.88

17.79

2000

2004

359,300

% Change 2008

309,433

2012

364,495

2004–2000 2008–2004 2012–2008

8.22

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12

not been one of quickening, or even steady, progress. There is a need for further quickening the rate of progress of reduction in the incidence of child labour to reach the target of zero incidence of child labour by 2025. When we look at the earlier ILO reports, we find that those child populations in different age groups have undergone changes between 2000 and 2012 (see Table 2.2). The child population data has been categorically divided into three groups—(i) 5–17; (ii) 5–14; (iii) 15–17. The general trend observed overtime mostly shows an increasing trend in the child population. However, child population growth for the age group (5–17) shows increasing trend though at decreasing rate for (2000–2008). However, for the period (2008–2012) the trend is however decreasing. The increasing trend is observed in the first two periods, i.e. from (2000–2008) for the (5–14) age group and the trend is just reversed for 2008–2012. A mixed trend is also found for the (15–17) age group. Even though the child population was increasing initially, but the period (2004–2008) reveals a negative growth. Surprisingly, the child population grows at a very high rate from (2008–2012) period. Table 2.3 presents the global estimates of employed children for different age groups during 2000–2012. We find that there has been a gradual reduction in the percentage of employed children for all categories and for all the years. An exception has been observed for (15–17) aged children in the period (2008–2012). The trend is increasing thereafter. An impressive picture has been found out for the child labourers in different age groups from Table 2.4. There has been a gradual decrease in the number of child labourers. Even though the child population is increasing but fewer child labourers are now employed in the job market. This is true for all age groups. This may be the response of different policy initiatives taken in favour of the curtailment of child labour. For all age groups, there has been a decline in the numbers of child labourers. In 12 years (2000–2012), highest fall in child labourer has been observed for (5–17) age group and the lowest fall has been noted for (15–17) age group. Table 2.5 gives estimates of children in hazardous work for different age groups for the period, 2000–2012.

2.2 Global Estimates

13

Table 2.3 Global estimates of employed children for different age groups, 2000–2012 Age group (years)

Children in employment 2000

2004

2008

2012

% Change 2004–2000 2008–2004 2012–2008

5–17 Number (‘000)

351,900 (23.0)

322,729 (20.5)

306,669 (19.3)

264,427 (16.7)

−2.5

−1.2

−2.6

5–14 Number (‘000)

211,000 (17.6)

196,047 (16.2)

176,452 (14.5)

144,066 (11.8)

−1.4

−1.7

−2.7

15–17 Number (‘000)

140,900 (42.4)

126,682 (35.2)

129,217 (32.0)

120,362 (33.0)

−7.2

−3.2

1

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12 Table 2.4 Global estimates of child labourers for different age groups, 2000–2012 Age group (years)

Child labourers

% Change

2000

2004

2008

2012

2004–2000 2008–2004 2012–2008

5–17 Number (‘000)

245,500 (16.0)

222,294 (14.2)

215,269 (13.6)

167,956 (10.6)

−1.8

−0.6

−3

5–14 Number (‘000)

186,300 (15.5)

170,383 (14.1)

152,850 (12.6)

120,453 (9.9)

−1.4

−1.5

−2.7

15–17 Number (‘000)

59,200 (17.8)

51,911 (17.8)

624,19 (16.9)

47,503 (13.0)

0

−0.9

−3.9

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12 Table 2.5 Global estimates of children in hazardous work for different age groups, 2000–2012 Age group (Years)

Children in hazardous work 2000

2004

2008

2012

% Change 2004–2000 2008–2004 2012–2008

5–17 Number (‘000)

170,500 (11.1)

128,381 (8.2)

115,314 (7.3)

85,344 (5.4)

−2.9

−0.9

−1.9

5–14 Number (‘000)

111,300 (9.3)

76,470 (6.3)

52,895 (4.3)

37,841 (3.1)

−3

−2

−1.2

15–17 Number (‘000)

59,200 (17.8)

51,911 (14.4)

62,419 (16.9)

47,503 (13.0)

−3.4

2.5

−3.9

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12

14

2 Incidence of Child Labour—Global and Indian Estimates

It is interesting to note that the exploitation of child labourers in hazardous work has been gradually reducing over the period. For all age groups, we find a declining trend in children engaged in hazardous work. One exception has been found in 2008 for the age group 15–17 whereas compared to 2004, more children are recruited in a hazardous activity. The ILO report cited above, identifies a number of actions that have driven progress, including the political commitment of governments, the increasing number of ratifications of the ILO Convention No. 182 on the worst forms of child labour and the parallel. The surge of the ILO Convention No. 138 on the minimum age for admission to employment, the two principal legal pillars for the global fight against child labour, sound policy choices, as well as solid legislative frameworks. The decline in child labour was greatest during the 4-year period (2008–2012) when the number of child labour decreased by 47 million from 215 to 168 million. This has happened despite the social hardship caused by the global economic crisis of 2008–09 and its aftermath. There are two explanations. First, while developing economies were not spared by the crisis, they have generally been quicker to rebound from its effects, although often on a lower growth path. Second, economic growth in the aftermath of the crisis has dampened labour demand. But another key reason for this decline is that attention to child labour by states, employers and workers and other key stakeholders has borne some fruits. In Tables 2.6a, 2.6b and 2.6c, we present the estimates of child population by sex, employed chidren by sex and of child labourers by sex for the period 2000–2012. It may be noted that even though the world number of both boys and girls child population has increased over time but the number of boys always exceed that of girls. Second, all the increase in the child population as has not been channelized to the job market. This is clear from the figure of employed children. Less number of both boys and girls are employed globally. Third, over time we find a reduced number of both male and female child labourers. A little increase in the recruitment of boys has been observed in 2008 as compared to the previous period 2004. In all other cases, we get a downward trend. Fourth, the girls always lag behind the boys in participation in the job market, and this is true even in case of hazardous work. The engagement of both boys and girls children in hazardous work is declining continuously. Most of the observed decline in child labour is in the number of girls. There were 15% (and 15 million) fewer female child labourers in 2008. The number of girls in hazardous work has declined altogether by 24%. For boys, on the other hand, an increase has been noted, both in terms of incidence rates and in absolute numbers. There were 7% more male child labourers in 2008 than 4 years earlier, i.e. 8 million more. The extent of hazardous work among boys remained relatively stable, however (74 million in total). Further, the difference between boys and girls in child labour employment becomes more pronounced with increasing age and danger faced in the workplace. For instance, among children in hazardous work in the 15–17 age group, boys outnumber girls by two to one. But it should be noted that these figures might underestimate girl’s involvement in child labour relative to that of boys as they do not reflect involvement in household chores, particularly hazardous chores, a dimension of child labour that is not included in the global estimates. Other

2.2 Global Estimates

15

Table 2.6a Global estimates of child population by sex Age group (years)

Child population

% Change

2004

2008

2012

2008–2004

2012–2008

World number (‘000)

1,566,300

1586,288

1,585,566

1.28

−0.05

Boys number (‘000)

804,000

819,891

819,877

1.98

0.001

Girls number (‘000)

762,300

766,397

766,690

0.53

0.04

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12 Table 2.6b Global estimates of employed childern by sex Age group (years)

Children in employment

% Change

2004

2008

2012

2008–2004

2012–2008

World number (‘000)

322,729 (20.6)

305,669 (19.3)

264,427 (16.7)

−1.3

−2.6

Boys number (‘000)

171,150 (21.3)

175,777 (21.4)

148,327 (18.1)

0.1

−3.3

Girls number (‘000)

151,579 (19.9)

129,892 (16.9)

116,100 (15.2)

−3

−1.7

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12 Table 2.6c Global estimates of child labourers by sex Age group (years)

Child labourers

% Change

2004

2008

2012

2008–2004

2012–2008

World number (‘000)

222,294 (14.2)

215,294 (13.6)

167,956 (10.6)

−0.6

−3

Boys number (‘000)

119,575 (14.9)

127,761 (15.6)

99,766 (12.2)

Girls number (‘000)

102,720 (13.5)

87,508 (11.4)

68,190 (8.9)

0.7

−3.4

−2.1

−2.5

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12

information available to the ILO indicates that girls may also be more present in less visible and therefore underreported forms of child labour such as domestic work in private households. Regarding regional spread of the child population, we find that in the age group of 5–14 years, Asia and the Pacific region is home to more than 60% of child population, and Latin America and the Caribbean region has least child labour among the world. An increasing trend in child population has been observed for Asia and the Pacific, Latin America and the Caribbean and for sub-Saharan Africa (Table 2.7). Regarding the employment opportunity of children, we find a mixed trend. For Asia and the Pacific, as well as for Latin America and the Caribbean even though the number of employed children was initially showing a decreasing trend, sub-Saharan Africa shows consistently an increasing trend pattern, whereas for other regions the trend

16

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.7 Global trends in child population by region (2000–2012) (5–14 age group) Region

Child population (‘000)

% Change

2000

2004

2008

2012

2004–2000 2008–2004 2012–2008

Asia and the Pacific

655,100

650,000

651,815

835,334

−0.78

0.28

28.16

Latin America and the Caribbean

108,100

111,000

110,566

142,693

2.68

−0.39

29.06

SubSaharan Africa

166,800

186,800

205,319

275,397

11.99

9.91

34.13

Other regions

269,300

258,800

249,154

110,411

−3.89

−3.73

−55.69

World

1,199,300 1,206,500 1,216,854 1,363,835

0.60

0.86

12.08

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12 Table 2.8 Global trends of child in employment by region (2000–2012) (5–14 age group) Region

Child in employment (‘000)

% Change

2000

2004

2008

2012

2004–2000 2008–2004 2012–2008

Asia and the Pacific

127,300

122,300

96,397

129,358

−3.93

−21.18

34.19

Latin America and the Caribbean

17,400

11,047

10,002

17,843

−36.51

−9.46

78.39

SubSaharan Africa

48,000

49,300

58,212

83,570

2.71

18.08

43.56

Other regions

18,300

13,400

10,700

13,307

−26.78

−20.15

24.36

World

211,000

196,047

176,452

244,078

−7.09

−9.99

38.33

Source ILO Report: Marking Progress against child labour, Global estimates and trends 2000–12

is again mixed in nature (Table 2.8). The picture changed in subsequent years (see Table 2.1).

2.3 Estimates of Child Labour in India India is home to the largest number of the world’s working children. In India, every third child is a working child and every fourth child in the age group 5–15 is employed

2.3 Estimates of Child Labour in India Table 2.9 Working children in India (5–14 years) (census)

17

Year

No. of children (in millions)

No. of child workers (in millions)

Proportion of working children (%)

1961

113.09

14.5

12.07

1971

150.07

10.8

7.01

1981

179.05

13.6

6.01

1991

209.09

11.3

5.37

2001

253.16

12.7

5.0

2011

259.64

4.35

1.68

Source Census of India (1961, 1971, 1981, 1991, 2001, 2011)

(ILO 2007). Thus a considerable number of children are not only losing their childhood but also opportunities for education. Census data on child labour according to main and marginal workers have been presented in Table 2.9, which shows that there were 14.5 million child workers in 1961 which declined to 10.8 million in 1971 then increased to 13.6 million in 1981. The number again fell to 11. 3 million in 1991 and then rose to 12.7 million in 2001. Notwithstanding fluctuations in absolute numbers, the proportion of working children in the workforce has declined over the decades. The data given in Table 2.1 shows a continuous declining trend in the child labour participation rate. The child population in India has increased from 113.09 million in 1961 to 253.16 million in 2001, in the meantime the proportion of working children has declined from 12 to 5% that is a fall of 7 percentage point. The table shows that over time an increasing trend has been observed in the number of children. This is consistent with the increasing population size in India. However, looking to the number of child workers a mixed trend is observed. Even though there has been a reduction in the number of child workers in 50 years but the trend is not always unidirectional. The year 1971 marked improvement over 1961 and 1991 over 1981. A reverse trend has been observed for the years 1981 and 2001. However, when we look at the figure for the proportion of working children, we find a declining trend over the period. Even though more a number of children are joining the job market, but their proportionate share is reducing in view of increasing population trend. When we look to the statewide distribution of working children according to 1991, 2001 and 2011 census in the age groups 5–14 years (Table 2.10), we find a mixed picture, though most of the states experiencing a diminution of the incidence of child labour. The trend on the magnitude of child labour is not uniform across the country, as per the census data. Three kinds of trends have been observed: (i) Fifteen states reveal throughout declining trend. The states are (a) Andhra Pradesh, (b) Gujarat, (c) Jammu and Kashmir, (d) Karnataka, (e) Kerala, (f) Madhya Pradesh, (g) Maharashtra, (h) Chhattisgarh, (i) Jharkhand, (j) Uttaranchal, (k) Orissa, (l) Tamil Nadu, (m) Tripura, (n) Dadra and Nagar Haveli, (o) Puducherry; (ii) Three states which initially (1991–2001) exhibited a declining trend, reversed the pattern in the later period

18

2 Incidence of Child Labour—Global and Indian Estimates

(2001–2011). The states are (a) Daman and Diu, (b) Goa, (c) Lakshadweep; (iii) Seventeen states initially (1991–2001) exhibited an increasing trend which reversed in the later period (2001–2011). The states are (a) Assam, (b) Bihar, (c) Haryana, (d) Himachal Pradesh, (e) Manipur, (f) Meghalaya, (g) Nagaland, (h) Punjab, (i) Rajasthan, (j) Sikkim, (k) Uttar Pradesh, (l) West Bengal, (m) Andaman and Nicobar Islands, (n) Arunachal Pradesh (o) Chandigarh, (p) Delhi, (q) Mizoram. Further, it may be noted that: (1) In 1991, the incidence of child labour was highest in Andhra Pradesh and it was lowest in Lakshadweep; (2) In 2001, incidence of child labour was highest in Uttar Pradesh and it was lowest in Lakshadweep; (3) In 2011, the highest and lowest rank holding states were the same as that for 2001; (4) Magnitude of child labour is comparatively higher in (a) Andhra Pradesh, (b) Madhya Pradesh, (c) Maharashtra, (d) Uttar Pradesh; (5) Magnitude of child labour is comparatively lower in (a) Sikkim, (b) Tripura, (c) Andaman and Nicobar Islands, (d) Dadra and Nagar Haveli, (e) Daman and Diu, (f) Lakshadweep, and (g) Puducherry. We now present the changes in the work participation rate of child labour between 1991 and 2004 (Table 2.11), state-wise Labour Force Participation Rate: 2011–12

Table 2.10 State-wise distribution of working children according to 1991, 2001, and 2011 census in the age group 5–14 years Sl. No.

Name of the state/UT

1991

2001

2011

1

Andhra Pradesh

1,661,940

1,363,339

404,851

2

Assam

327,598

351,416

99,512

3

Bihar

942,245

1,117,500

451,590

4

Gujarat

523,585

485,530

250,318

5

Haryana

109,691

253,491

53,492

6

Himachal Pradesh

56,438

107,774

15,001

7

Jammu and Kashmir

–

175,630

25,528

8

Karnataka

976,247

822,615

249,432

9

Kerala

34,800

26,156

21,757

10

Madhya Pradesh

1,352,563

1,065,259

286,310

11

Maharashtra

1,068,427

764,075

496,916

12

Chhattisgarh

–

364,572

63,884

13

Manipur

16,493

28,836

11,805

14

Meghalaya

34,633

53,940

18,839

15

Jharkhand

–

407,200

90,996

16

Uttaranchal

–

70,183

28,098

17

Nagaland

16,467

45,874

11,062

18

Orissa

452,394

377,594

92,087

19

Punjab

142,868

177,268

90,353

20

Rajasthan

774,199

1,262,570

252,338 (continued)

2.3 Estimates of Child Labour in India

19

Table 2.10 (continued) Sl. No.

Name of the state/UT

1991

2001

21

Sikkim

5,598

16,457

2011 2704

22

Tamil Nadu

578,889

418,801

151,437

23

Tripura

16,478

21,756

4998

24

Uttar Pradesh

1,410,086

1,927,997

896,301

25

West Bengal

711,691

857,087

234,275

26

Andaman and Nicobar Island

1265

1960

999

27

Arunachal Pradesh

12395

18,482

5766

28.

Chandigarh

1870

3779

3135

29

Dadra and Nagar Haveli

4416

4274

1054

30

Delhi

27,351

41,899

26,473

31

Daman and Diu

941

729

774

32

Goa

4656

4138

6920

33

Lakshadweep

34

27

28

34

Mizoram

16,411

26,265

2793

35

Pududcherryy

2680

1904

1421

Total

11,285,349

12,666,377

4,353,247

Source Census of India (1991, 2001, 2011)

(Rural) (Table 2.12a), and state-wise Labour Force Participation Rate: 2011–12 (Urban) (Table 2.12b). From Table 2.11, we observe the following: (i) Some states reveal throughout the declining trend. The states are: (a) Andhra Pradesh, (b) Assam, (c) Gujarat, (d) Jharkhand, (e) Karnataka, (f) Kerala, (g) Madhya Pradesh, (h) Maharashtra; (ii) One state, Goa initially exhibited a declining trend which reversed in the later period; (iii) Six states initially exhibited an increasing trend which reversed in the later period. The states are: (a) Bihar, (b) Delhi, (c) Himachal Pradesh, (d) Orissa, (e) Punjab, (f) Rajasthan, (g) Haryana; and (iv) Work participation rates of child labour always remain very high in Andhra Pradesh. It occupies the highest position both in 1991 and 2004. It was ranked third in 2001 while Rajasthan ranked first. From Tables 2.12a and 2.12b, we observe the following pattern: (i) Himachal Pradesh occupies the first position in child workforce participation rate for both rural and urban India; (ii) Andhra Pradesh, Chhattisgarh, Gujarat, Karnataka, Maharashtra are some of the states with high incidence of child labour in both rural and urban areas; (iii) Bihar is characterized by lowest incidence of child labour in both rural and urban areas. The incidence of child labour is also low in Uttar Pradesh; (iv) Child labour is more acute in rural areas for all the states as compared to the urban areas.; (v) Number of male child labour always dominates their female counterparts. When we look into the changes in the percentage share of child labour between 1991 and 2010 (Table 2.13) in different states of India, and their gender-wise dis-

20

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.11 Changes in the work participation rate of child labour between 1991 and 2004 Sl. No.

State

Work participation rate 1991

2001

2004

1

Andhra Pradesh

9.98

7.7

6.61

2

Arunachal Pradesh

5.65

6.06

–

3

Assam

5.46

5.07

1.82

4

Bihar

3.99

4.68

1.36

5

Chhattisgarh

–

6.96

4.58

6

Delhi

1.27

1.35

0.26

7

Goa

1.95

1.82

2.70

8

Gujarat

5.26

4.28

2.53

9

Haryana

2.55

4.78

1.71

10

Himachal Pradesh

4.55

8.14

2.73

11

Jammu and Kashmir

–

6.62

–

12

Jharkhand

–

5.47

2.48

13

Karnataka

8.81

6.91

4.66

14

Kerala

0.58

0.47

0.20

15

Madhya Pradesh

8.08

6.71

2.82

16

Maharashtra

5.73

3.54

3.42

17

Orissa

5.87

4.37

4.87

18

Punjab

3.04

3.23

1.73

19

Rajasthan

6.46

8.25

4.86

Source Compiled from Census of India 1991 and 2001 and NSS Unit Level Records of 2004–05 Table 2.12a State-wise labour force participation rate among children: 2011–12 (rural) Sl. No.

State

Work participation rate (rural) Female

Male

Total

1

Andhra Pradesh

44.8

61.2

52.8

2

Arunachal Pradesh

28.2

49.2

38.9

3

Assam

12.9

56.4

35.9

4

Bihar

5.8

48.7

28.4

5

Chhattisgarh

41.6

56.3

49.0

6

Delhi

14.6

54.4

37.1

7

Goa

21.2

58.6

39.9

8

Gujarat

27.9

60.2

44.8

9

Haryana

16.4

53.2

36.5

10

Himachal Pradesh

52.9

54.7

53.8

11

Jammu and Kashmir

26.3

55.9

41.5 (continued)

2.3 Estimates of Child Labour in India

21

Table 2.12a (continued) Sl. No.

State

Work participation rate (rural) Female

Male

Total

12

Jharkhand

20.4

54.2

37.8

13

Karnataka

28.9

62.0

45.4

14

Kerala

25.8

58.3

41.0

15

Madhya Pradesh

23.9

56.4

40.7

16

Maharashtra

38.9

58.2

49

17

Orissa

25.1

60.6

42.7

18

Punjab

23.7

57.9

41.4

19

Rajasthan

34.90

50.0

42.7

Source National Sample Survey Office, 68th Round, July 2011–June 2012 Notes 1. Figures are based on usual status approach and include principal status and subsidiary status persons of all ages 2. The figures represent the size of labour force as a percentage of the population Table 2.12b State-wise labour force participation rate among children: 2011–12 (urban) Sl. No.

State

Work participation rate (urban) Female

Male

Total

1

Andhra Pradesh

18.0

57.6

38.0

2

Arunachal Pradesh

13.9

47.5

31.8

3

Assam

9.7

57.3

34.8

4

Bihar

5.4

44.1

26.7

5

Chhattisgarh

25.2

51.7

39.3

6

Delhi

10.9

54.8

34.9

7

Goa

17.4

52.6

35.3

8

Gujarat

13.5

60.7

38.7

9

Haryana

10.2

53.5

33.1

10

Himachal Pradesh

23.6

61.2

43.3

11

Jammu and Kashmir

14.5

56.3

36.2

12

Jharkhand

7.3

50.3

30.0

13

Karnataka

17.1

59.4

38.8

14

Kerala

22.2

56.7

38.6

15

Madhya Pradesh

11.9

53.3

33.4

16

Maharashtra

17.2

56.0

37.4

17

Orissa

15.8

60.3

39.5 (continued)

22

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.12b (continued) Sl. No.

State

Work participation rate (urban) Female

Male

Total

18

Punjab

14.1

58.6

37.9

19

Rajasthan

14.4

50.7

33.6

Source National Sample Survey Office, 68th Round, July 2011–June 2012 Notes 1. Figures are based on usual status approach and include principal status and subsidiary status persons of all ages 2. The figures represent the size of labour force as a percentage of the population

tribution between 2004 and 2009–10, rural (Table 2.14a), and urban (2.14b), we observe the following interesting pattern: The change in percentage share of child labour was highest in Andhra Pradesh for all the three successive time periods. In Bihar, Karnataka, Madhya Pradesh, Maharashtra and Rajasthan the share is quite high (above 6%). In Gujarat and Orissa, moderate share is reflected (above 4%). In the other states, a very low percentage share of child labour is exhibited. Arunachal Pradesh, Goa and Kerala are the states to be ranked in the lowest category in this aspect.

Table 2.13 Changes in the percentage share of child labour between 1991 and 2010 Sl. No.

State

% Share of child labour in the states 1991

2001

2004

2010

1

Andhra Pradesh

14.7

10.8

13.2

4.71

2

Arunachal Pradesh

0.1

0.1

–

–

3

Assam

2.9

2.8

1.5

3.80

4

Bihar

8.3

8.8

4.0

5.55

5

Chhattisgarh

0.0

2.9

2.9

0.23

6

Delhi

0.2

0.3

0.1

0.37

7

Goa

0.0

0.0

0.1

–

8

Gujarat

4.6

3.8

3.3

7.84

9

Haryana

1.0

2.0

1.1

1.45

10

Himachal Pradesh

0.5

0.9

–

0.15

11

Jammu and Kashmir

0.0

1.4

–

0.59

12

Jharkhand

0.0

3.2

2.3

1.65

13

Karnataka

8.7

6.5

6.3

4.54

14

Kerala

15

Madhya Pradesh

16 17

0.3

0.2

0.1

0.06

12.0

8.4

5.4

3.83

Maharashtra

9.5

6.0

8.6

5.23

Orissa

4.0

3.0

4.8

2.70 (continued)

2.3 Estimates of Child Labour in India

23

Table 2.13 (continued) Sl. No.

State

% Share of child labour in the states 1991

2001

2004

2010

18

Punjab

1.3

1.4

1.1

0.98

19

Rajasthan

6.9

10.0

9.0

8.14

Source Compiled from Census of India 1991 and 2001 and NSS Unit Level Records of 2009–10 Table 2.14a Gender-wise child workforce participation rates in different states of rural India from 2004–05 to 2009–10 Sl. No.

State

(2004–05) (per ‘000 of total worker)

(2009–10) (per ‘000 of total worker)

Male

Male

Female

Female

1

Andhra Pradesh

39.78

69.97

88.16

110.91

2

Assam

27.92

24.19

144.66

31.91

3

Bihar

21.06

23.85

224.29

38.66

4

Chhattisgarh

27.07

48.30

3.67

7.32

5

Gujarat

20.36

24.87

150.49

207.97

6

Haryana

11.50

3.97

226.64

17.47

7

Himachal Pradesh

6.43

14.19

2.30

2.94

8

Jharkhand

21.42

50.88

63.68

14.66

9

Karnataka

31.38

47.91

89.79

113.43

10

Kerala

1.07

0.00

1.18

0.00

11

Madhya Pradesh

21.29

57.50

91.45

32.81

12

Maharashtra

24.19

35.44

66.37

127.99

13

Orissa

33.33

64.23

54.39

38.29

14

Punjab

17.58

35.00

16.80

6.43

15

Rajasthan

33.96

98.03

93.06

261.87

16

Tamil Nadu

7.48

14.74

0.00

13.88

17

Uttaranchal

28.58

7.02

14.81

7.24

18

Uttar Pradesh

33.29

50.43

1012.29

546.32

19

West Bengal

25.76

69.04

357.27

433.45

739.56

All India

Source Derived from respective unit level records of NSS

2701.3

134.66 1714.76

24

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.14b Gender-wise child workforce participation rates in different states of urban India from 2004–05 to 2009–10 Sl. No.

State

1

Andhra Pradesh

2 3

(2004–05) (per ‘000 of total workers)

(2009–10) (per ‘000 of total workers)

Male

Male

Female

Female

22.23

48.59

20.77

15.55

Assam

5.16

55.43

11.83

0.76

Bihar

23.44

49.11

11.02

2.55

4

Chhattisgarh

26.70

21.16

0.64

0.00

5

Gujarat

12.76

28.18

15.95

16.28

6

Haryana

8.69

5.07

28.07

3.99

7

Himachal Pradesh

0.00

3.78

2.16

0.00

8

Jharkhand

19.52

52.57

4.12

0.00

9

Karnataka

6.55

14.87

20.79

2.48

10

Kerala

11

Madhya Pradesh

12

Maharashtra

13 14 15

Rajasthan

16 17 18

Uttar Pradesh

19

West Bengal

1.14

4.95

0.00

1.58

13.59

32.17

57.69

9.06

7.34

6.97

54.23

12.08

Orissa

15.83

36.29

36.52

5.36

Punjab

8.45

0.00

15.66

9.94

38.13

56.46

43.18

7.83

Tamil Nadu

6.93

20.52

3.47

0.00

Uttaranchal

3.85

4.23

3.22

2.10

39.78

132.69

147.82

68.89

All India

21.40

76.52

31.95

27.72

281.49

649.56

509.09

186.17

Source Derived from Respective Unit Level Records of NSSO

1. That there are discrepancies between the work participation rates of male and female child labourers is well known—there are certain jobs from which female workers are excluded. Domination of male child labours are observed for overall India as well as for urban India. During 2004–05, in Andhra Pradesh, Bihar, Chhattisgarh, Gujarat, Himachal Pradesh, Jharkhand, Karnataka, Madhya Pradesh, Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh, West Bengal female child labour incidence was more prominent. In the other states like Assam, Haryana, Kerala, Uttaranchal the trend was reversed. In all India, the former trend followed. During 2009–10, the trend of female labour domination continued in Andhra Pradesh, Chhattisgarh, Gujarat, Himachal Pradesh, Karnataka, Maharashtra, Rajasthan, Tamil Nadu is continued. However, Bihar, Jharkhand, Madhya Pradesh, Orissa, Punjab, Uttar Pradesh, West Bengal revealed an opposite trend in 2009–10 as compared to 2004–05. Over time, however, no change was observed in the states where male child labour was

2.3 Estimates of Child Labour in India

25

more actively engaged in the job market. This situation may be attributable to the implementation of various schemes taken in favour of women for their protection and empowerment enhancement. During 2004–05, in urban India, Chhattisgarh, Haryana, Madhya Pradesh, Punjab are the only states where male workforce participation in the job market was dominating. In 2009–10, Gujarat is the only state where female workers participated in a slight higher than their male counterpart. A surprising result was observed for the other 18 states, where male child labour was much higher than female child labour. The number of such states was only four in 2004–05. For the overall rural–urban incidence of child labour in India during the period 2004–05 to 2009–10, we may turn to Table 2.15. Two observations are worth noting from this table: (i) In 2004–05, Assam, Bihar, Kerala, Tamil Nadu, Uttar Pradesh and West Bengal are the few states where rural incidence of child labour was more prominent than urban incidence; (ii) However, in 2009–10, except Kerala and Punjab all other states exhibited uniform trend.

Table 2.15 Rural–urban incidence of child labour in different states in India Sl. No.

State

Rural (2004–05) (per ‘000 of total workers)

Urban (2004–05) (per ‘000 of total workers)

Rural (2009–10) (per ‘000 of total workers)

Urban (2009–10) (per ‘000 of total workers)

1

Andhra Pradesh

109.75

70.82

199.07

36.32

2

Assam

52.11

60.59

176.57

12.59

3

Bihar

44.91

72.55

262.95

13.57

4

Chhattisgarh

75.37

47.86

10.99

0.64

5

Gujarat

45.23

40.94

358.46

32.23

6

Haryana

15.47

13.76

244.11

32.06

7

Himachal Pradesh

20.62

3.78

5.24

2.16

8

Jharkhand

72.3

72.09

78.34

4.12

9

Karnataka

79.29

21.42

203.22

23.27

10

Kerala

1.07

6.09

1.18

1.58

11

Madhya Pradesh

78.79

45.76

124.26

66.75

12

Maharashtra

59.63

14.31

194.36

66.31

13

Orissa

97.56

52.12

92.68

41.88

14

Punjab

52.58

8.45

23.23

25.6

15

Rajasthan

131.99

94.59

354.93

51.01

16

Tamil Nadu

22.22

27.45

13.88

3.47

17

Uttaranchal

35.6

8.08

22.05

5.32 (continued)

26

2 Incidence of Child Labour—Global and Indian Estimates

Table 2.15 (continued) Sl. No.

State

18

Uttar Pradesh

19

West Bengal

Rural (2004–05) (per ‘000 of total workers)

Urban (2004–05) (per ‘000 of total workers)

Rural (2009–10) (per ‘000 of total workers)

Urban (2009–10) (per ‘000 of total workers)

83.72

172.47

1558.61

216.71

94.8

97.92

491.93

59.67

Source Derived from Respective Unit Level Records of NSSO Table 2.16 Rural–urban incidence of child labour in India (2009–10) Sl. No.

State

Male (rural)

1

Andhra Pradesh

2

Arunachal Pradesh

–

3

Assam

144,655

31,909

11,833

757

4

Bihar

224,292

38,665

11,017

2548

5

Chhattisgarh

3669

7321

363

0

6

Delhi

18,576

0

7

Goa

–

–

–

–

8

Gujarat

150,487

207,973

15,945

16,282

9

Haryana

22,664

17,471

28,073

3988

10

Himachal Pradesh

2300

2942

2156

0

11

Jammu and Kashmir

11,274

16,872

1139

0

12

Jharkhand

63,684

14,661

4123

0

13

Karnataka

89,796

113,429

20,793

2479

14

Kerala

1182

0

0

1583

15

Madhya Pradesh

91,454

32,812

57,688

9063

16

Maharashtra

66,370

127,996

54,230

12,077

17

Orissa

54,390

38,288

36,522

5363

18

Punjab

16,802

6433

15,664

9937

19

Rajasthan

93,055

261,871

43,184

7826

88,156

–

Female (rural)

Male (urban)

Female (urban)

110,191

20,767

15,548

–

–

–

–

Source Annexure Referred to in Reply to Part (A) of Lok Sabha Starred Question No. 191 for Answer on 11.3.2013 Regarding Child Labour by Shri Ashok Tanwar and Shrimati Supriya Sule

For 2009–10, the state-wise distribution of rural–urban incidence of child labour also has a gender-wise classification. Table 2.16 summarizes. It is clear from Table 2.16 that no uniform trend is observed in male–female analysis. In some states, male child labour predominates the female counterpart

2.3 Estimates of Child Labour in India

27

whereas its reverse is also observed in some other states. In rural Bihar, participation of male child labour is highest whereas for Kerala it is lowest for both male and female. In Gujarat, however, participation of female child labour is highest for both rural and urban areas, while in urban Madhya Pradesh participation of male child labour is highest and for Kerala it is lowest.

2.4 Conclusion The above description of the estimates of Child Labour in the world and in India, reveals the following important information. First, despite a declining trend, the absolute size of the child labour use is still very high. The incidence of child labour is relatively large in Africa and the Asia Pacific region, although in other regions of the world the size has not diminished significantly. The slow progress in the decline of child labour in all parts of the world indicates that governments should adopt various policies-trade and non-trade—to accelerate the diminution of the use of child labour in order to achieve the UN deadline of 2025 to achieving the zero child labour target in the world so as to achieve the sustainable development goals. Secondly, in India, too, where the stock of child labour use in different states is very high, the progress towards elimination of child labour has been rather slow, and policy interventions of various kind are urgently required to accelerate the declining trend. This calls for orientation of production structure and redesigning the forces of the market in the economy through suitable reforms. The blind pursuit of policies like banning child labour may not suffice as there are a complex set of factors that govern the use of children in jobs rather than sending the children to school.

References Census of India. (1961, 1971, 1981, 1991, 2001 and 2011). Government of India. ILO. (1996). Economically active populations—Estimates and projections, 1956–2010. Geneva: ILO. ILO. (2006a). The end of child labour: Within reach, global report under the follow-up to the ILO declaration on fundamental principles and rights at work. In: Report of the Director-General, Report I (B), International Labour Conference, 95th Session. Geneva. ILO. (2006b). The international programme on the elimination of child labour (IPEC) what it is and what it does (p. 2006). Geneva: ILO. ILO. (2013a). World report on child labour: Economic vulnerability, social protection and the fight against child labour. Geneva: ILO. ILO. (2013b). Marking progress against child labour, global estimates and trends 2000–12. Geneva: ILO. ILO. (2017). Global estimates of child labour: Results and trends, 2012–2016. Geneva. International Labour Organization. (2006). Global child labour trends 2000–2004 (p. 2006). Geneva: ILO.

28

2 Incidence of Child Labour—Global and Indian Estimates

National Sample Survey Office. (2005). NSS unit level records of 2004–05. National Sample Survey Office. (2009). NSS unit level records of 2009–10. National Sample Survey Office. (2012). 68th Round Employment and Unemployment Survey Report, July 2011–June 2012. UNICEF. (2013). Impact of Unpaid Household Practices on the Measurement of Child Labour, MICS Methodological Papers no 2. October 2013.

Chapter 3

Types and Causes of Child Labour

Abstract The practice of sending children to work is prevalent widely throughout the globe, particularly in developing countries. Although there have been diminutions in the incidence of child labour over the recent periods, including in India, the size of child labour employment in various forms, continues to be quite large in such economies. The different types or forms that working children are found to be in are unique in the sense that in each category, they perform specific work and face specific problems. In this chapter, we describe different types and categories in which working children perform, and try to locate the proximate causes for the existence of child labour.

Notes: This chapter draws from and reuses with permission the materials contained in the following papers of the author(s): (a) Biswajit Chatterjee and Runa Ray (2016), “Impact of Trade Restriction on Child Labour Supply and the Role of Parents’ Utility Function: A Two Sector General Equilibrium Analysis” in Malabika Roy and Saikat Sinha Roy (eds.), International Trade and International Finance, Springer 2016. (b) Runa Ray (2015), “Trade versus Non-Trade Policy in a Two Sector General Equilibrium Framework”, Business and Economics Journal 2015, Volume 6 Issue 2, 2015, Business Economic Journal, an open access journal. (c) Runa Ray (2014), “Import Restrictions, Capital Accumulation and Use of Child Labour–A General Equilibrium Analysis”, Chapter 5 of the book edited by Ambar Nath Ghosh and Asim K. Karmark (eds.), Analytical Issues in Trade, Development and Finance, Essays in Honour of Biswajit Chatterjee, Springer India, 2014, as Part I of Chapter 6 of the present book; (d) Runa Ray and Biswajit Chatterjee (2010c), “Education Subsidy, Adult Unemployment and the Incidence of Child Labour in an Open Economy: A Three Sector General Equilibrium Analysis”, Vol 57, No 4, Jan–March 2010, The Indian Economic Journal, pages 37–59. Also used is (i) ILO Report “Global Estimates of Child Labour: Results and Trends, 2012–2016”, Published by International Labour Office (ILO), Geneva, 2017. (ii) www.alliance.org. © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_3

29

30

3 Types and Causes of Child Labour

3.1 Types of Child Labour According to the ILO Report (2017), more than two-thirds of all children in child labour work as contributing family labourers, while paid employment and ownaccount workers account for roughly 27% and 4%, respectively, of those in child labour. Most children in child labour are not in an employment relationship with a third-party employer, but rather work on family farms and in family enterprises. There are categories of children at work who are forced labour. As per the said report, we may distinguish between the following nomenclatures used to portray the status of child labour spread across the globe engaged in different types of occupations: • Children in employment are those working in any form of market production and certain types of non-market production (principally, e.g. the production of agricultural produce for own use). This group includes children in forms of work engaged in both the formal and informal economy; inside and outside family settings; for pay or profit (in cash or in kind, part time or full time); and domestic work outside the child’s own household for an employer (paid or unpaid). • Children in child labour is a narrower category than children in employment, as it excludes children in employment who are in permitted light work and those above the minimum age whose work is not classified as a worst form of child labour, or in particular, as ‘hazardous work’. • Children in the worst forms of child labour comprise: (a) all forms of slavery or practices similar to slavery, such as the sale and trafficking of children, debt bondage and serfdom and forced or compulsory labour, including forced or compulsory recruitment of children for use in armed conflict; (b) the use, procuring or offering of a child for prostitution, for the pornographic performances; (c) the use, procuring or offering of a child for illicit activities, like drug trafficking, etc; and (d) work, which by its nature or environment, may affect the health, safety and morals of children. The last category is often described as belonging to Children in hazardous work, which may include night work and long hours of work, exposure to physical, psychological or sexual abuse; work underground at dangerous heights or in confined spaces; and work in an unhealthy environment, etc. Hazardous work by children is often treated as a proxy category for the worst forms of child labour because primarily reliable national data on the worst forms of child labour other than hazardous work, such as children in bonded and forced labour or in commercial sexual exploitation, are still difficult to obtain and second, children in hazardous work account for the overwhelming majority of those in the worst forms of child labour. It is clear that the above categories of child labour are referred in official estimates and reports are overlapping in nature and clear-cut boundaries between each category seem impossible in view of the arbitrariness involved. There are also Children performing household Chores as referring to those performing domestic and

3.1 Types of Child Labour

31

personal services for consumption within their own households, and constitute a ‘non-economic’ form of production and are excluded from consideration in the UN System of National Accounts (UNSNA). The real problem of children at work is that in this case, the children are definitely excluded from school attendance. For those who are also engaged in light domestic work apart from performing their designated functions in the school, we are not including them in our discourse on child labour, as such categories do not pose any challenge to policymakers, and in size are quite thin compared to other categories. When the UNO calls for the eradication of half of the world stock of child labour by 2025 as a part of achieving the target of sustainable development, they exclude the abolition of this category, and we shall adhere to such norms in our discussions. Working children in each category have their own specific problems, and we discuss below these problems. (a) Children in wage-earning employment Working children are regarded as a source of income for the family (Grootaert and Kanbur 1995). Children in this category are the ones who are most easily exposed to the danger of exploitation and hazardous working conditions. However, the children are often employed in small-scale industries, petty commerce and domestic service to which labour legislation, if any, would not usually apply. Labour administration alone would have limited influence in this field and cooperation with other administration of social services is imperative. (b) Children bounded with artisans The same danger of exploitation is to be found with children with artisans’ family, as they are often bound by an advance paid to their parents and as they often live with their patrons, their duty often includes service for the family of their patron. The working conditions of these children labourers could be improved by their employers or patrons as a result of enforcement of legislation and administrative advice (Burra 1995), and they may be provided with adequate training by setting standards of work and training by organizations of employers. The reluctance of the employers to provide training to such workers is often guided by the fact that they might leave such jobs when they are better equipped with additional skills formed through training. (c) Children as ‘unpaid family workers’ They are supposed to form the majority of working children engaging mostly in agriculture. Their main deprivation seems to be lack of educational opportunities, not to speak of the lack of vocational preparation for employment of modern sectors. Provision to open school should be of prime importance. Informal educational facilities such as classes by cooperations or voluntary organizations could replace formal school education. Measures for children of this category would be most directly related to the development programmes, in general, where cooperation of various agencies would be necessary.

32

3 Types and Causes of Child Labour

(d) Children working and schooling1 Some children have a working and schooling life combined. Their school attendances are on a part-time basis because their working hours prevent them from full-time attendance. On the other hand, those children who work after full-time schooling are in danger of exhaustion because of the double burden. Ensuring schooling hours should come first. In order to ensure schooling hours and to limit the total amount of the burden, regulation of working hours would be inevitably required. Also, climate and practice for ensuring adequate schooling hours among employers and parents should be developed. It may be argued that the goal of the total abolition of child labour is beyond the immediate reach of most countries. Some might even despair that this objective of the total elimination of child labour is of limited use, because it is too long run in character. Yet, the goal of child labour abolition is itself a strong guiding principle in policy decisions. It can be argued that progress towards the reduction in the incidence of child labour can be made if there is a revival of sustained economic growth, the pattern of growth is more egalitarian, the modern sector is expanding rapidly, universal compulsory schooling of children and adult literacy programmes be launched, and implemented. Since the socio-economic policies will only bear results in the long term, there is a need for reinforcing these broad measures by additional action directly targeting working children as is being done, for instance, by the ILO’s International Programme on the Elimination of Child Labour (IPECL). Legislation and enforcement of child labour laws beyond the formal sector, where the vast majority of the working children are found, is often impractical and costly. Socio-economic policies, along with growth promoting and diversifying strategies may be geared towards reducing the incidence of child labour simply Despite their long term significance, it is clear that the socio-economic policies are a key solution to the problem of child labour, simply because of their significant economy-wide influences, even in the long run. It may be noted that a country’s full commitment to the total abolition of child labour should be judged not merely on the basis of its official pronouncements, but on whether the child labour objective is consciously considered in the above outlined social and economic policy framework. Practical actions and standard setting need be combined with direct economic policies which address the root causes of child labour, particularly in the developing countries. In this book, two alternative policies have been analysed—one is protectionist trade policy where the impact of increased protectionism in import competing sector has been analysed. Another is economic expansion. Child labour problem can be reduced by adopting policies favourable to economic growth or by adopting restrictive trade policy in the sector where child labour is used. A ban on child labour is 1 This

part draws from the author’s research article—Ray (2015), Trade versus Non-Trade Policy in a Two Sector General Equilibrium Framework. Bus Eco J 6: 138. https://doi.org/10.4172/21516219.1000138.

3.1 Types of Child Labour

33

not necessary. Hence, we can argue that a country should not always blindly follow every aspect of the WTO prescribed policy package. Given the institutional and technological characteristics and the trade pattern of the relevant country, an appropriate menu of non-trade policies to address the problem.

3.2 Different Causes for Child Labour2 Though there are clear provisions in the Indian Constitution to safeguard the interest of children by ensuring that they receive education and are not forced to work for a living, it is unfortunate that the problem of child labour exists to a large extent in our country. We distinguish below the demand-side and supply-side factors that influence child labour. (a) Demand-Side Factor: 1. Low wage and absence of Trade Union Pressure: It is a common phenomenon that child workers are invariably underpaid as compared to current market wages. They can be coaxed, admonished, pulled up and punished for defaults without jeopardizing relations. Moreover, children are not organized on lines of trade unions which fight for their rights. (b) Supply-Side Factors: 1. Poverty: The benefits to the household of sending a child to work are the wages of that child and the reduced education expenditures from not sending the child to school. Households that are very poor are much more likely to send their children to work, and child labour contributes to poverty in the next generation by reducing schooling attainment. This circular constellation of the link between poverty and child labour may lead to a vicious cycle of poverty. If schooling attainments can be improved, then the next generation’s income could be higher and their children can, in turn, become yet better educated. The most effective way to draw children out of damaging work is by reducing the poverty of parents. 2. Adult Literacy: Gross inequalities in educational field arise from differences in the home environment. A child from a rural household or an urban slum having nonliterate parents does not get the same opportunity, which a child from an upper class home with highly educated parents gets. Socio-economic status is a crucial determinant of success at school. Education is a key area of development as this enables the child to realize his/her potential as a human 2 Refer

to Chatterjee and Ray (2016), Op Cit.

34

3 Types and Causes of Child Labour

being. Unless the parents are educated, they cannot feel the values of education. Illiteracy and parents lack of interest in education has been regarded as the major cause for educational backwardness. For the illiterate parents, education is viewed as useless since it provides no immediate returns, i.e. an income. 3. Expenditure on Education: Hike in expenditure on education will curtail school enrolment rate. It is often observed that due to increased educational expenses, increasing dropout rates are observed in higher stages of education. The maximum number of students are enrolled in the primary stage while many of them dropout in higher secondary stage and the dropout is more severe among girls. 4. Access to School: Shortage of numbers of schools, inadequate infrastructure in schools or distance of schools from remote rural residences will curtail the school enrolment rate. Now, we will try to derive a specific child labour supply function from the utility-maximizing behaviour of the representative adult worker and will examine the impact of different determinants on the supply of child labour.

3.3 Quantitative Derivation of Supply Function of Child Labour* *This Section Draws from the Authors’ Chapter, “Impact of Trade Restriction on Child Labour Supply and the Role of Parents’ Utility Function: A Two Sector General Equilibrium Analysis” in Malabika Roy and Saikat Sinha Roy (Eds.), International Trade and International Finance, Springer 2016. In our model, the supply function of child labour is derived from the utility— maximizing behavior of the representative adult worker. His utility function is given by   U = U (X, Y ) − U L CS

(3.1)

The worker derives utility from the consumption of the final goods and disutility from child labour. For analytical simplicity, let us consider the following specific algebraic form of the utility function:  α 0 0 − ⎣ (3.15) Z β α−1

As the adult labour endowment increases, the supply of child labour in the economy increases. Case 5: Impact of increase in tariff rate 1 1 ∂ L CS =− S ∂t Z LC



 α 1 {P x(1 − t)} α−1 (−1) < 0 α−1

(3.16)

Higher is the rate of tariff imposed by the rest of the world on the exported product of the small open economy, lower will be the incentive for the producers to produce the product. Hence, child labour supply will fall. Thus, the aggregate child labour supply function in functional form can be written as   ∂ L CS ∂ L CS ∂ L CS ∂ L CS ∂ L CS > 0, > 0, > 0, 0, 0 < a, b, c < 1

(3.18)

It satisfies all the standard properties and it is homogeneous of degree (a + b – c) where (a + b − c) ≤ 1 or ≥ 1. The adult worker maximizes the utility function subject to the budget constraint  P x(1 − t)X + PY Y =

  WA L CS + W A L + r K /L β

The Lagrangian expression for the constrained optimization problem is

(3.19)

   A    C W + λ PX (1 − t)X + PY Y − L CS + W A L + r K /L (3.20) M = AX a Y b / L CS β

38

3 Types and Causes of Child Labour

The First-order Conditions (FOCs) for maximization are  −c δ M/δ X = Aa X a−1 Y b L CS + λPX (1 − t) = 0

(3.21)

 −c δ M/δY = Ab X a Y b−1 L CS + λPY = 0

(3.22)

 −c−1   δ M/δL CS = −AcX a Y b L CS − λ W A /β /L = 0

(3.23)

 δ M/δY = PX (1 − t)X + PY Y −

  WA L CS + W A L + r K /L = 0 (3.24) β

From FOCs, we get ⎤  A W /β L S X = (a/c) ( L ) C /PX (1 − t) ⎦  A W /β L S Y = (b/c) ( L ) C /PY

(3.25)

Substituting the values of X and Y in the budget constraint, we get    W A /β L CS PX (1 − t)(a/c) /PX (1 − t) L     A  S W A /β L CS W /β L C rK S +WA + + PY (b/c) /PY L C /PY = L L L     β 1 + r KA β LW A + r K c (3.26) ∴ L CS =  a b L W 1  = A W [(a + b) − c] +c−L c This is the aggregate supply function of child labour. We shall now analyse its properties.   log L CS = log β + log c + log L W A + r K − log W A − log{(a + b) − c} (3.27) ∴

1 L −r K 1 ∂ L CS  0 LW A + r K L CS ∂r

(3.29)

1 ∂ L CS r = >0 S ∂K A LW + r K LC

(3.30)

1 ∂ L CS WA = >0 S ∂L LW A + r K LC

(3.31)

3.3 Quantitative Derivation of Supply Function …

39

Hence, the aggregate supply function in functional form can be written as   ∂ L CS ∂ L CS ∂ L CS ∂ L CS < 0, > 0, > 0, >0 L CS = f W A , r, K , L , where ∂W A ∂r ∂K ∂L (3.32) Thus, the determinants of child labour supply remains similar to that of the earlier case.

3.4 A Summing-Up It is clear that child labour in various forms are influenced by both demand- and supply-side factors. The demand-side factors are primarily influential in inducing the producers to hire child labour, mainly because the child wage rate is lower than the adult wage rate. Of course, the scale effect of the volume of production, and the rental rate on capital are also important factors that govern demand for child labour, particularly in a labour surplus underdeveloped country. But the supply-side factors are also important as they influence children’s participation in the labour market poverty of parents, educational access, prevalence of tariff protection against the exports of developing countries, and also the uncertainty about the return on education are important factors that tend to influence child labour supply. The fact that a vast majority of child labourers are engaged in hazardous types of occupations are also guided by similar demand-side and supply-side factors plus the proliferation of illegal and dangerous activities that span over different parts of the globe and the legal frameworks in such countries that permit such high incidence of child labour use.

References Burra, N. (1995). Born to work: Child labour in India. Delhi: Oxford University Press. Chatterjee, B., & Ray, R. (2016). Impact of trade restriction on child labour supply and the role of parents’ utility function: A two sector general equilibrium analysis. In M. Roy, & S. S. Roy (eds.), International trade and international finance. Springer. Grootaert, C., & Kanbur, R. (1995). Child labour: An economic perspective. International Labour Review, 134(2), 187–203. ILO. (2017). Global Estimates of Child Labour: Results and Trends, 2012–2016. ILO: Geneva. Ray, R., & Chatterjee, B. (2010). Education subsidy, adult unemployment and the incidence of child labour in an open economy: A three sector general equilibrium analysis. The Indian Economic Journal, 57(4), 37–59. Ray, R. (2014). Import restrictions, capital accumulation and use of child labour—A general equilibrium analysis. In A. N. Ghosh & A. K. Karmark (eds.), Analytical issues in trade, development and finance: Essays in Honour of Biswajit Chatterjee. India: Springer. Ray, R. (2015). Trade versus non-trade policy in a two sector general equilibrium framework. Business and Economics Journal, 6(2), 138. https://doi.org/10.4172/2151-6219.1000138 (An open access journal). www.alliance.org.

Chapter 4

Survey of Literature

Abstract This chapter provides a detailed review of the general issues in the literature on child labour and development, the empirical and theoretical issues dealt with in the literature and their implications for the choice of the trade and non-trade policies advocated for the elimination or curbing the problem of widespread use of child labour in developing nations. Some of the problems associated with the implementation of alternative policies for curbing the incidence of child labour are also discussed in the chapter.

4.1 Introduction The problem of child labour is an important issue today not only for economist but also for government, social groups and concerned citizens of every nation. There is no denying the fact that child labour use is undesirable and needs to be abolished. Instead, children should be enrolled in educational or vocational programmes, possessing skills for a better future. However, with imperfectly working institutions and inadequate resources (as in the case in most developing nations), abolition of child labour by law may not ensure better educational opportunities for these children. The prevalence of child labour employment in the developing economies has been rooted in economic reasons, which can be addressed in terms of a demand–supply framework (Ray 2014, 2015; Chatterjee and Ray 2016a, b). The demand for child labour comes from all the sectors of the economy. Children employed in the agricultural sector are seen to assist their parents in a variety of agricultural operations, viz. sowing, weeding, harvesting and threshing etc. and also in allied activities like fishing, village household, cottage industries etc., and there are a number of reasons for such engagement of children. First, children, not being organized through trade unions, can be asked to work for long hours with a minimum wage payment, and rural household industries could easily evade legal restrictions. The producers in such industries prefer cheap child labour to adult workers whose wages are high relatively. Second, there is the ‘nimble finger’ argument which states that since the human body of a child is soft, flexible and pliable in childhood, such child workers can learn quite easily how to twist, bend, stretch as required by a particular job. Moreover, children © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_4

41

42

4 Survey of Literature

are more amenable to discipline and control than adult labourers. The key motive for employment of child labour in labour surplus developing countries is primarily to minimize costs and thus to maximize profits, so as to have a comparative advantage in the world market.

4.2 General Issues The World Trade Organization (WTO) was established on 1 January 1995, following the Maracas Agreement of the Uruguay Round (UR) with the objective of ensuring fair and free trade practices among member countries, so as to initiate trade negotiations for the reduction of trade barriers, establish rules of behaviour and resolution of disputes. Members of WTO are guided to create and alter rules governing trade policies and trade liberalization. Two fundamental elements can be identified in this context: achieving an agreement, and enforcing it. International trade theory suggests small open economies cannot improve their national welfare through the practice of restrictive trade policies, whereas large countries through trade restriction policies may be able to influence the international terms of trade to their own favour and gain in national welfare. For the world as a whole, however, the imposition of trade restrictions by one or more countries can only reduce welfare. Therefore, for both small and large open economies, adoption of free trade policies is Pareto superior alternative to restrictive trade policies. On the world scale, liberal trade is a positive sum game, and there are incentives to cooperate to lower trade barriers through multilateral agreements. After its inception, the discussions in the WTO, among others, focused on the issues relating to the adoption of labour standards and environmental standards in the domain of world trade. In the Seattle meeting held from 29 November to 3 December 1999, the developed countries tried to force these issues. The developed nations raised the twin issues of core labour standards, and of use of child labour in the meetings of WTO and tended to argue that many developing countries are engaged in exporting child labour-intensive commodities, and that such trade be banned as unfair because on ethical considerations, the children should go to school rather than to work. It is important to note that such a position by the developed nations ignores the fact that in poor developing countries, the use of child labour is conditioned by economic compulsions of the poor parents. In such poor developing countries, children are sent to work because their families are too poor to send them to school, and a policy of banning child labour would further worsen the family incomes of such poor parents, with the result that children in these countries would be compelled to move onto even worse forms of work, often hazardous in nature. When the countries from the European Union boycotted the exports of the garments manufactured in Bangladesh on the plea that such manufacturing of garments uses child labour on a large scale, the result was that many working children in Bangladesh lost their jobs in the garments sector, and landed up in other worse paid jobs. Thus, the developing countries were not in a position to take trade measures to eliminate

4.2 General Issues

43

the incidence of child labour, primarily because these countries do not have adequate employment opportunities for their adult labour force as well. Banning- quantitative or tariff-ridden is unlikely to have a dent on the use of child labour- if anything, such measures may aggravate the incidence of child labour. ‘Preventing child labour use either by legislation, or by use of restrictive trade policies may not be successful in eliminating the incidence of child labour use in such labour abundant poor countries like India, they end up losing their international market access or worsening trade balances and lowering the level of national welfare, while other non-trade policies may be somewhat effective in reducing this incidence’ (Ray 2015). In fact, the clamour by the developed countries for the adoption of minimal ‘International Labour Standards’ and ‘social clauses’ as a prerequisite for trade, basically hinges on the argument that use of child labour in the developing countries gives them a trading advantage in labour-intensive goods. The developed nations argued that such advantage is illegitimate because according to them, child labour is socially unacceptable. Such an advocacy by the developed nations for adopting restrictive trade policies or of banning exports of products containing child labour as an input is basically to encourage protectionism, which is supposed to hurt not only workers but also consumers in developing countries. Naturally, the meetings of WTO witnessed opposition by the developing countries against such trade sanction or trade restriction measures by the developed countries. Thus, it seems inappropriate to use policies of trade restrictions to deal with the presence and persistence of child labour in developing countries. According to Gupta (2002), in many developing countries, financial markets are often dysfunctional, such that use of child labour in such economies could be curtailed, if only the parents could finance ‘their children’s exit from the labour force and entry into schooling from the increased future earnings of the child’ (Gupta 2002). In fact, there exists considerable time lag between the investment in child education and the return to that investment that may accrue to the parents (the adult labourers), so that the parents do not undertake the risk for such investment in education of their child and instead prefer to send them to work which yield immediate return. There may be a limited or marginal impact on the total employment of the children in developing countries that such international pressure like boycott of the child labour-intensive product may bring in. But, given the wide range of activities which use child labour in many developing nations, the costs of enforcement of a ban on the use of child labour are likely to be huge, and evading enforcement becomes easy. Also, children pushed out of one industry can often readily move to another, and maybe with worse economic conditions. This would aggravate the misery of the working children, and not eliminate its prevalence in the short run. In an expanding economy, demand for both adult and child labour would expand, but the increase in income of the adult labour force may not be sufficient to induce them to finance children’s education and immediately contract the supply of child labourers. May be the introduction of compulsory education to children could have a longer term impact on the use of child labour, as the experiences of countries like Rwanda, Zambia and Sri Lanka tend to suggest. Many countries like South Korea, Taiwan, China and Sri Lanka have made primary education compulsory even when their per capita incomes were

44

4 Survey of Literature

low and poverty was widespread. In fact, establishment of compulsory education is often viewed as a necessary condition for the reduction and abolition of child labour, although it may not be sufficient. Education by itself may not eradicate child labour, yet it reduces its incidence by discouraging it, and it is easier to keep a tab on school enrolment and attendance than to enforce legal ban on the use of child labour in production. Such a view is endorsed by Weiner (1996), who argues that NGOs, religious and social organizations in conjunction with the parent’s association could play a crucial and important role in encouraging formal and/or non-formal education of the children and discourage dropouts so that over time engagement of children to work get discouraged. According to Weiner (1996), one need not wait for legal ban or trade sanctions to reduce the incidence of working children—a more effective way is to make education compulsory to all children, and may be made free, and monitor and improve the conditions of such education.

4.3 Empirical Literature There exists vast empirical literature on child labour covering different causes and magnitudes of child labour. Some papers are also broadly suggesting different remedial measures for child labour problem. The types of activities in which children participate, both within and between countries, are quite diverse—most working children support their families in domestic work, household enterprises or in family farms. Edmonds (2002) commented on the link between globalization and poverty and argues that since ‘Poverty creates child labor’ and since globalization tend to usher in growth in the developing countries, the incomes of the hitherto poor parents tend to improve which induces them to send their children to school rather than to work. He argues that since poor parents tend to depend on the incomes of their children to support their household living standards, persistence of child labour use is quite widespread in poor developing countries, and when the earning opportunities of the parents improve due to enhance growth induced by globalization process (through increased inflow of foreign capital, or larger access to developed countries’ markets etc.), they prefer sending their children to school with the expectation of improving their skills and potential incomes in the longer run. Edmonds (2002) cites the experience of Vietnam where during 1993–1998, the work participation rates for children declined by around 25% as the country experienced high boom whereby the incomes of the poorer households experienced steady high growth. Thus ‘improvements in living standards can explain 94% of the decline in child labor for households near the poverty line in Vietnam’ (Edmonds 2002). There are two important caveats to this argument. First, globalization-induced growth in developing countries also tends to raise the wages of child labourers, which further induces their supply in the labour market. Second, and most importantly, globalization also has increased ‘the ability of rich countries to influence policy in the world’s poorer nations’—the rich nations can use the threat of trade sanctions and trade policies to curtail child labour use in

4.3 Empirical Literature

45

developing countries. In December 2001, Pascal Lamy, the then EU Trade Commissioner announced that the foreign ministers of EU had approved a preferential tariff scheme for countries such that they could adhere to ILO’s labour standards including child labour. Both the Bush and the Clinton administrations had made similar policy pronouncements, and in 1999, the U.S. imposed quotas on Cambodia’s garment industry on grounds of use of child labour. Such protectionist threats by developed nations had a worsening impact on the conditions of child workers, and forced its spread, instead of curtailing the incidence. Edmonds and Pavcnik (2002) in their paper ‘Does Globalization Increase Child Labour? Evidence from Vietnam’ have tried to establish empirically the relation between globalization and child labour. To explain the connect between child labour and globalization, they considered the link between changes in the price of rice in Vietnam and changes in child labour use, and argued that during 1993–1998, the average domestic price of rice in Vietnam rose by 29% relative to the consumer price index, and around 9% (quite large) reduction in child labour was observed. They argued that households that own land experienced larger reductions in child labour when relative price rose. An increase in the relative price of rice (potentially stemming from liberalized trade policy) resulting in the enhancement of rural household income, which resulted in the decline of child labour use and an increase in school in enrollment. A range of issues about use of child labour, its extent and main characteristics had been noted by Edmonds (2005), who argued that although poverty reduction may ultimately reduce the incidence of child labour, in the intermediate run, however, engagement of some economic events or policies can have ambiguous effects on child labour. Since the distribution of the benefits of growth across generations is uneven and the effect on inducement of sending children to school rather than to work is weak and time consuming (Edmonds 2005). Edmonds (2005) also assesses the different policy options to reduce the worldwide prevalence of child labour use. According to him, improvements in educational systems and provision of financial incentives to poor parents to induce them to send their children to school, rather than to work, seem more practical solutions than punitive measures like outright ban or trade sanctions. Ahmed’s (1999) empirical cross-country study attempts to analyse ‘whether and to what extent changes in GNP per capita, poverty, inequality, structure of the economy, demographic factors and basic education could effectively reduce the employment of child labour’ (vide Ahmed 1999). To explain a substantial proportion of the crosscountry variations in child labour, he has used an econometric model to find that ‘poverty (percentage of the population below the poverty line), income inequality (measured by the ‘Gini coefficient’), the percentage of population below 15 years of age, and the share of agriculture in GDP are positively associated with the percentage of children working’ (Ahmed 1999). But, a negative association of child labour is expected with GNP per capita, school enrolment (percentage of primary age group enrolled in education) and with the adult and female literacy rate, argues Ahmed (1999). Of these, poverty is found to be only a minor factor influencing the incidence of child labour, but school enrollment is found to be a major one, although it must be realized that since enrollment data in most countries are highly exaggerated in

46

4 Survey of Literature

official statistics, and there are combination of working as well school attending children, such a relationship with school enrollment is somewhat blurred and suspect. Ahmed’s (1999) multiple regression analysis confirms strong association between the high incidence of child labour, and high dependence on agriculture, and very unequal income distribution in developing economies. According to Ray (2000), the exploitation of working men, women and children is intrinsic to the capitalist society. The issue of rights cannot be resolved as long as the predatory, rapacious, capitalists system is at the root of the system. Ray (2000) mentioned two schools of thought regarding the incidence of child labour. The proponents of the poverty school (first one) mention about compulsions of poverty of the parents as reasons to send their children to work, rather than to school. The employers of the child labour, on the other hand, perpetuate this scheme of things because it enables them to exploit the labour and increase their profits (i.e. the child wage rate is low). The children are considered to be cheaper than adult labour. The other school of thought is the education school—the proponents of this school argue that not poverty per se, but inadequacy of schools, limited access to education, irrelevance of the courses taught and the poor quality of education imparted, etc., together contribute to the apathy of parents to send their children to school. The post facto return on education and skill formation of the children is considered as too uncertain to rely on and persuade the poor parents to send their children to school rather than send them to work which brings immediate return, however, paltry it might be. There have been interesting debates in the literature over the definition of child labour and child work and the contribution of children’s work in the informal economy. Burra (2005) argues that the distinction at the conceptual level between child labour and child work is essentially flawed, and be abandoned. In fact, since unorganized children are unlikely to demand higher wages, take longer breaks from work, shirk their jobs, etc., that are typically observed in cases of adult labour force, the presumption that children are preferred to adult labour because of low wages is also not correct, as many studies point towards the preference for child workers on grounds of easier controllability. Children who are working outside the formal labour market and are not attending schools have been defined as ‘nowhere’ children by economist like Chaudhri (1996), who does not seem to consider the economic contributions of children as being significant. According to him, about 100 million children in the world are categorized as ‘nowhere children’. In India, the important contribution of children to the informal economy are recognized in official discourse as well, and there has been a strong plea to include all out-of-school children in the estimate of child labour. Many children start working for full-time work because schools within their neighbourhoods are often not available. The programme of universalization of education as advocated by the Ministry of Human Resource Development (MHRD), Government of India, also hinges on the elimination of child labour first. The ‘Sarva Shiksha Abhiyan’ programme of the Government of India attempts to bring the working children within the fold of formal school system through bridge courses, informal evening schooling, etc., and at present, such efforts are in operation in India. In fact, universalization of elementary education appears to be a necessary, if not sufficient,

4.3 Empirical Literature

47

condition for the elimination of both child labour and end of educational deprivations in a developing country like India. Cigno et al. (2002) argue that the view that international trade and increasing economic integration among countries necessarily raise the incidence of child labour is not supported by available empirical evidence. They argue that the costs and benefits of education influence the decisions of parents about whether a child would go to school or work. Their decision regarding investment on child education may be influenced by the liquidity constraints of households, and in developing countries endowed with a large stock of uneducated workforce, the wage rates of uneducated workers may rise relative to the educated following globalization and this could reduce the incentive to educate a child, and make the child work at the earliest opportunity. However, if such a rise in the wage rate of uneducated workers rises takes place both in absolute as well as relative terms, there could be positive income effect on the demand for education. For both the categories of least developing and relatively advanced developing countries, the net effects of globalization are ambiguous, available cross-country evidences (after controlling for GDP, skill composition of the adult workforce, and weight of public health expenditure in GDP, as noted by Cigno et al. (2002), indicate that trade openness does have a negative, yet insignificant, effect on child labour. If the economy’s share of educated workforce and that of public health expenditure in GDP are high, there are low incidences of child labour use and efforts to reduce child labour are found to be relatively successful. Thus, exposure to international trade and globalization process, except for countries with a largely uneducated workforce (who may be left out of the globalization process), does not necessarily augment the size of child labour workforce. It is not globalization per se, but the ability to take part in it, and the associated circumstances of deprivation that may be related to explain the prevalence and persistence of child labour engagements in the labour markets of developing countries. One important tool for reducing child labour could be a direct targeted subsidy to families, who send their children to school. Through such direct grant to the family of any child who is enrolled in school, such programme tries to address the root causes of sending children to work by balancing the current cost of moving a child out of the labour force and into school with a current grant (Ambreen 2017). In Mexico, the Progresa programme (later named as ‘Opportunidades’) was introduced in 1998 in a randomized sequence to provide ‘mothers of enrolled students in rural Mexico with grants that have a value slightly less than the wage that would be earned by the child if she is working full time’ (Schultz 2001). According to the estimates by Schultz (2001), the programme has resulted in an increase in the schooling of about two-thirds of a year (off a baseline attainment of 6.8 years), and that child labour use correspondingly fell. The most dramatic effects are found for girls enrolled in secondary school, the estimated drop in labour force participation was by almost fifty percentage points, and the real rate of return, estimated by Schultz (2001) was of the order of around 8%. Nicaragua also instituted the similar programme, ‘Red de Protection Social (RPS)’ in 2000, which reportedly had a massive impact, i.e. to reduce the proportion of working children by almost 9 percentage points (from a base of 27%) (vide Maluccio 2003).

48

4 Survey of Literature

On the basis data from Nepal and Pakistan, Ray (2001) investigated the interaction between the child labour hours and child schooling experience, while Ravallion and Wodon (2000) have investigated the effects of an enrollment subsidy on schooling and child labour in Bangladesh. According to the estimates of Ravallion and Wodon (2000), in rural Bangladesh following completion of primary schooling by children, the expected per capita consumption of households increased by 9% (controlling for a range of individual and household characteristics). And other benefits from the higher school attendance induced by the programme, like better health care and greater ability to participate in society, also accrued. Ravallion and Wodon (2000) further argue that the FFE programme [the Food-For-Education (FFE) programme] in rural Bangladesh had enhanced primary school enrollment from approximately 75% to over 90%, and there was a drop in the participation rate of children workers by about 30% for boys and by about 20% for girls. The impact of midday meal scheme in India has been studied by Khera (2006) and Dreze (2004)—their argument has been that midday meals schemes by fostering higher school enrolment, more regular attendance and better learning achievements, also enlarge the educational opportunities of children and induce parents to send their children to school, rather than to work. Thus, it is clear that whenever there have been direct targeted monetary subsidies for school enrollment or a direct food subsidy in the form of midday meal or FEE programme, the incidence of children’s work seems to diminish in developing countries. The real issue is how to manage such programmes with efficiency and effectiveness, and since in many developing economies such governance issues create problem, the effectiveness of such programmes in discouraging child work has been rather moderate. Two broad reasons have been identified in the empirical literature in explaining children’s work. The first one is that abject poverty of parents compels them to send their children to work, rather than to send them for education (see Bhalotra 2000; Ray 1999). Second, in many developing countries, the rate of return to basic, primary-level education as provided to poor children is very low, ‘not only because of high rates of time discount but also due to the poor quality of education. In addition, since decisions on child labour are inherently inter-temporal, credit markets play a significant role in influencing them. Even if returns to education were significantly high, sending a child to school instead of to work entails a sacrifice of current income in favour of future earnings’ (Mukherjee and Sinha 2009). Mukherjee and Sinha (2009) have developed a model, where the choice confronting parents is child labour versus education, and the factors that crucially influence such decision include the existence of informal sector, uncertainty in job prospect, and above all the paternal attitude towards education. If, and only if, the poor households have reasonable access to credit markets, ca one expect such parents to send their children to school rather than to work, because in such cases, they do not have to sacrifice their current consumption significantly, and if such schooling is free and is supported by provisions of midday meal to the students, there are additional supplementary nutritional benefits to such children, and this may thus outweigh the disadvantages of poor quality schooling infrastructure. Mukherjee and Sinha (2009) argued that ‘even the altruistic parents might withdraw their children from school and put them to work in the informal

4.3 Empirical Literature

49

sector’. The entire discourse has been to focus on intergenerational aspects of parental decisions which have a significant bearing on the incidence of child labour in the locality or the region. Regarding the effectiveness of trade sanctions in reducing the incidence of child labour, Mukherjee and Sinha (2009) argued that while there may be immediate effect of reducing the incidence of child labour, certification would have ambiguous effect as was argued by Ranjan (2001), Gupta (2002) and others. Even a policy of compulsory schooling might induce early age school enrollment of children, but there might be substantial drop out at the later ages. Such drop out children after some schooling at relatively higher ages may often join the informal sector as workers, and thus the total size of child workers may not significantly diminish over time in such economies. There is a marked paucity of empirical work on the intergenerational aspects of child labour, although there are some theoretical work (e.g. Baland and Robinson 2000; Basu 1999; Ranjan 1999). The empirical works on child labour have primarily focused on isolating the determinants of child labour using survey data (see, e.g. Ray 1999, 2000; Jensen and Nielsen 1997; Patrions and Psacharopoulos 1997; Grootaert and Kanbur 1995). Emerson and Souza (2002), using the Brazilian household survey data, however, have found strong evidence that there exists a direct link between the child labour status of the parents and their children actually working as it affects the permanent family income, i.e. if the children in the poor families actually go to the labour market and thereby augment family incomes, the parents in such families do not take initiatives to withdraw their children from work. In the Indian context, the preponderance of child labour in Uttar Pradesh has surfaced in the discussion. Many of the studies on child labour in India have been conducted in the local carpet industry, the glass bangle industry, the leather industry or the brass industry (Singh 1990, 1992; Singh and Mahanty 1993; Kulshreshtra 1994). Lieten (2000) conducted a microstudy, in the Faizabad, district of state of Uttar Pradesh, where, going around the villages in the late 1990s, he found that children are attending schools in high numbers, both the private schools (by the children of families who can afford the extra expenses) and the public schools. The figures suggest that around 25% of the girls and 60% of the boys go to school. Only a few of them are registered as working. His field level survey data also suggested that involvement in child labour, as it is officially defined, starts from around age of 12 onwards in the case of boys, whereas, in the case of girls, it remains rather limited even at that age. The fact remains that despite the government’s attempt to achieve the goal of 100% enrolment and retention of children with schooling facilities in all habitations, through the ‘Sarva Shiksha Abhiyan’, India’s record in achieving the goal of universalization of elementary education has remained rather poor, particularly for girl children. India, like many developing nations, has introduced zero fees in primary education, and the 86th amendment to the Indian constitution has made free and compulsory education a fundamental right for all children in the age group 6–14 years. Chandrasekher and Mukhopadhyay (2006) have sought to incorporate the direct costs of attending primary schools in India—they have argued that making primary education completely free will not necessarily increase the attendance rates to 100%. The econometric model of Duraisamy (2000), using a large scale national level survey data for the

50

4 Survey of Literature

year 1994, has sought to identify the determinants of schooling and work participation of boys and girls. He has brought into focus the third category of children referred to as the ‘invisible’ children a, and sought to integrate the child schooling and work participation decisions. According to him, parental education, availability of middle schools and family income significantly increase the probability of children’s school attendance and reduce the likelihood of children participating in work. Of these, maternal education increases more the likelihood of a girl child’s school enrollment than boys and also reduces more the work participation of girls over boys. Jayachandran (2001) has referred to ‘bhonga shalas’ experiments in the brick kiln industry, which provide employment opportunities to tribal youth and ensure continuity of educational process. Through his survey, he has identified several constraints facing the running of ‘bhonga shalas’ schools, which include reluctance and opposition of brick kiln owners, reluctance on part of the district administration to allocate funds for the functioning of bhonga shala schools for tribal children, oppose health check-up services conducted by the Vidhayak Sansad, deny active participation by the education department in the teacher training process, provision of educational material and some food incentive to the students, visits and inspections by the district education officers, etc., and the insistence of the brick kiln owners not to grant leave to child labourers during examination days, and in fact forcing them to work at kilns on such days. There are also obstacles and hurdles towards the participation of girl children in the activities of bhonga sala schools, although, in the samples chosen by Jayachandran (2001), girls children outnumber those of boys in the engagement of the activities of brick kilns. As a result, the effectiveness of such schools in perpetuating education among tribal child workers has remained sluggish and weak. Biggeri et al.’s study (2009) on ‘Child Labour in Industrial Outworker Households in India’ investigate how the activities of home-based working children affect their well-being in terms of education and health and if they are more likely to work more than other children, and indicate the policy implications. Based on the data and information collected over 2000 to 2001 for five countries (India, Pakistan, Thailand, Indonesia and Phillippines), they tried to explain the specificity of homebased working households. Both quantitative and qualitative methods were used for the study on sub-contracted household workers in India. The 50% incidence of child labour in home-based working household from the survey data is found to be much higher than the incidence of child labour on a national scale in India. Regarding child schooling, the authors observe that among the younger children, three-fourths are attending school in home-based working household. Among older, the share of children attending school is much lower. About 16% of the younger children and less than a tenth of older fall in the neither studying nor working category. The authors have also mentioned some determinants of child labour in home-based working households—age of a child (with marginal effect on the probability of full-time work around 7.5%), religion of the household and of course, the human capital endowment of the household, etc. Gender does not seem to affect the probability of studying. The age dependency ratio within the household is non-significant too. Income per capita and membership of a home-based work organization show a positive effect (Biggeri et al. 2009). The main findings of this study have shown that education, joint

4.3 Empirical Literature

51

action and social protection happen to be the main factors that influence the human development level of home-based working households, and that if provisions of community-based child care are made, the older girls of such home-based enterprises could go the schools and work part-time in household activities. For recent estimates of working children in the age group of 5–14 years, their distribution between boys and girls and across different regions and occupations, one may refer to ILO (2017). Bacolod and Ranjan (2004) offer a rigorous theoretical and empirical analysis of schooling child labour and ‘idleness’ (neither work nor school). The authors show theoretically that even the poorest households will send working children to school if their child is sufficiently able. Using rich data from the Philippines, they argue that apart from child ability and household wealth, factors, e.g. mother’s labour supply, the presence of a family business, access to good school quality, etc. act as the most important determinates of child idleness and the use of child labour. Appropriate policy interventions that encourage investments in human capital, rather than blanket ban on child labour, should be taken to address the problems of child labour and idle children. Lahiri and Sylwester (2005) have examined the interactions between investments in physical capital and that in human capital, when the latter is financed by taxing the group the makes the former. In particular, they develop a twoperiod general equilibrium model in which there are two groups: rich and poor. The rich are responsible for investments in physical capital. The government decides on the quality of education by investing in schools financed by taxation of the rich. The poor families decide on the school participation rate. The theoretical part of the paper examines the channels through which a rise in the tax rate affects both the level of investments in physical capital and the school participation rate. The empirical part of the paper estimates these relationships econometrically using cross-country data for the year 2001. The study confirms that endeavours to increase the human capital stock by taxation can adversely affect investment in physical capital and thus longterm job opportunities for the poor, which in turn can affect investments in human capital. In the other words, it is very important that the international community demonstrates its concern for child labour and low school participation rates in the developing countries by assisting those countries more in improving the quality of education. An important dimension to this problem is the gender disparity in school enrollment. Attempts to portray the rights of child labour and mitigate its incidence in specific industry during the 1990s resulted in several private initiatives. One such ‘labelling’ initiative is the Rugmark scheme which seeks to provide economic incentives for manufacturers to desist child labour use in the carpet industry of Uttar Pradesh. Labelling schemes like Rugmark are considered as measures which convey the consumers’ indignation towards the plight of child labourers in the industry. Based on field research in the districts of Mirzapur and Bhadohi in UP, Ravi (2001) sought to examine the effectiveness of such labelling initiatives towards containing child labour. Although such labelling initiatives helped in creating the widespread awareness against the use of child labour in the carpet industry, these measures did not prove to be sufficient for the elimination of child labour in the industry, argued Ravi (2001).

52

4 Survey of Literature

From the different empirical investigations, it emerges that there is no ‘ideal’ or no single policy prescription for the elimination of child labour. What may matter and be relevant are combinations of a package of policies which could have significant cumulative contributions to the permanent reduction of child labour. The moot question is whether trade policies or a combination of non-trade policies could be relied upon to effectively diminish the incidence of child labour in a developing country or not.

4.4 Theoretical Literature* *This Section Draws with Appropriate Permissions, from the Authors’ Own Research Paper, Runa Ray & Biswajit Chatterjee (2012), “Trade and Child Labour—a Review of Theoretical Literature”, Journal of International Economics, Institute of Public Enterprise, Vol 3, Issue 1, January–June, 2012, pp. 93–113 Two sets of questions primarily dominate the theoretical discussions literature on child labour: (i) why child labour originates, and (ii) how the employment of child labour as an input in a structurally differentiated open economy under conditions competitive general equilibrium is affected by the different forms of trade and nontrade policies, with implications for factor prices, composition of output, national income, national welfare, trade balance and extent of urban adult unemployment and so on. These issues are important as explaining the microeconomic motivations for the genesis of child labour, particularly in a developing economy where formal and informal labour markets coexist, and there exist complementarities and sector specificities of certain factors in the production process. Trade sanctions policies as are advocated under the WTO’s norm of labour standard may not be as effective means to reduce the incidence of child labour, some combinations of non-trade policies by the government may have desired results under the specified market and production structures. Condemnation of child labour use on moral and humane grounds apart, the real issue for the analysts is to understand the determinants of child labour so as to evaluate its policy implications of various instruments to alter them. The needs for basic survival of poor families in developing nations constitute the basis of supplying child labour in the market as the same households, if endowed with high income and wealth, shall refrain from supplying child labour. Thus, children work in developing nations because of economic necessity and not due to a lack of potential consciousness. In other words, one can conceive of at least two reasonable equilibria—one where children work and the other where they do not. Basu (2000), Maskus (1997) and Ray (2002) point to higher family incomes, improved educational opportunities, etc., while Bhagwati (1995), and Dixit (2000) point to a general increase in the living standards of the poor through liberalized trade and improved access to world markets as factors that are expected to reduce the incidence of child labour. In such economies, the household or family supplies child labour in order to maximize its current income from the employment of the child labour in preference to the income expected from

4.4 Theoretical Literature*

53

their employment in future after the schooling of its children. Such households, being myopic because of low life expectancy of children, decide to maximize the present income rather than the future one, and the presence of occupational or sectoral dualism prompts such decision of poor parents in favour of short run economic gains. In the informal segment of such developing nations, children are engaged to work for paid and unpaid jobs, may be in family occupations. The existence of such dualistic structure at the techno-economic level of production and distribution is considered responsible for the supply of child labour. Basu and Van (1998) have sketched a model to show that for a class of situations, labour markets in the developing countries often exhibit multiple equilibria, which call for governmental interventions even in the absence of interpersonal externalities and altruistic behaviour of parents. The neoclassical theory of child labour supply seeks to explain that. Moreover, the low life expectancy of children in developing countries is another reason for households to supply child workers in the market. It is often argued that since a large part of the labour market in developing economies is unorganized and informal, the households there simply supply their family labour and child labour in such informal segments of the economy. A class of models tries to analyse intra-household decisionmaking, particularly with respect of the supply of child labour. Jafarey and Lahiri (2002) focused on the inter-temporal nature of decision-making of the households to find the effects of trade sanction against child labour. First, they have considered the child labour decisions taken in equilibrium by poor households—they have developed an inter-temporal model of child labour versus child education in which low value added of schooling, subjective parental attitudes towards education and the nature of credit markets, all play a role. To be precise, they have developed a two-period, two-good model in which both rich and poor families decide how many of its children should go to school in the first period. If a child goes to school, he/she becomes skilled and could earn higher wage in the second period. Jafarey and Lahiri (2002) consider the effects of a trade sanction against child labour products on these decisions. The inter-temporal nature of decision-making means that child labour will not only be directly affected by credit markets but also be affected via the response of credit market outcomes to trade sanctions (Jafarey and Lahiri 2002). They considered three alternative credit market scenarios: (i) all households can borrow or lend freely in a perfectly international capital market at an exogenous interest rate (the country is small); (ii) borrowing is subject to quantitative restrictions; (iii) all households can participate in a domestic credit market, but cannot borrow from or invest in foreign capital market. The first case represents the first-best scenario in which credit markets are perfect globally. While far from realistic, it represents a benchmark for policy makers at the international level to aspire to. It will also serve as a stepping stone for the analysis of the other two cases. The case of quantity-constrained borrowing is more realistic since it depicts a situation in which poor borrowers in developing countries lack easy access to credit markets. The third scenario represents a goal which domestic policy makers should aspire to. In many developing countries, poor rural households often take informal loans from rich households, and therefore, the third case possibly comes closest to reality. The different forms of imperfections and monopolistic practices prevailing in credit markets of developing economies call for

54

4 Survey of Literature

analysis of the constraints faced by such poor households in deciding whether their children should go for work or to schools with the expectation of a larger future income emanating from their resultant skill formation. On the nature of equilibrium, Jafarey and Lahiri (2002) show that the incidence of child labour is likely to fall as borrowing opportunities improve, from binding borrowing constraints to a domestic loan market in which the poor can freely borrow from the rich to a world capital market on which the country can really borrow. On the effects of trade sanctions on child labour they break them down into three components: (a) a ‘substitution’ effect, i.e. a reduction in the opportunity cost of sending a child to school, (b) an ‘income’ effect, i.e. a reduction in the parent’s marginal utility from educating the children, and (c) a ‘discount rate’ effect which affects the present value of the wage premium from attending school. The first effect encourages schooling while the second discourages it. While both of these are independent of the exact credit scenario, the third effect is crucially dependent on it. Clearly, this effect is absent under a perfect international credit market (scenario (i). However, it unambiguously discourages schooling under a binding borrowing constraint (scenario (ii)) and, with plausible restrictions, also discourages schooling under a domestic credit market (scenario (iii)), but by less than under borrowing constraints. Overall then, a trade sanction can induce a perverse effect on child labour (via the negative income effect) even the first-best case of perfect access to world capital markets, but the possibility of a perverse effect rises (due to the additional discount rate effect) as access to credit weakens (Jafarey and Lahiri (2002), Mukherjee and Sinha (2009). Maskus (1997) set out a series of hypotheses about trade and labour standards and examines whether and how deficient labour standard affect competitive advantage and trade among nations. Assuming prevailing standards as exogenous, Maskus (1997), with the help of a simple model of schooling choice where a representative household maximizes two-period utility function defined over consumption of food and non-food items, has shown that school going children incur a first period cost, but obtains a higher second period wage, the extent of premium depending on the productivity of education. An increase in the cost of education and in the minimum food consumption requirement tends to decrease the demand for education. Demand also depends on the (uneducated) adult wage, the discount rate, the stock of first period children and commodity prices. Incentives for school attendance would include a rise in parental wages, reduction in the costs of education and improvements in the productivity of education. In a small open economy framework with the adult labour-intensive exportable product X, and a capital-intensive importable final good Y, together with an non-traded intermediate good (N) which enter into the X sector, Maskus (1997) has shown that changes in demand for the export good translate directly into changes in demand for child labour. The supply of child labour into sector N is taken as a rising function of the child wage. This assumes that the substitution effect of child wage changes dominates any income effects. To capture the linkage between household income and child labour, he has made supply curve a negative function of the adult wage. The demand for child labour depends negatively on the wage and positively on output in sector X. Then he has considered different policies for eliminating child labour.

4.4 Theoretical Literature*

55

Regarding the derivation of the supply function of child labour, we may also refer to the works of Christopher and Kaboor (2003), who identified several factors that influence the level of child labour in a family, e.g. Size of family (+), nature of the Community (e.g. children belonging to communities like scheduled caste and scheduled tribes are found to work as child labourers in large numbers); Literacy level of parents (-); Occupation of parents (if the parents are engaged in an occupation where the wages earned are small in amount, they are forced to send their children to work in order to increase family earnings); Family income (inadequate level of earnings of a family encourages employment of children at work so that the deficit of earnings could be made good); Possession of property (owning of property is an indication of wealth, and families, which own a considerable amount of property, need feel under no compulsion to send their children to work). Following the lead by Eswaran (1996) and Basu and Van (1998), who consider child labour supply as the outcome of poverty of parents, but do not make welfare comparisons of alternative bequest possibilities, Baland and Robinson (2000) have developed an alternative micro-theoretic framework to show that the phenomenon of child labour can be Pareto inefficient, even when parents are fully altruistic toward their children. The main presumption in their model is the existence of a tradeoff between the supply of child labour at present and the accumulation of human capital if children are instead sent to school, and parents fail to internalize fully the negative effects of child labour because of the imperfections in the capital market, as is often the case in developing economies. Child labour may nonetheless persist ‘either when parents leave their children no bequests or when capital markets are imperfect’ (Baland and Robinson 2000), as in both the cases, the failure of the parents fail to internalize the socially efficient trade-off between child labour in the current period and earning ability in future get manifested in the persistence of child labour. When bequests are positive, and the parents could completely internalize the possible ‘adverse impact of child labour on the future income of their children’, they could adjust the size of their bequests and send their children to school. However, they argue that, ‘even if bequests are interior, child labour may be inefficiently high because parents face capital market imperfections, which stop their transferring the future reduction in bequest into the present’. The non-negativity constraint on bequests partly generates the inefficiency of child labour, and therefore it can be reasonably argued that with reverse altruism, where children transfer to their parents, efficiency of resource allocation may be restored. ‘However, the inefficiency of child labour stemming from imperfect capital markets is unaltered by the presence of reverse altruism’ (Baland and Robinson 2000). Thus child labour exists as a socially inefficient outcome when parents use it either as a substitute for negative bequests to transfer children’s income to them, or as a substitute for borrowing to transfer income from the future to the present (due to imperfections in the capital market). Such inefficiency cannot be cured through intra-family contracting as such contracting would neither be legal, nor self-enforcing in nature. In case the model is extended to accommodate heterogeneity of both firms and families, there may be a set of potentially Pareto improving policies that can be used to compensate the set of firms for not using child labour by taxing other groups of firms—i.e. one has to take into

56

4 Survey of Literature

account the distributional impact of alternative policy interventions to discourage the use of child labour. Pouliot (2006) has shown that even when parents are not credit constrained and bequests are interior, the presence of child labour can be inefficiently high when return to human capital is uncertain and insurance markets are incomplete. Since parents often tend to increase period 1 levels of child labour beyond the efficient label so as to offset uncertainty in period 2 income, such results are possible outcomes, but when there is uncertainty in the return to human capital and complete insurance markets are absent, child labour can be inefficiently high, since such uncertainty cannot be offset through reallocation of savings across periods or by altering the amount of bequests. The overlapping generation general equilibrium model of Ranjan (2001) highlights the importance of credit constraints faced by poor parents in developing countries, which ‘generate inefficient outcome of child labour’. Set to derive a positive relationship between inequality in the distribution of income, and the incidence of child labour, Ranjan (2001) argued that trade sanctions against such countries may not succeed in reducing the ‘incidence of child labour as it does not alter/reduce the inequality in the steady-state distribution of income’. On the other hand, ‘if a redistributive policy takes into account the ability of individuals, then even some inequality enhancing redistributions can reduce the incidence of child labour. For example, redistribution of income between two parents with equal income who are both below the threshold level of income and their children have identical ability can reduce the incidence of child labour’ (Ranjan 2001). Since in practice, it is ‘extremely difficult for a policymaker to perfectly observe the ability’, it is unlikely that such redistributive policies would have a significant impact on the reduction of the incidence of child labour. Here, one can make a distinction between a one-shot redistribution and a permanently redistributive policy. In the steady-state equilibrium, argued Ranjan (2001), an one-shot policy of redistribution of income, e.g. a small redistributive tax subsidy scheme (small proportional tax and a lump sum subsidy) that leaves the future wages unaffected, can have permanent impact on the reduction of the incidence of child labour, but the impact of a permanently redistributive policy is ambiguous because such a policy will tend to reduce, for example, the net of tax skilled wage and increase the net of tax unskilled wage. Since sending the children to work is the outcome of an optimal decision for poor parents, a legislative ban, by putting an additional constraint on their choices, would tend to diminish their welfare. ‘Any policy that induces poor parents to send their children to school is to be preferred over policies that coerce them into doing so, basically … such a policy would have to provide consumption support to the families of poor children to compensate for the foregone earnings of the children attending school’. Such a policy of redistribution would effectively bypass ‘the enforcement problems encountered in the case of a legislative ban’ (Ranjan 2001). Sahin and Ghosh (2016) have developed an overlapping generations model with imperfect monitoring to show the implications of a ban on exports with child labour imposed by importing countries. For a small open economy with elastic demand for exports, a ban on exports with child labour content would increase the cost of employing child labour and would thus tend to contract output of the export

4.4 Theoretical Literature*

57

sector, which uses child labour. ‘The final outcome will depend on whether the aggregate supply of unskilled labour by child workers in the pre-ban equilibrium is greater or less than the demand for unskilled labour in the import competing sector’ (Sahin and Ghosh 2016). If the former exceeds the latter, such a ban will have virtually no effect on the incidence of child labour use, and the opposite would obtain in the reverse case. A ban on child labour use would raise the wage rate of skilled adult workers and reduce the wage rate of child workers making them worse off compared to the pre-ban scenario. The rise in skilled worker’s wage leads to substitution of unskilled labour for skilled adult workers in the import competing sector, whose output increases at the expenses of the child labour using export sector. The increase in parental income raises children’s welfare as well as average welfare in both the short and the long run. Sahin and Ghosh (2016) recommend the use of strict laws to ban child labour coupled with policies to subsidize school education to encourage parents to send their children to school and opt for larger return from human capital formation as against the income of the children to supplement their current income. Sarkar and Sarkar (2012) have also developed a theoretical model to highlight the role of income inequality as a channel of persistence of child labour. ‘The interplay between income inequality and investments in human capital gives rise to a non-convergent dynamic path of income distribution characterised by clustering of steady state relative incomes around local poles. The two effects reinforce the intrahousehold substitution of health for schooling among the poor, thereby sustaining inequality and child labour’ (Sarkar and Sarkar 2012). While redistributive policies like public provision of education can be helpful to poor parents and may encourage private health investment, a ceteris paribus ban on child labour is likely to aggravate the problem and force them to send their children to work. In respect of the two-way link between poverty and child labour, Udry (2003) points out that since the poor parents of the developing nations send their children to work and not to school, child labour persists as long as poverty of parents continues. Children not going to school cannot improve their skills and hence future income, they, therefore, tend to remain poor even when they grow up as adults and the cycle continues. Since the financial markets in developing nations are imperfect, the separation in time between the immediate benefits and long-term ‘delayed costs of sending children to work’ often leads to persistence of widespread use of child labour, argues Udry (2003). Second, the costs and benefits of child labour are borne by different people at different points of time, and there are important externalities associated with education when children go to school, and this should be counted in the calculation of the costs of child labour. Even if the sacrifice of such non-market returns to education is ignored, ‘the primary costs of child labour are the lower future earnings of the child when she enters the adult labour market with low educational attainment’ (Udry 2003). He noted certain empirical regularities in case of child labour—e.g. studies reported in Fallon and Tzannatos (1998) indicate a strong negative relation between the incidence of child labour and household income, particularly in less developed countries, whereas Ray (2000) could find a strong negative correlation between child labour and household income, and a positive relationship between household income and school enrollment in Pakistan, but

58

4 Survey of Literature

no such relationship is found in Peru. When adult wage rate rises, the wages of child workers also tend to go up inducing a substitution effect favouring increase in child labour supply, and not increase school enrolment. On the other hand, Edmonds and Pavcnik (2002) have shown that in Vietnam, increases in rice prices were strongly associated with declines in child labour. Thus, Udry (2003) argues that one needs to look into the specific mechanisms that trap people ‘in the awful equilibrium of persistent poverty, excessive child labour and low education over generations’ (Udry 2003). Two such mechanisms that often work simultaneously are: (a) parents’ inability to seize advantageous long-run investments in children’s human capital because of credit market constraints; and (b) problems of agency within households. Thus, Udry (2003) argues that child labour is the outcome of a draconian choice made by poor households ‘faced with severely limited options’. Even if the parents are fully altruistic towards their children, so that they treat symmetrically the future costs to the current child labour with current benefits to the household, it is possible to find the existence of too much child labour and too little schooling because of the poorly functioning financial markets in developing countries. Since the benefits of child labour are realized immediately, poor households may find it too difficult to forego these immediate benefits, because they do not have access to credit markets. Thus, it can be argued that presence and persistence of child labour ‘is not a consequence of impatience or an unwillingness on the part of poor households to plan for the future, rather it is a reflection of poverty and inadequate access to capital markets’ (Udry 2003).

4.5 General Equilibrium Models There exists a rich literature which uses the general equilibrium framework of international trade in alternative versions to examine the effects of alternative economic policies—trade as well as non-trade, on the use of child labour and the associated macro indicators of development like employment, national welfare, trade balance, etc. The standard two-sector general equilibrium model of international trade (Jones 1965) derives a one-to-one relationship between factor prices and goods prices, which in turn, forms the basis of the Rybczynski, Heckscher Ohlin and Factor Price Equalization theorems, on the assumptions of the equality of the number of goods and factors and (2) constant returns to scale under perfect competition. However, the above relation of univalence between commodity and factor price ratios no longer holds, if we introduce economies of scale or the assumption of number of factors exceeding the number of goods. In the two-sector three-factor model (often referred to as the Ricardo-Viner or Specific Factors (SF) model in the literature) where each sector uses a sector-specific and a common mobile factor, Jones (1971) has shown that the one-to-one relationship between factor prices and goods prices no longer holds. The main result of this model is that an increase in the relative price of a good leads to a fall in wages in terms of that good but a rise in terms of the other good. Thus, the extent of mobility in the two factors reflects the basic properties of

4.5 General Equilibrium Models

59

this model—while Mussa (1982) and Grossman (1981) have tried to explore this issue further, their restrictive assumptions about the degree of mobility of alternative factors of production have marred their results from general applicability, a problem that was tackled effectively by Hill and Mendez (1983). The Hill and Mendez (1983) model provides a convenient framework for generating many of the fundamental results in neoclassical trade theory; in fact, they have shown that (i) many elements of the Stolper–Samuelson theorem (Stolper and Samuelson 1941) generalize to any degree of mobility, provided the two factors are equally mobile; and that (ii) if input use ratios are equal across sectors, then the final magnification effects obtained by Jones (1971) hinge only on the one factor being more mobile than the other, and not on the size of the mobility differential. Regarding growth balanced by sectors but uneven by factors, Hill and Mendez (1983) argued that when the two factors are equally mobile, the sector which pays a higher distributive share to the faster growing factor will experience an increased share of national product, irrespective of the extent of mobility of such factors, and again how mobile these factors are; and further if the distributive shares awarded to each factor are equal across sectors, then when the faster growing factor is the more (less) mobile of the two, the faster growing sector experiencing the greatest (least) factor substitutability. Jones (1971) analyses a restricted version of the three-factor, two-commodity case, the restriction being that although three factors of production are employed, only two enter into the production of any one commodity. This still allows the straitjacket of the factor price equalization theorem to be removed, while preserving most of the simple methods of analysis characteristic of the two-by-two model. Our analysis of child labour use in an open economy uses the basic structure of the three-by-two model developed by Jones (1971). Some theoretical papers have studied the efficacy of trade sanctions as a policy of reducing the incidence of child labour. Kar and Guha Khasnobis (2003) have examined that trade reform policies (both globalization and sanctions) affect both demand and supply of child labour and the equilibrium child wage rate by affecting simultaneously the production pattern and household labour market decisions. When individuals are characterized by a degree of relative risk aversion, the demand for child labour rises, while that of supply declines with a tariff cut only under certain conditions. In the model of Kar and Guha Khasnobis (2003), child wage rate is determined by a general discounting factor which is treated as an endogenous variable, being interpreted as the equilibrium level of effort exerted by a child labourer. Since a child may exert varying levels of effort at work and at school, the exact value of the variable is determined by the interaction of demand and supply for child labour. In their model, demand for child labour comes from import competing sector as well as from the sector producing intermediate good for this sector. In particular, they have determined the demand for child labour originating in the small open economy and compare that against child labour supply determined by household decisions. Thus, their model determines the equilibrium child wage rate endogenously, as a fraction of the equilibrium adult wage. Only under certain conditions, they tend to show that demand for child labour rises while the supply of child labour falls with a tariff cut. On the other hand, the supply of child labour increases after the imposition of import

60

4 Survey of Literature

ban. The significance of this result was not, however, explored further by Kar and Guha Khasnobis (2003). The paper of Bandyopadhyay and Bandyopadhyay (2005) provides a general equilibrium analysis of the effects of changes in terms of trade on child labour and the welfare of the unskilled households that supply child labour. The effect of the terms of trade on child labour depends critically on the pattern of substitutability (or complementarity) in the excess demand functions between the export good and the non-traded good. They argue that the welfare effect of terms of trade changes depends on the timing of the change. Under substitutability between the non-traded good and the export good, utility of the unskilled families must rise if the terms of trade improvement occur in period-1, while the utility may fall if the change occurs in period-2. Their findings have important policy implications. First, they show how trade liberalization (or trade policies in general) may affect the child labour problem as well as the utility of families supplying it. Second, the results suggest that if a sanction is effective in reducing child labour (i.e. if we have substitutability), it may be beneficial for the unskilled families to have these sanctions pre-announced and to go into effect in the future. Thus, the effect of trade sanction is analysed in a dynamic framework. Finally, a balanced budget policy of taxing the income of skilled families and subsidizing the education of unskilled families, must reduce child labour without any impact on aggregate welfare, argue Bandyopadhyay and Bandyopadhyay (2005). This is because such a policy has no trade-offs vis-a-vis child labour, as the tax (on the skilled) actually accentuates the reduction of child labour that will be obtained by providing the subsidy. Compared to a policy of trade sanction, such a policy does not have any adverse welfare effects. However, if foreign aid, say, is available to finance the educational subsidy towards unskilled families, that would still be a superior outcome, as the distortionary incentive effect associated with taxation shall be absent in this case. However, in case of education subsidy being financed by foreign capital inflow, the repatriation in the subsequent period may introduce additional costs in the form of reverse capital flows and may dampen national welfare. Since most developing countries having a high incidence of child labour are exporters of unskilled labour-intensive goods, Bandyopadhyay and Bandyopadhyay (2005) argued that a trade sanction for this economy will lower the relative price of the unskilled labour-intensive good, and from the familiar StolperSamuelson relationship between the product prices and factor price, this will tend to lower unskilled wage and raise skilled wage. A less developed economy suffers from the twin problems of adult unemployment and the presence of child labour. In such an open economy, the moot policy question is to examine the effects of imposing trade sanctions on products using child labour on the unemployment of adult workers and on the supply of child labour, and the competitive general equilibrium model of Gupta (2002) tries to address this issue. In his model, there is (i) unemployment in the adult labour market, plus an endogenous labour supply function, and (ii) there exists a third sector employing child labour, again with an endogenous supply function. Capital is assumed to perfectly mobile between all the three sectors and the supply function of labour in the two labour

4.5 General Equilibrium Models

61

markets includes both the wage rates as arguments. Some important conclusions of the model can be noted: (i) In contrast with the existing view that unemployment in the adult labour market is one of the important causes of the existence of the child labour market, in this model, the supply of child labour is shown to vary inversely with the level of unemployment in the adult labour market. The result is also interesting in the context of the proposition advanced by Basu (2000) who argues that imposing a minimum wage in the adult labour market would cause adult unemployment to rise, and this may aggravate the problem of child labour, as the adults also send their children to work to supplement their income; (ii) the rise (fall) in the price of the product produced by the capital-(labour) intensive adult labour sector raises the unemployment rate in the adult labour market and lowers the size of the child labour market; and (iii) the policy of imposing trade sanctions on the products of child labour is shown to reduce unemployment in the adult labour market and this may raise the welfare of the country but does not necessarily reduce the supply of child labour. Alternative policies like protection for the capital-intensive import competing sector and/or taxing the labour-intensive sector may result in the reduction of the supply of child labour. It may be noted that policy-induced changes in resource allocation and the resultant changes in the composition of output have important effects on the use of child labour as an input vis-à-vis adult labour in the general equilibrium framework, but the analyses in such models do not reflect on the factors which generate child labour as a result of parental choice. The interdependence between sectors and the resource reallocation effects following parametric changes get altered when the structure of the economy is modified to accommodate features of a real economy. For example, when the structure of general equilibrium framework is modified with special consideration to the interaction between the adult labour market and the child labour market, some interesting results follow. Consider four sectors in the economy—an urban formal sector, the urban informal sector and the rural sector (all employing adult labour), and a child labour sector. With positive and diminishing marginal productivity of each input, under constant returns to scale, the first three sectors use adult labour and capital as inputs, and the child labour sector uses only child labour as an input to produce either traded or non-traded goods. In such a framework, Gupta (2003) has investigated the general equilibrium effects of economic expansion (capital accumulation) on the wage rate of the child workers and on the size of the child labour market. An increase in capital stock, K, lowers the size of the child labour market and raises the wage rate there. Since it is assumed that the income effect on the demand for child labour using product is zero, and child labour sector does not use capital as an input, capital accumulation does not have any effect on the demand for child labour. Since the urban sector is assumed to be more capital intensive, following capital accumulation, the size of the formal sector’s labour use is enhanced and that of the informal sector is reduced with the result, size of the child labour market is reduced and the equilibrium wage rate goes up. Second, as Gupta (2003) argues, ‘if the child labour sector produces an inferior (normal) good, then capital accumulation raising total income lowers (raises) the demand for the product and hence lowers (raises) the demand for child labour, so that the reduction in the size of the child labour market

62

4 Survey of Literature

will be higher and the effect on the wage rate will be indeterminate in the case of an inferior good’ (Gupta 2003). However, the equilibrium wage rate is increased by larger amount and the effect on the size of the child labour market is indeterminate when the child labour sector produces a normal good. Moreover, in case where the demand for the product using child labour is price-elastic, an economic expansion induced by domestic capital accumulation may result in be ‘immiserizing’ growth in the sense of a diminution of the post-growth level of national welfare below that of the pregrowth level, and such a tendency towards immiserization would definitely be strengthened when ‘the additional capital stock is owned by the foreign country and its entire rental income is repatriated’ (Gupta 2003). Further, it may be noted that since the rental rate on capital as well as the total wage income of the adult workers, remain unaltered, the effect on national income would be determined by the effect on income generated by the child labour sector; again since the size of the child labour market contracts due to such capital expansion, its total product declines and hence price of that good is increased. If the demand for the product (which uses child labour) is price-elastic, national income is reduced, and this is strengthened when foreign-owned capital repatriates. Using a Harris and Todaro (1970) framework with a non-traded intermediate product, Chaudhuri (2001) has sought to examine the implications of foreign capital inflow in a small open economy on the welfare and urban unemployment. His model has two urban sectors and a rural sector, where the rural wage rate is flexible but the wage rates in the two urban sectors are exogenously given and exceed the rural wage rate. The first urban sector (sector Z) is tariff-protected producing import competing manufacturing intermediate input (e.g. fertilizer, pesticides or pump sets for irrigation) with the help of labour, capital and a non-traded intermediate input, which is produced in the second urban sector (Y sector) with the help of labour and capital. The rural sector (X sector) produces the exportable with labour and the non-traded intermediate input. The main results of his model are: (a) ‘An inflow of foreign capital is welfare improving under certain conditions and reduces urban adult unemployment’; (b)‘When the non-traded intermediate product is solely used in the export sector of the economy, an inflow of foreign capital income with full repatriation of foreign capital income improves the welfare of the economy and lowers the urban unemployment of labour if and only if the protected import competing sector is less capital intensive than the vertically integrated export sector’; and (c) ‘An inflow of foreign capital with full repatriation of foreign capital income reduces the welfare of the economy and raises the urban unemployment of labour when the non-traded intermediary is only used in the protected import competing sector’ (Chaudhuri 2001). Chaudhuri (2003a) finds ‘that under reasonable conditions, trade sanctions may produce perverse effect on the incidence of child labour while a liberalized trade policy is effective in bringing down the prevalence of the evil in the system’. Similar results also obtain from Chaudhuri (2003b, 2006, 2007, 2010). It has been argued by Chaudhuri and Gupta (2004) that the effect of trade liberalization on the incidence of child labour crucially depends on the relative factor intensities of the two sectors of the economy—i.e. if the tariff-protected import competing sector is capital intensive, a reduction of import tariff may lower the supply of child labour. With slight modifications in the utility

4.5 General Equilibrium Models

63

function of the representative adult worker as used by Chaudhuri and Gupta (2004), Ghosh (2007) has assumed that the worker derives utility from the consumption of final goods and children’s leisure. Interestingly, this modification gives just the opposite result—if the tariff-protected import competing sector is more capital intensive than the export sector, a reduction in import tariff may have an adverse effect on the incidence of child labour—the opposite result would obtain when the intensity ranking between the import competing and the export sector is reversed. Chaudhuri and Dwibedi (2007), while accommodating the case of foreign capital inflow, also seek to investigate the effectiveness of an education subsidy policy and a lump sum tax on the richer people as a means of financing the subsidy on education to ameliorate prevalence of child labour. While a policy of overall economic expansion with foreign capital might lower or raise the employment of child labour depending on whether the traded sector is more or less child labour intensive than the non-traded sector, an education subsidy policy is expected to produce the desired result on child labour in both situations. The technological and institutional factors of the economy, and not the relative child labour intensities of the sectors, determine whether education subsidy be financed by lump sum taxation of the tax on the wealthier people, they argue. The moot question is why despite increases in the inflow of foreign capital in developing countries during the last two decades, poverty-induced supply of children to the labour markets in these countries continued unabated, albeit at differential rates in these economies. The importance of using the general equilibrium framework, with adequate extensions to number of sectors, introduction of factor specificities and/or factor market distortions has been to study the interactions between sectors and understand the underlying mechanisms that exposure to different parametric shocks in the form of alternative policy interventions (trade or non-trade) tend to generate within an economy. The framework, therefore, extends the microeconomic foundations based on different forms of choices, and tries to portray the working of the economy within the frame of general economic equilibrium. Biswajit Chatterjee and Runa Ray in a series of papers (vide Chatterjee and Ray 2007, 2008a, b, 2009, 2011, 2013, 2016a, b; Also see Ray 2014, 2015; Ray and Chatterjee 2010a, b, c) have directly examined the effectiveness of trade and non-trade policies on the incidence of child labour use in terms of two as well as three sector models of general equilibrium having structural features like sector specificity, wage rigidity, and unemployment in the adult labour market. They have also examined the possible effects of education subsidy policy on the incidence of child labour by using a structurally specified framework for developing economies. The detailed framework of their analysis and the comparative statics results are analysed in Chaps. 5 and 6 of the book.

4.6 A Step Forward In our analysis, we shall be using a specific factor general equilibrium framework to study the effectiveness of different trade and non-trade policies to control child labour employment, as well as their impacts on different other macroeconomic vari-

64

4 Survey of Literature

ables, like employment, national welfare, trade balance and so on. This comprises an extension of the specific factor model for the closed economy framework by Ronald Jones. In our two-sector model of Chap. 5, child labour, adult labour and capital are taken as three factors of production. Child labour is specific in export sector and capital is specific in import competing sector of the economy. However, wage rates of child labour and adult labour are correlated. In the three sector version of Chap. 6, we consider a competitive general equilibrium model of a small open, less developed economy, which is divided into one rural and two urban sub-sectors. Adult labour in this model is assumed to be perfectly mobile between all sectors, but child labour is perfectly mobile within informal sectors, and capital is a specific factor of production in the import competing Y sector. Factor market distortion in this model is captured by the existence of a factor price differential between urban and rural sectors. The trade policy in the form of protectionism in import competing sector of the economy, and non-trade policy in terms of economic expansion, either by an increase in domestic capital stock, or by an increase in adult labour endowment is captured in this model. Our exercises regarding the effectiveness of alternative trade and non-trade policies on the incidence of child labour employment depart from the analysis of Bandyopadhyay and Bandyopadhyay (2005) in the following ways: (i) We have treated child labour as a specific factor in two-sector model, which has not been so in the work of Bandyopadhyay and Bandyopadhyay; (ii) The model of analysis of Bandyopadhyay and Bandyopadhyay is based on a two-period dynamic analysis whereas our analysis is entirely static in nature; (iii) In their model, households are heterogeneous differentiated by skills, but in our model skillwise division of labour has not been introduced; (iv) In both the analysis (theirs, as well as ours), it has been shown that trade sanctions on the export good produced by child labour may not reduce child labour, but effectiveness of trade sanction policies are totally different in two models; (v) Although they as well as we analyse the role of education subsidy in eliminating child labour problem, the resultant analysis differs between the two models. In the three sector versions of our model, we make structural departures from models of Chaudhuri and Gupta (2004) in the following ways: (i) No specific factor is introduced in Chaudhuri and Gupta (2004), but our model introduced factor specificity; (ii) Although child and adult labour are two different factors of production with different wage rates, in the model of Chaudhuri and Gupta, price cost equality condition does not mention this difference; (iii) In our model, we have constructed two separate labour endowment equations instead of using effective labour endowment equations of Chaudhuri and Gupta; (iv) Following, Chaudhuri and Gupta (2004), in our model, the aggregate supply function of child labour in functional form can be written as   ∂LSC ∂LSC ∂LSC ∂LSC ∂LSC < 0, > 0, > 0, < 0, >0 LSC = f WA , r, K, L, t , ∂r ∂K ∂L ∂t ∂WA

4.6 A Step Forward

65

In our model, child labour supply function depends on the tariff rate and the relation is positive. But in Chaudhuri and Gupta (2004), it depends on the tariff revenue which varies positively with the tariff rate only if the import demand function is inelastic; (v) We assume that government does not redistribute the tariff revenue as income transfer to the worker. Here, we differ from Chaudhuri and Gupta, where each worker receives a fraction of the tariff revenue as income transfers from the Government. The results of our exercises in Chaps. 5 and 6 connote to the conclusion that use of trade policy interventions to minimize the incidence of child labour may not yield the desired impact and worsen national welfare while a set of non-traded policies may be more suitable to the task of eliminating the prevalence of child labour, particularly in developing countries.

References Ahmed, I. (1999). Getting rid of child labour. Economic and Political Weekly, pp. 1815–1822. Ambreen, F. (2017). The effect of globalization and credit market imperfections on the incidence of child labour. International Journal of Social Economics. Bacolod, M. P., & Ranjan, P. (2004). Why children work, attend school, or stay idle: The roles of ability and household wealth. Economic Development and Cultural Change, 56(4), 791–828. Baland, J. M., & Robinson, J. A. (2000). Is child labour inefficient? Journal of Political Economy, 108(4), 663–679. Bandyopadhyay, S., & Bandyopadhyay, S. (2005). Trade and child labour: A general equilibrium analysis. IZA discussion papers, No. 1514. Institute for the Study of Labour (IZA). Basu, K. (1999). Child labour: Cause, consequence, and cure, with remarks on international labour standards. Journal of Economic Literature, 37(3), 1083–1119. Basu, K. (2000). The intriguing relationship between adult minimum wage and child labour. The Economic Journal, 110(462), C50–C61. Basu, K., & Van, P. H. (1998). The economics of child labour. American Economic Review, 88(3), 412–427. Bhagwati, J. (1995). Trade liberalization and ‘fair trade’ demands: Addressing the environment and labour standards issues. World Economy, 18, 745–759. Bhalotra, S. (2000). Is child work necessary? Discussion paper, no. 26, STICERD. London School of Economics. Biggeri, M., Mehrotra, S., & Sudarshan, R. (2009). Child labour in industrial outworker households in India. Economic and Political Weekly, pp. 47–56. Burra, N. (2005). Crusading for children in India’s informal economy. Economic and Political Weekly, pp. 5199–5207. Chandrasekher, S., & Mukhopadhyay, A. (2006). Primary education as a fundamental right: Cost implications. Economic and Political Weekly, pp. 3797–3804. Chatterjee, B., & Ray, R. (2007, September–December). Growth, national welfare and trade balance in a small open economy: A general equilibrium analysis. Artha Vijnana, XLIX(3 & 4), pp. 203–222. Chatterjee, B., & Ray, R. (2008a, December). A three sector model with child labour and adult market unemployment: Impact of trade restriction on national welfare and trade balance. Artha Beekshan, 17(3).

66

4 Survey of Literature

Chatterjee, B., & Ray, R. (2008b, January & July). Trade policy and impact on child labour supply—A two sector general equilibrium analysis. Journal of Quantitative Economics, 6(1 & 2) (New Series). Chatterjee, B., & Ray, R. (2009, September). Incidence of child labour, factor market distortion and export barrier by rest of the world: Impact on national welfare and trade balance in a two sector general equilibrium analysis. Artha Beekshan, 18(2). Chatterjee, B., & Ray, R. (2011, November). Economic expansion and incidence of child labour supply (with Runa Ray). Indian Journal of Quantitative Analysis, XXIV–XXV (1 & 2). Chatterjee, B., & Ray, R. (2013, September). Trade restriction, adult unemployment and incidence of child labour: A three sector general equilibrium analysis. Artha Vijnana, Journal of the Gokhale Institute of Politics and Economics, LV (3), pp. 239–251. Chatterjee, B., & Ray, R. (2016a, July). Impact of trade vs. non-trade policies on the incidence of child labour, a two-sector general equilibrium framework. Foreign Trade Review, pp. 287–297. Chatterjee, B., & Ray, R. (2016b). Impact of trade restriction on child labour supply and the role of parents’ utility function: A two sector general equilibrium analysis. In M. Roy & S. S. Roy (Eds.), International trade and international finance. Springer. Chaudhuri, D. P. (1996). A dynamic profile of child labour in India 1951–91. New Delhi: ILO. Chaudhuri, S. (2001). Foreign capital inflow, non-traded intermediary, urban unemployment and welfare in a small open economy: A theoretical analysis. The Pakistan Development Review, 40(3), 225–235. Chaudhuri, S. (2003a). Trade sanctions and child labour: Quest for an alternative policy. Mimeo, University of Calcutta. Chaudhuri, S. (2003b). How and how far to liberalize a small open economy with informal sector and factor market distortions. Journal of International Trade and Economic Development, 12(4), 403–428. Chaudhuri, S. (2006). Trade liberalization in agriculture in developed nations and incidence of child labour in a developing economy. Bulletin of Economic Research, 4. Chaudhuri, S. (2010). Incidence of child labour, informal sector and economic liberalisation. Revisiting the Informal Sector. Chaudhuri, S., & Dwibedi, J. K. (2007). Foreign capital inflow, fiscal policies and incidence of child labour in a developing economy. The Manchester School, 75(1), 17–46. Chaudhuri, S., & Gupta, M. R. (2004). Child labour and trade liberalization in a developing economy. The Japanese Economic Review, 55(2), 201–211. Christopher, S., & Kaboor, A. (2003). Child labour and its correlates. In R. K. Sen & A. Dasgupta (Eds.), Problems of child labour in India. Deep and Deep Publications. Cigno, A., Rosati, F. C., & Guarcello, L. (2002). Does globalization increase child labour? World Development, 30(9), 1579–1589. Dixit, A. (2000). Comment on “A transaction cost politics analysis of international child labour standards. In A. V. Deardonff & R. Stern (Eds.), Social dimensions of U.S. trade policies. Dreze, J. (2004). Mid-day meals and children’s rights. Economic and Political Weekly, 19, 1937–1938. Duraisamy, M. (2000). Child schooling and child work in India. Econometric Society World Congress (2000). Contributed papers 0837, Econometric Society. Edmonds, E. V. (2002, February 23/24). Reduziert die Globalisierung die Kinderarbeit [Globalization and the economics of child labor]? Neue Zürcher Zeitung, p. 29. Edmonds, E. (2005). Child labour in the global economy. Journal of Economic Perspectives, 19(1), 199–220. Edmonds, E., & Pavcnik, N. (2002). Does globalization increase child labour? Evidence from Vietnam. NBER working paper, No. 8760. Emerson, P. M., & Souza, A. P. (2002). The inter-generational persistence of child labour. Unpublished paper. Eswaran, M. (1996). Fertility, literacy and the institution of child labour. The University of British Columbia (CRESP) discussion paper: DP No.36.

References

67

Fallon, P., & Tzannatos, Z. (1998). Child labor: Issues and directions for the World Bank. World Bank, Mimeo. Ghosh, A. (2007). Incidence of child labour and trade liberalization: Comments. Unpublished paper. Grootaert, C., & Kanbur, R. (1995). Child labour: An economic perspective. International Labour Review, 134(2), 187–203. Grossman, G. M. (1981). Partially mobile capital: A general approach to two-sector trade theory. International Finance section discussion paper, G-81–01. Princeton University. Gupta, M. R. (2002). Trade sanctions, adult unemployment and the supply of child labour: A Theoretical analysis. Development Policy Review, 20(3), 317–332. Gupta, M. R. (2003). Child labour and economic development: A theoretical analysis. Arthaniti, 2, 97–107. Harris, J. R., & Todaro, M. P. (1970). Migration, unemployment and development: A two sector analysis. American Economic Review, 60, 126–142. Hill, J. K., & Mendez, J. A. (1983). Factor mobility and the general equilibrium model of production. Journal of International Economics, 15(1), 19–25. ILO. (2017). Global estimates of child labour: Results and trends, 2012–2016. Geneva. Jafarey, S., & Lahiri, S. (2002). Will trade sanctions reduce child labour? The role of credit markets. Journal of Development Economics, 68(1), 137–156. Jayachandran, U. (2001, September 1). Taking schools to children: Bhonga Shalas. Economic and Political Weekly, pp. 3347–3350. Jensen, P., & Nielsen, H. S. (1997). Child labour or school attendance? Evidence from Zambia. Journal of Population Economics, 10(4), 407–424. Jones, R. W. (1965). The structure of simple general equilibrium models. The Journal of Political Economy, 73(6), 557–572. Jones, R. W. (1971). A three-factor model in theory, trade, and history. In J. Bhagwati et al. (Eds.), Trade, balance of payments and growth: Essays in honour of Charles P. Kindleberger. Kar, S., & Guha Khasnobis, B. (2003). Wage determination of child labour & effect of trade reform. Mimeo, Centre for Studies in Social Sciences. Khera, R. (2006). Mid-day meals in primary schools, achievement and challenges. Economic and Political Weekly, pp. 4742–4750. Kulshreshta, C. J. (1994). Indian child labour. New Delhi: Uppal Publishing House. Lahiri, S., & Sylwester K. (2005). Education expenditure and school participation—A theoretical and empirical analysis. In S. Lahiri & P. Maiti (Eds.), Economic theory in a changing world. Oxford University Press. Lieten, G. K. (2000). Children, work and education-I: General parameters. Economic and Political Weekly, pp. 2037–2043. Maluccio, J. A. (2003). Education and child labour: Experimental evidence from a Nicaraguan conditional cash transfer program. In P. F. Orazem, G. Sedlacek, & Z. Tzqnnatos (Eds.), Child labour and education in Latin America. Washington, DC: Inter-American Development Bank & World Bank. Maskus, K. E. (1997). Should core labour standards be imposed through international trade policy? World Bank, Policy research working paper, No. 1817. Mukherjee, D., & Bhanu Sinha, U. (2009). Attitude to schooling, wage premium and child labour. Indian Growth and Development Review, 2(2), 113–125. Mussa, M. (1982). Imperfect factor mobility and the distribution of income. Journal of International Economics, 12(1–2), 125–141. Patrinos, H., & Psacharopoulos, G. (1997). Family size, schooling and child labour in Peru: An empirical analysis. Journal of Population Economics, 10(4), 387–405. Pouliot, W. (2006). Introducing uncertainty into Baland and Robinson’s model of child labour. Journal of Development Economics, 79, 264–272. Ranjan, P. (1999). An economic analysis of child labour. Economic Letters, 64(1), 99–105.

68

4 Survey of Literature

Ranjan, P. (2001). Credit constraints and the phenomenon of child labour. Journal of Development Economics, 64(1), 81–102. Ravallion, M., & Wodon, Q. (2000). Does child labour displace schooling?: Evidence from behavioral responses to an enrolment subsidy. Economic Journal, 110(462), 158–176. Ravi, A. (2001). Combating child labour with labels: Case of Rugmark. Economic and Political Weekly, pp. 1141–1147. Ray, R. (1999). Poverty, household size and child welfare in India. University of Tasmania, Mimeo. Ray, R. (2000). Analysis of child labour in Peru and Pakistan: A comparative study. Journal of Population Economics, 13, 3–20. Ray, R. (2001). Simultaneous analysis of child labour and child schooling: Comparative evidence from Nepal and Pakistan. ASARC working Papers 2001–10. Australia South Asia Research Centre. Ray, R. (2002). The determinants of child labour and child schooling in Ghana. Journal of African Economics, 11(4), 561–590. Ray, R. (2014). Chapter 5, Import restrictions, capital accumulation and use of child labour—A general equilibrium analysis. In A. N. Ghosh & A. K. Karmark (Eds.), Analytical issues in trade, development and finance, essays in honour of Biswajit Chatterjee. India: Springer. Ray, R. (2015). Trade versus non-trade policy in a two sector general equilibrium framework. Business and Economics Journal, 6(2). BEJ, an open access journal. Ray, R., & Chatterjee, B. (2010a). Impact of restrictive trade policy on adult unemployment, welfare and the incidence of child labour—A three sector general equilibrium analysis. Journal of Quantitative Economics, 8(1), 148–161. Ray, R., & Chatterjee, B. (2010b). Education subsidy, adult unemployment and the incidence of child labour in an open economy: A three sector general equilibrium analysis. The Indian Economic Journal, 57(4), 37–59. Ray, R., & Chatterjee, B. (2010c). Child labour and restrictive trade policy—A three sector general equilibrium framework. Asian Economic Review, Journal of The Indian Institute of Economics, 52(1), 179–190. Ray, R., & Chatterjee, B. (2012). Trade and child labour—A review of theoretical literature. Journal of International Economics, 3(1), 93–113. Sahin, S., & Ghosh, A. N. (2016). Effect of ban on exports containing child labour in a dynamic model in presence of imperfect monitoring. Foreign Trade Review, 51(1), 26–45. Sarkar, J., & Sarkar, D. (2012, June). Why does child labour persist with declining poverty? Working paper #84 (Revised), NCER working paper series. Schultz, T. P. (2001). School subsidies for the poor: Evaluating the Mexican progress poverty program. Yale Economic Growth Centre, Paper no. 834. Singh, A. N. (1990). Child labour in India: Socio-economic perspective. Shipra: New Delhi. Singh, I. S. (1992). Child labour in India. New Delhi: Oxford and IBH. Singh, B. P. & Mahanty, S. (1993). Children at work: Problems and policy options. BR: Delhi. Stolper, W. F., & Samuelson, P. A. (1941). Protection and real wages. Review of Economic Studies, 9, 58–73. Udry, C. (2003, June). Child labor. Discussion paper no. 856. New Haven: Economic Growth Center, Yale University. http://www.econ.yale.edu/~egcenter/CENTER. Weiner, M. (1996). Child labour in India: Putting compulsory education on the political agenda. Economic and Political Weekly, pp. 3007–3014.

Chapter 5

Impact of Trade Versus Non-trade Policy on Child Labour in a Two Sector General Equilibrium Framework

Abstract Following the sanctions suggested by WTO, the developed nations have been advocating that trade in commodities using child labour in many developing nations is unfair, and be banned. They, therefore, advocate the use of restrictive trade policies by the rest of the world or by the developing countries themselves to restrict the use of child labour in the exporting of their products. Had it not be a fact that the use of child labour is conditioned the economic compulsions of poor parents in developing countries, this position would have been acceptable as fair on ethical and normative grounds. In the present chapter, we have considered a competitive twosector general equilibrium model of a small open less-developed economy, which exports child labour-intensive products to examine the effectiveness of two kinds of trade restrictive policies on the incidence of child labour: (i) in one case, rest of the world imposes trade restrictions on the exported product of the small open economy, which is produced using child labour and (ii) in the second other case, the policy of protectionism in import competing sector has been taken into consideration. In both Notes: I. This chapter draws from and reuses with permission the materials contained in the following papers of the author(s): (a) Biswajit Chatterjee & Runa Ray (2016), “Impact of Trade Restriction on Child Labour Supply and the Role of Parents’ Utility Function: A Two Sector General Equilibrium Analysis” in Malabika Roy and Saikat Sinha Roy (eds), International Trade and International Finance, Springer 2016. (b) Runa Ray & Biswajit Chatterjee (2010c), “Education Subsidy, Adult Unemployment and the Incidence of Child Labour in an Open Economy: A Three Sector General Equilibrium Analysis”, Vol 57, No 4, Jan–March 2010, The Indian Economic Journal, pages 37–59. (c) Biswajit Chatterjee & Runa Ray (2016), “Impact of Trade veraus Non-trade Policies on the Incidence of Child Labour, A Two-sector General Equilibrium Framework”, Foreign Trade Review, July 2016, pp. 287–297; (d) Biswajit Chatterjee & Runa Ray (2007), “Growth, National Welfare and Trade Balance in a Small Open Economy: A General Equilibrium Analysis”, Artha Vijnana, vol XLIX Nos 3 and 4, pp 203–222, September–December, 2007. (h) Runa Ray (2015), “Trade versus Non-Trade Policy in a Two Sector General Equilibrium Framework”, Business and Economics Journal 2015, Volume 6, Issue 2, 2015, BEJ, an open access journal. II. The notational and structural similarity follows from the pioneering papers on General Equilibrium Analysis by Ronald W. Jones (1965,1971), which have been used widely in the literature on trade and development. © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_5

69

70

5 Impact of Trade Versus Non-trade Policy on Child Labour …

cases, we have separately investigated the impact on demand for and supply of child labour, as well as on equilibrium level of employment. One interesting result of our work suggests that restrictive trade policy may fail to reduce the perverseness of the incidence of child labour supply. In fact, the first three propositions derived from our model indicate that due to the imposition of trade restriction on the exported product of the small open economy by the rest of the world, (a) capital owners shall gain, while both adult and child workers shall be adversely affected, but (b) the incidence of child labour supply may or may not be perversely affected, when the adult worker’s utility function is assumed to be additive separable in nature and (c) may in fact increase the supply of child labour unambiguously when the adult worker’s utility function is assumed to be Cob–Douglas in nature. This result, thus, challenges the popular view in favour of imposing trade sanctions on the import of those goods from the developed countries which are produced using child labour.

5.1 General Issues After the formation of the World Trade Organization (WTO) in 1 January 1995, the focus of discussions on trade policy has shifted to ensuring fair and free trade practices among member countries, and as sequel, the issues relating to labour standards and environmental standards, have been brought to the forefront of policy discussions at the world level. The developed nations argued that trade in commodities using child labour in many developing nations is unfair, and be banned—the children should go to school rather than to work. This position ignores the fact that in the poor developing countries, many children are sent to work because their families are too poor to send them to school. The outright ban on children from work will not necessarily ensure their enrollment to school from the next day, but the resultant loss of income will mean a further setback for the already poor families. Although the use of child labour is also prevalent in developed countries such as the USA, UK, Italy, Portugal and Spain in some measures, the suggestion that child labour use be banned in the developing countries is counterproductive as these developing nations are not in a position to take trade measures against them to eliminate the incidence of child labour employment on practical grounds. During the 1970s and 1980s, there had been proliferation of non-tariff barriers in world trade by OECD governments, driven largely by exogenous shocks (the collapse of fixed exchange rates; successive oil price hikes by the OPEC cartel) and political developments (détente) that reduced the primacy of foreign policy considerations in maintaining cooperation in trade. Such responses were considered inappropriate to the prevailing structural adjustment pressures and the successful completion of the Uruguay round revealed that the major trading nations were willing to maintain multilateral cooperation and strengthen disciplines regarding the use of non-tariff measures. Such trade negotiations were conducted to reduce barriers to trade, to agree to rules of behaviour and resolve disputes. In this context, one must distinguish between achieving an agreement, and enforcing it. The basic international trade

5.1 General Issues

71

theory suggests that economic rationale for trade restrictions from national welfare point of view, depend largely on the market power of a country, such that a small country that cannot influences international prices and therefore has little to lose and much to gain from a multilateral agreement to lower trade barriers, but the large countries may be able to change the terms of trade—the price of their exports relative to the price of imports—in their favour by restricting trade. For the world as a whole, the imposition of trade restrictions by one or more countries can, however, only reduce national welfare. Yet, pressures for protection continued to confront WTO member governments, and one of the forms in which demands for protection cropped up by the developed nations has been the imposition of trade restrictions, including sanctions, on the products exported by the developing countries on grounds of child labour clause. Use of child labour in production is seen as one way of violating labour standards, and imposing sanctions on the import of goods produced by child labour is one such social clause. The argument in favour of the clause is that the use of child labour gives the developing countries a trading advantage in labour-intensive goods and this advantage is illegitimate because child labour is socially unacceptable. Those who oppose the inclusion of the clause argue that trade sanctions will encourage protectionism, which will hurt not only workers in developing countries but also consumers in the developed countries. In labour-abundant poor countries, preventing child labour use either by legislation, or by use of restrictive trade policies may not be successful in eliminating the incidence of child labour use as such—they often end up losing their international market access or worsening trade balances and lowering the level of national welfare, while other non-trade policies may be somewhat effective in reducing this incidence. For them, the choice is to select a batch of non-trade policies vis-à-vis trade policy instruments like tariffs to tackle the issue of prevalence of child labour. In such poor developing nations, trade restrictions are particularly inappropriate tools for dealing with the challenge of child labour International pressure through the boycott of the product produced by child labour may have a limited impact. Since children pushed out of one industry can often readily move to another, a ban by importers of goods produced by children, even if there were an effective way of labelling goods, would have only a marginal impact on the total employment of the children. The moral, as well as the legal sanctions implied by a ban, may deter some employers from hiring children, but those who choose not to obey will find it easy to evade enforcement. As a practical matter, the monitoring cost required minimally to enforce a ban on child labour would be enormous, given the number of workshops, restaurants, landowners and households employing children. So long as the number of children in the labour force is as high as they are in many countries in the developing world, a legal ban on employment, is likely to have only a limited impact on child labour. Moreover, even if governments could effectively ban child labour, the consequences could be dire for those poor households who are forced to send their children for work resorting out of economic compulsions. An effective ban on child labour would make these poor households worse off. Prevalence of child labour in many activities of poor developing countries has raised a number of analytical issues. These issues include parental

72

5 Impact of Trade Versus Non-trade Policy on Child Labour …

choices (Basu 1999; Basu and Van 1998), nature of parents’ utility function (Baland and Robinson 2000; Pouliot 2006; Chaudhury and Gupta 2004), child labour and poverty (Edmonds 2005; Udry 2004; Ranjan 1999), credit constraints and financial market imperfections (Ranjan 2001; Ambreen 2017), education subsidy (Basu 2000; Baland and Robinson 2000; Udry 2004; Jafarey and Lahiri 2002) and so on. All these issues are addressed in the literature in terms of decisions of microeconomic agents using a partial equilibrium framework. Such discussions, therefore, ignore interlinkages between sectors some of which uses child labour. By contrast, general equilibrium models point to the resource allocation/relocation effects between sectors in response to changes in certain policy parameters. *There exists a rich literature which uses the general equilibrium framework of international trade in alternative versions to examine the effects of alternative economic policies trade as well as nontrade, on the use of child labour and the associated macro indicators of development like employment, national welfare, trade balance, etc. Important contributions to the literature on general equilibrium analysis have been pioneered by the seminal work of Jones (1965) on two-sector model and extended to three-sector framework with immobility of factors by Jones (1971).The framework has become integral in the subsequent research incorporating different forms of rigidities and imperfections in the operation of markets to capture realities prevalent in many developing countries. One important line of extension of this general equilibrium framework is to consider an economic sector where child labour is used in many developing countries. The contributions of Gupta (2000, 2002, 2003), Bandyopadhyay and Bandyopadhyay (2009), Chaudhury (2003, 2006, 2010), Chaudhuri and Gupta (2004) and Chaudhuri and Dwibedi (2007, 2016) are important to note in this context. We want to use the general equilibrium framework using child labour to examine the effectiveness of various trade and non-trade policies designed to curtail the incidence of child labour, particularly focusing on resource allocation effect, unemployment, welfare and growth. In a series of papers, we have focused on the different dimensions of the general equilibrium effects of trade policy effects and compare them with different non-traded policies so as to see whether the incidence of child labour necessarily declines or nor following such policy changes (vide Chatterjee and Ray 2007, 2008a, b, 2009, 2011, 2013, 2016a, b, and Ray and Chatterjee, 2010a, b, c, 2012; Ray 2014, 2015). The following sections of this chapter focus on the two-sector version of the general equilibrium framework as have been developed in our abovementioned papers.

5.2 Child Labour in a Two-Sector General Equilibrium Framework1 In this section, we consider a competitive two-sector general equilibrium model of a small open (typically a less developed economy), which exports products using 1 Refer

to Chatterjee and Ray (2016a, b), (2007), Ray and Chatterjee (2010c) and Ray (2015) for details.

5.2 Child Labour in a Two-Sector General Equilibrium Framework

73

child labour. Child labour is specific to the export sector and capital is specific to import competing sector and adult labour is mobile between the sectors. In this section, we have derived the supply function of child labour by using different kinds of utility function of the adult worker. One interesting result of our work suggests that trade restrictive policy may fail to reduce the perverseness of the incidence of child labour supply. This result, thus, challenges the popular view in favour of imposing trade sanctions on the import of those goods from the developed countries which are produced using child labour. We omit separate discussions on the impacts of restrictive trade policies on national welfare since for such a small open economy, such impacts on welfare are unambiguously negative, and the effects of such policies on trade balance are ambiguous in nature. The Model The assumptions of the model are as follows: (i) We consider a small open economy. (ii) The economy consists of two sectors and three factors of production. Sector 1 produces an exportable commodity (X) with the help of adult labour and child labour. In particular, we assume that Rest of the World (ROW) imposes trade restrictions in such a way that exporters of the small open economy have to pay tax t on per unit export of X, resulting in reduction of the effective producer’s price of X. Sector 2 is import competing sector of the economy. It produces importable commodity (Y) with the help of adult labour and capital. Thus, adult labour in our model is perfectly mobile between the sectors but child labour and capital are specific factors of production. Child labour is specific in our model in contrast to the model by Chaudhuri and Gupta (2004), and it is specific in use in the export sector and capital is specific in import competing sector of the economy. (iii) We assume that adult labour is a perfect substitute for child labour. It is assumed that an adult labour is equivalent to b number of child worker where b > 1. Each adult worker earns a wage WA . The child wage rate WC must be (WA /b) when adult wage rate is WA . (iv) Production functions exhibit constant returns to scale with diminishing marginal productivity to each factor. (v) We assume that all inputs are fully employed and all markets are perfectly competitive. The stock of capital and adult labour force are exogenously given. (vi) The prices of the traded goods X and Y are given internationally, due to our assumption of a small open economy. (vii) Each firm maximizes profit. (viii) In our model, the demand for child labour originates from the general equilibrium structure of the small open economy and supply from the household comparison of parental utilities from sending the child to work or to school. The following symbols will be used in the formal presentation of the model: aLa i aLC X

Adult labour-output ratio in the ith sector, i = X, Y Child labour-output ratio in the X sector

74

5 Impact of Trade Versus Non-trade Policy on Child Labour …

aKY θij

Capital-output ratio in the Y sector Distributive share of the ith input in the jth sector, i = La , Lc , K and j = X, Y Proportion of the ith input employed in the jth sector, i = La , Lc , K λij and j = X, Y World price of the ith good, i = X, Y Pi t Ad-valorem tariff rate on the export of X WA   Adult wage rate A Child wage rate WC = Wβ r Rate of return on capital L Adult labour endowment Aggregate supply of child labour LSC K Capital stock of the economy ∧ Proportional change X Domestic production of exportable Y Domestic production of importable The General Equilibrium Analysis Given the assumptions of perfectly competitive markets, the following equations display the zero profit conditions for the economy:   aLaX WA + aLC X WA /β = PX (1 − t)

(5.1)

aLaY WA + aKY r = PY

(5.2)

Child and adult labour are two different factors of production with different wage rates, WA and WA /β. The full employment conditions for the economy are aLaX X + aLaY Y = L

(5.3)

  aLC X X = LSC = LSC WA , r, K, L, t

(5.4)

aKY Y = K

(5.5)

In our model, we have constructed two separate labour endowment equations instead of using effective labour endowment equation as was attempted by Chaudhuri and Gupta (2004). There are five endogenous variables: WA , r, X, Y and LSC and five Eqs. (5.1–5.5). The parameters in the model are: PX, PY, β, L, K, t. We note that the system possesses decomposition property since the unknown input prices WA (hence WC ) and r can be solved from the price system alone independent of the output system. Once the factor prices are known factor coefficients, aij ’s are also known. Y is solved from Eq. (5.5). Substituting the value of Y in (5.3), we shall solve for X. LSC will be solved from (5.4).

5.3 Comparative Static Exercises

75

5.3 Comparative Static Exercises In this section, we shall discuss the effectiveness of trade restrictive policy on the incidence of child labour. In the first place, following Maskus (1997), the impact of stringent trade restriction in our model is captured by an increase in the tariff rate that the rest of the world imposes on the exported product of the small open economy. Second, we discuss the effects of protectionism in the import competing sector of the economy on the selected endogenous variables.

5.3.1 Imposition of Stringent Trade Restriction by Rest of the World on the Exported Product of the Small Open Economy (A) Effect on Factor Price and Child Labour Supply To find out the impact of restrictive trade policy on child labour supply, we differentiate Eq. (5.4) with respect to tariff rate t and the following expression is obtained: ∂LSC dWA ∂LSC dr ∂LSC dK ∂LSC dL ∂LSC dLSC = + + + + dt ∂r dt ∂K dt ∂L dt ∂t ∂WA dt

(5.6)

Now, we shall first discuss the impact on factor prices. Taking total differentiation of (5.1), we get aLaX dWA + WA daLaX + aLcX d(WA /β) + (WA /β)daLcX = − PX dt ˆ A + θLaX aˆ LaX + θLcX W ˆ A + θLcX aˆ LcX = −dt or, θLaX W

(5.7)

Now, the cost minimization condition of the producer entails that   daLcX /daLaX = − WA /WC or, WC daLcX + WA daLaX = 0 or, θLcX aˆ LcX + θLaX aˆ LaX = 0

(5.8)

ˆ A (θLcX + θLaX ) = −dt ∴ From (5.7), we get W A

ˆ = −dt (since θLcX + θLaX = 1) or, W

(5.9)

WC = (WA /β)

(5.10)

Since,

76

5 Impact of Trade Versus Non-trade Policy on Child Labour …

∴ dWC = (1/β)dWA ˆC=W ˆ A = −dt < 0 ∴ W

(5.11)

Differentiating (5.2) we get aLaY dWA + WA daLaY + aKY dr + rdaKY = 0 ˆ A + θLaY aˆ LaY + θKY ˆr + θKY aˆ KY = 0 or,θLaY W

(5.12)

From condition of cost minimization, θLaY aˆ LaY + θKY aˆ KY = 0Z

(5.13)

∴ From (5.12), we can write ˆ A + θKY ˆr = 0 θLaY W or r = (θLaY /θKY )dt > 0 

(5.14)

Proposition 1 Imposition of trade restriction on the exported product of the small open economy by the rest of the world adversely affect employment of both adult and child workers, and capital owners will gain. The intuitive explanation behind Proposition 1 is straightforward: Due to the fall in effective producer’s price in the exportable sector, they will offer less to the factors of production engaged in the sector. It indicates a shock in labour (both adult and child) market. On the other hand, the price of import competing good remains unchanged. Hence to ensure normal profit producers, one must pay more to the capital owners. Differentiating (5.4) we get ∂LSC ∂LSC ∂LSC A dr + dt dW + ∂WA   ∂r  ∂t 

dLSC =  S

or, Lˆ C =



∂LSC WA ˆA+ W ∂WA LSC

∂LSC r ˆr + ∂r LSC

(5.15)

∂LSC t ˆ A + Erˆr + Etˆt ≷ 0 (5.16) ˆt = EWA W ∂t LSC

Proposition 2 If we assume adult worker’s utility function to be additive separable in nature, then imposition of trade restriction on child labour using exported product by rest of the world, may or may not have perverse impact on incidence of child labour supply.

5.3 Comparative Static Exercises

77

5.3.2 Alternative Specification of Utility Function and Effect on Child Labour Supply The result will be different if there is a change in the form of utility function. Let us consider the following specific Cob–Douglas type of utility function:  C U = AXa Yb / LSC where A > 0, 0 < a, b, c < 1

(5.17)

It satisfies all the standard properties and it is homogeneous of degree (a + b − c) where (a + b − c) ≤ 1 or ≥ 1 The adult worker maximizes the utility function subject to the budget constraint  Px(1 − t)X + PY Y =



 WA S Lc + WA L + rK /L β

(5.18)

The Lagrangian expression for the constrained optimization problem is

 A

   S C  W a b S A Lc + W L + rK /L + λ PX (1 − t)X + PY Y − M = AX Y / LC β (5.19) The First-order Conditions (FOCs) for maximization are  −c + λPX (1 − t) = 0 δM/δX = Aa Xa−1 Yb LSc

(5.20)

 −c δM/δY = AbXa Yb−1 LSc + λPY = 0Z

(5.21)

 −c−1   δM/δLSc = −AcXa Yb LSc − λ WA /β /L = 0Z

(5.22)

 δM/δλ = Px(1 − t)X + PY Y −



 WA S A Lc + W L + rK /L = 0 β

(5.23)

From FOCs, we get   A W /β LS X = (a/c) ( L ) c /Px(1 − t)  A  W /β)LSC Y = (b/c) ( /P L

Y

Substituting the values of X and Y in the budget constraint, we get    WA /β LSc PX (1 − t)(a/c) /Px(1 − t) L

(5.24)

78

5 Impact of Trade Versus Non-trade Policy on Child Labour …

    A  S WA /β LSc W /β Lc rK S + WA + + PY (b/c) /PY Lc /PY = L L L     rK β 1 + LW A β LWA + rK c S = A ∴ Lc =  a b W [(a + b) − c] + c − L1 c

(5.25)

This is the aggregate supply function of child labour. We shall now analyse its properties.   log LSc = log β + log c + log LWA + rK − log WA − log{(a + b) − c} ∴

1 1 ∂LSc L −rK  0 LSc ∂r LWA + rK

(5.28)

1 ∂LSc r = >0 S ∂K A Lc LW + rK

(5.29)

1 ∂LSc WA = >0 S ∂L Lc LWA + rK

(5.30)

Hence, the aggregate supply function in functional form can be written as   ∂LSC ∂LSC ∂LSC ∂LSC LSC = f WA , r, K, L , where > 0, > 0, >0 < 0, ∂r ∂K ∂L ∂WA

(5.31)

Now, due to imposition of stringent trade restriction on the exported product of the small open economy by the rest of the world, there will be change in aggregate supply of child labour. To see how it changes, we differentiate (5.31). dLSC =

∂LS ∂LS ∂LSC ∂LS dWA + C dr + C dK + C dL A ∂r ∂K ∂L ∂W

(5.32)

In our model dK = dL = 0, ∴ dLSC = S or, Lˆ C =



∂LSC WA ∂WA LSC



ˆA+ W



∂LSC r ∂r LSC



∂LSC ∂LSC A dr dW + ∂r ∂WA

(5.33)

ˆr > 0

Proposition 3 Due to the imposition of trade restriction on the exported product of the small open economy which uses child labour, a paradoxical result is obtained—supply of child labour goes up unambiguously .

5.3 Comparative Static Exercises

79

(B) Effect on Composition of Output We shall first try to find out the impact on output of various sectors due to restrictive trade policy. In the specific factor model,2  A  ˆ −r >0 aˆ La Y = −θKY σY W   A (5.34) ˆ −r 0 ˆ = −ˆaKY = −θLaY σY W Y

(5.35)

Thus, the import competing sector will expand. Now, differentiating (5.3) we get XdaLaX + aLaX dX + YdaLaY + aLaY dY = dL ˆ + λLaY aˆ LaY + λLaY Y ˆ =0 or, λLaX aˆ LaX + λLaX X  A    ˆ = σY W ˆ − ˆr (λLaY /λLaX ) < 0 ∵ aˆ Lax = 0 or, X

(5.36)

Thus, the output of the export sector will contract. Proposition 4 Due to the imposition of trade restriction by rest of the world on the exported product of the small open economy, export sector will contract and the import competing sector will expand. The intuitive explanation can be given as follows: To produce per unit Y, now, less unit of capital will be required. This is obvious because capital has become more costly as the rental rate has increased. Now, capital is specific in Y production. So, full employment of capital must result in the higher output of Y and to support the additional production of Y, more unit of adult labour will be switched from X to Y production. Thus, we can say that the export sector will contract. Alternatively, as there is fall in effective producer’s price, so the producers of X will be less willing to produce the commodity. Hence, there will be a fall in output of X.

2 See

Appendix 1 for detail derivation.

80

5 Impact of Trade Versus Non-trade Policy on Child Labour …

5.3.3 Alternative Policy Option for Child Labour Problem—Effect of Protectionism in Import Competing Sector Instead of trade restriction on exported product, now suppose the government decides to protect the import competing sector of the economy. Due to protectionism in import competing sector, the adult worker’s budget constraint becomes  PX X + PY (1 + t)Y =



 WA S A Lc + W L + rK /L β

(5.37)

Proceeding like the earlier case, the supply function of child labour now becomes α

(WA ) α−1 + 1



(β L) α−1

WA /β L

1  α−1  rK 

L

LSc = α   A  α−1 α α W /β α−1 α−1 (Px) + {PY (1 + t)} − L

(5.38)

It will follow all the properties assumed in 5.31. Moreover, differentiating (5.38) with respect to tariff rate we get 1 ∂LSc = − LSc ∂t



1 α

α

(Px) α−1 + {PY (1 + t)} α−1 −



WA /β L

α  α−1



α 1 {PY (1 + t)} α−1 > 0 α−1 (5.39)

Higher is the rate of protection in import competing sector of the economy, higher is the engagement of adult labourers in that sector. Ceteris paribus, child labour supply must rise in the export sector of the economy. Thus, the aggregate child labour supply function in functional form can be written as   ∂LSC ∂LS ∂LS ∂LS ∂LS < 0, C > 0, C > 0, C > 0, C > 0 LSC = f WA , r, K, L, t , A ∂r ∂K ∂L ∂t ∂W (5.40) One notable difference of our model with Chaudhuri and Gupta (2004) lies in the specification of our child labour supply function, which in their model, depends on the tariff revenue which varies positively with the tariff rate only if the import demand function is inelastic. Now, we try to investigate the effect of protectionism on various macroeconomic variables. The zero profit conditions for the economy are rewritten in the form

5.3 Comparative Static Exercises

81



WA β

= Px

(5.41)

aLa Y WA + aKY r = PY (1 + t)

(5.42)

aLa X W + aLC X A

To find out the impact of trade restriction on child labour supply, we differentiate Eq. (5.24) with respect to tariff rate t and the following expression is obtained: ∂LSC dWA ∂LS dr ∂LS dK ∂LSC dL ∂LSC dLSC = + C + C + + A dt ∂r dt ∂K dt ∂L dt ∂t ∂W dt

(5.43)

(A) Effect on Factor Prices Now, we shall first discuss the impact on factor prices. Since PX remains unchanged, so there will be no change in adult, as well as in child wage rate due to protectionism. To find out the effect on rental rate, differentiating Eq. (5.41) we get aLaY dWA + WA daLaY + aKY dr + rdaKY = PY dt or, θLaY aˆ LaY + θKY ˆr + θKY aˆ KY = PY dt

(5.44)

From condition of cost minimization, θLaY aˆ LaY + θK aˆ KY = 0

(5.45)

θKY ˆr = PY dt or, ˆr = (PY /θKY )dt > 0

(5.46)

Proposition 5 Due to protectionism in import competing sector, only the capital owners will be benefited. There will be no change in adult as well as child wage rate. Due to constancy of price in the export sector, neither wage rate will be affected due to protectionism. Hence, the entire gain of increased price in import competing sector of the economy will be channelized to the capital owner. (B) Effect on Child Labour Supply Since there are no changes in the wage rate, stock of capital and adult labour endowment, therefore (5.42) can be written as ∂LSC dr ∂LSC dLSC = + dt ∂r dt ∂t  S  S ∂LC ∂LC ∴ dLSC = r ˆr + dt > 0 ∂r ∂t

(5.47) (5.48)

Hence, the policy of protectionism unambiguously increases the incidence of child labour supply in our model.

82

5 Impact of Trade Versus Non-trade Policy on Child Labour …

Proposition 6 Due to protectionism in import competing sector where child labour is not used, supply of child labour goes up unambiguously. In our model, child labour supply is influenced by several factors. In case of protectionism in import competing sector, only two determinants (r and t) which have one-to-one correspondence with child labour supply, have been affected. Hence, the increase in rental rate and tariff rate increases child labour supply. (C) Effect on Composition of Output From (5.35),  A  ˆ = −ˆaKY = −θLaY σY W ˆ − ˆr = θLaY σY ˆr > 0 Y

(5.49)

Thus, the import competing sector will expand. As obtained from (5.36),  A  ˆ − ˆr (λLaY /λLaX ) ˆ = σY W X = −σY ˆr(λLaY /λLaX ) < 0

(5.50)

Proposition 7 Due to protectionism in import competing sector, output of this sector will be expanded at the cost of the other sector. Since the price of the import competing goods is increased, so profit-maximizing producers will be more willing to produce more output of Y. Since capital is specific in Y production, the supply of capital will act as a constraint in the path of expansion of import competing sector. Hence, producers will try to use less units of capital in per unit Y production. Also, there is a rise in price of capital leaving the wage rate of the adult workers unchanged, which will again motivate producers to minimize the use of capital and increase the use of adult labour in per unit Y production. Thus, in our model, we see that aˆ KY = −θLa Y σY ˆr < 0

(5.51)

aˆ La Y = θKY σY ˆr > 0

(5.52)

Full employment of capital indicates expansion of the import competing sector. Since total adult labour supply is constant and there is no barrier in inter-sectoral labour mobility, so labourers will be released from X production and will be absorbed in Y production. Hence, the output of X falls.

5.4 Concluding Remarks

83

5.4 Concluding Remarks In this chapter, we have considered a competitive two-sector general equilibrium model of a small open economy, which exports products using child labour. We have examined the effectiveness of two kinds of trade restrictive policies on the incidence of child labour: (i) in one case, rest of the world imposes trade restrictions on the exported product of the small open economy, which is produced using child labour and (ii) in the second other case, the policy of protectionism in import competing sector has been taken into consideration. In both cases, we have separately investigated the impact on demand for and supply of child labour, as well as on equilibrium level of employment. One interesting result of our work is that trade restrictive policy may fail to reduce the perverseness of the incidence of child labour supply. In fact, the first three propositions derived from our model indicate that due to the imposition of trade restriction on the exported product of the small open economy by the rest of the world, (a) both adult and child worker will be adversely affected and capital owners will gain, but (b) incidence of child labour supply may or may not be perversely affected, if the adult worker’s utility function is assumed to be additive separable in nature and (c) may in fact increase the supply of child labour unambiguously if the adult worker’s utility function is assumed to be Cob–Douglas in nature. This result thus challenges the popular view in favour of imposing trade sanctions on the import of those goods from the developed countries which are produced using child labour. There are several laws regulating the employment of the children in various occupations. Any legislation totally prohibiting child labour, would virtually amount to inflicting on these children an unbearable suffering. What, therefore, sounds more sensible and pragmatic is to eliminate or minimize the impact of adverse and undesirable conditions attendant to child labour and to promote healthy working conditions. It was observed in some cases that a child rarely knew the terms of contract between him and his employer and the rate of remuneration, as the contract was entered into between the parent and the employer. In some other cases, the child knew his wages, but it was paid by the employer directly to the parents. The existence of such practices creates a doubt whether the terms of contract were really opened. Imposition of trade restriction on the exportable commodities of the developing countries that are produced using child labour is the most controversial and hotly debated policy suggested by representatives of developed countries in this context in the WTO meeting. Trade restrictions are a particularly inappropriate tool for dealing with the challenge of child labour, although international pressure through the boycott of the product produced by child labour may have limited impact. The moral, as well as the legal sanctions implied by a ban, may deter some employers from hiring children, but those who choose not to obey will find it easy to evade enforcement. So long as the number of children in the labour force are as high as they are, a legal ban on employment, though helpful, is likely to have only a limited impact on child labour.

84

5 Impact of Trade Versus Non-trade Policy on Child Labour …

References Ambreen, F. (2017). The effect of globalization and credit market imperfections on the incidence of child labour. International Journal of Social Economics. Baland, J. M., & Robinson, J. A. (2000). Is child labour inefficient? Journal of Political Economy, 108(4), 663–679. Bandyopadhyay, S., & Bandyopadhyay, S. C. (2009). Trade and child labor: A general equilibrium analysis. Emerging Markets Finance and Trade. Basu, K., & Van, P. H. (1998). The economics of child labour. American Economic Review, 88(3), 412–427. Basu, K. (1999). Child labour: Cause, consequence, and cure, with remarks on international labour standards. Journal of Economic Literature, 37(3), 1083–1119. Basu, K. (2000). The intriguing relationship between adult minimum wage and child labour. The Economic Journal, 110(462), C50–C61. Chatterjee, B., & Ray, R. (2007). Growth, national welfare and trade balance in a small open economy: A general equilibrium analysis. Artha Vijnana, XLIX(3, 4), 203–222. Chatterjee, B., & Ray, R. (2008a). A three sector model with child labour and adult market unemployment: Impact of trade restriction on national welfare and trade balance. Artha Beekshan, 17(3). Chatterjee, B., & Ray, R. (2008b). Trade policy and impact on child labour supply—A two sector general equilibrium analysis. Journal of Quantitative Economics, New Series, 6(1, 2). Chatterjee, B., & Ray, R. (2009). Incidence of child labour, factor market distortion and export barrier by rest of the world: Impact on national welfare and trade balance in a two sector general equilibrium analysis. Artha Beekshan, 18(2). Chatterjee, B., & Ray, R. (2011). Economic expansion and incidence of child labour supply (with Runa Ray). Indian Journal of Quantitative Analysis, XXIV–XXV (1, 2). Chatterjee, B., & Ray, R. (2013). Trade restriction, adult unemployment and incidence of child labour: A three sector general equilibrium analysis. Artha Vijnana Journal of The Gokhale Institute of Politics and Economics, LV (3), 239–251. Chatterjee, B., & Ray, R. (2016a). Impact of trade versus non-trade policies on the incidence of child labour, a two-sector general equilibrium framework. Foreign Trade Review, July 2016 (pp. 287–297). Chatterjee, B., & Ray, R. (2016b). Impact of trade restriction on child labour supply and the role of parents’ utility function: A two sector general equilibrium analysis. In M. Roy & S. S. Roy (Eds.), International Trade and International Finance, Springer. Chaudhuri, S. (2003). Trade sanctions and child labour: Quest for an alternative policy. Mimeo: University of Calcutta. Chaudhuri, S., & Gupta, M. R. (2004). Child labour and trade liberalization in a developing economy. The Japanese Economic Review, 55(2), 201–211. Chaudhuri, S. (2006). Trade liberalization in agriculture in developed nations and incidence of child labour in a developing economy. Bulletin of Economic Research. Chaudhuri, S., & Dwibedi, J. K. (2007). Foreign capital Inflow, fiscal policies and incidence of child labour in a developing economy. The Manchester School, 75(1), 17–46. Chaudhuri, S. (2010) Incidence of child labour, informal sector and economic liberalisation. Revisiting the Informal Sector. Edmonds, E. (Winter, 2005). Child labour in the global economy. Journal of Economic Perspectives, 19(1), 199–220. Gupta, M. R. (2000). Wage determination of a child worker: A theoretical analysis. Review of Development Economics, 4(2), 219–228. Gupta, M. R. (2002). Trade sanctions, adult unemployment and the supply of child labour: A theoretical analysis. Development Policy Review, 20(3), 317–332. Gupta, M. R. (2003). Child labour and economic development: A theoretical analysis. Arthaniti, 2, 97–107.

References

85

Jafarey, S., & Lahiri, S. (2002). Will trade sanctions reduce child labour? The role of credit markets. Journal of Development Economics, 68(1), 137–156. Jones, R. W. (1965). The structure of simple general equilibrium models. The Journal of Political Economy, 73(6), 557–572. Jones, R. W. (1971). A three-factor model in theory, trade, and history. In J. Bhagwati et al. (Eds.), Trade, Balance of Payments and Growth: Essays in honour of Charles P. Kindleberger. Maskus, K. E. (1997). Should core labour standards be imposed through international trade policy? World Bank, Policy Research Working Paper, No. 1817. Pouliot, W. (2006). Introducing uncertainty into Baland and Robinson’s model of child labour. Journal of Development Economics, 79, 264–272. Ranjan, P. (1999). An economic analysis of child labour. Economic Letters, 64(1), 99–105. Ranjan, P. (2001). Credit constraints and the phenomenon of child labour. Journal of Development Economics, 64(1), 81–102. Ray, R., & Chatterjee, B. (2010a). Impact of restrictive trade policy on adult unemployment, welfare and the incidence of child labour—A three sector general equilibrium analysis. Journal of Quantitative Economics, The Indian Econometric Society, 8(1), 148–161. Ray, R., & Chatterjee, B. (2010b). Education subsidy, adult unemployment and the incidence of child labour in an open economy: A three sector general equilibrium analysis. The Indian Economic Journal, 57(4), 37–59. Ray, R., & Chatterjee, B. (2010c). Child labour and restrictive trade policy—A three sector general equilibrium framework. Asian Economic Review, Journal of The Indian Institute of Economics, 52(1), 179–190. Ray, R., & Chatterjee, B. (2012). Trade and child labour—A review of theoretical literature. Journal of International Economics, Institute of Public Enterprise, 3(1), 93–113. Ray, R. (2014). Chapter 5, Import restrictions, capital accumulation and use of child labour—A general equilibrium analysis. In A. N. Ghosh & A. K. Karmark (Eds.), Analytical Issues in Trade, Development and Finance, Essays in Honour of Biswajit Chatterjee. Springer India. Ray, R. (2015). Trade versus non-trade policy in a two sector general equilibrium framework. Business and Economics Journal, 6(2) (BEJ, an open access journal). Udry, C. (2004). Child labour. Yale School of Management Working Papers, ysm 399, Yale School of Management.

Chapter 6

Effectiveness of Trade and Non-trade Policies on the Incidence of Child Labour—A Three-Sector General Equilibrium Framework

Abstract This chapter considers a 3 × 3 × 3 competitive general equilibrium framework of a small open economy which exports product produced by child labour. In the first part, we abstract from unemployment to begin with and assume that adult labour market maintains full employment. Our objectives have been to investigate the

Note: I. This chapter draws from and reuses with permission the materials contained in the following papers of the author(s): (a) Runa Ray (2014): “Import Restrictions, Capital Accumulation and Use of Child Labour—A General Equilibrium Analysis”, Chapter 5 of the book edited by Ambar Nath Ghosh and Asim K. Karmark (eds.), Analytical Issues in Trade, Development and Finance, Essays in Honour of Biswajit Chatterjee, Springer India, 2014, as Part I of Chapter 6 of the present book; (b) Runa Ray and Biswajit Chatterjee (2010a), “Child Labour and Restrictive Trade Policy—A Three Sector General Equilibrium Framework”, Vol. 52, No. 1, April 2010, Asian Economic Review, Journal of The Indian Institute of Economics, Pages 179–190, as Part I of Chapter 6 of the present book; (c) Runa Ray and Biswajit Chatterjee (2010b), “Impact of Restrictive Trade Policy on Adult Unemployment, Welfare and the Incidence of Child Labour—A Three Sector General Equilibrium Analysis,” Journal of Quantitative Economics, The Indian Econometric Society, vol. 8(1), pages 148–161, January 2010 as Part II of Chapter 6 of the present book; (d) Biswajit Chatterjee and Runa Ray (2013), “Trade Restriction, Adult Unemployment and Incidence of Child Labou: A Three Sector General Equilibrium Analysis”, Artha Vijnana Journal of The Gokhale Institute of Politics and Economics, Vol. LV No. 3, September 2013 Page No. 239–251 as Part II of Chapter 6 of the present book; (e) Runa Ray and Biswajit Chatterjee (2010c),“Education Subsidy, Adult Unemployment and the Incidence of Child Labour in an Open Economy: A Three Sector General Equilibrium Analysis”, Vol. 57, No. 4, Jan–March 2010, The Indian Economic Journal, pages 37–59, as Part III of Chapter 6 of the present book; (f) Biswajit Chatterjee and Runa Ray (2016), “Impact of Trade versus Non-trade Policies on the Incidence of Child Labour, A Two-sector General Equilibrium Framework”, Foreign Trade Review, July 2016, pp. 287–297; (g) Biswajit Chatterjee and Runa Ray (2016), “Impact of Trade Restriction on Child Labour Supply and the Role of Parents’ Utility Function: A Two Sector General Equilibrium Analysis “ in Malabika Roy and Saikat Sinha Roy (eds.), International Trade and International Finance, Springer 2016. (h) Biswajit Chatterjee and Runa Ray (2007), “Growth, National Welfare and Trade Balance in a Small Open Economy: A General Equilibrium Analysis,” Artha Vijnana, vol. XLIX, Nos. 3 and 4, pp. 203–222, September–December, 2007. © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_6

87

88

6 Effectiveness of Trade and Non-trade Policies …

effectiveness of trade and non-trade policies in the presence of child labour problem and also to examine the impacts on national welfare and trade balance of such an economy. In particular, in this chapter, we have examined the effectiveness of three different kinds of policies on the incidence of child labour. Among these three, one is trade policy and the rest are non-trade policies. First part of this chapter reveals that economic expansion via increase in adult labour endowment is ineffective in curtailing the incidence of child labour. The other two policies seem to be efficient in this ground. However, any policy designed to control child labour incidence must have an impact on welfare and trade balance situation of the economy. Trade balance situation worsens in case of economic expansion via increase in domestic capital stock. However, the other non-trade policy leaves it as unchanged. But the impact is ambiguous for the case of trade policy. Again, welfare impact is positive only in case of economic expansion via increase in adult labour endowment but ambiguous in the other two cases. Hence, from the above analysis, we observe that none of the policies simultaneously can have favourable impact on child labour supply, trade balance and welfare of the small open economy. In part two of the chapter, we introduce unemployment problem in the adult labour market on one hand, and retain the existence of a child labour market on the other hand. The economy is divided into one rural and two urban sub-sectors. Four inputs are used in the model among which three are specific in nature. The representative adult worker in this model not only supplies his own labour but also sends his children out to work. Factor market distortion in this model is captured by the existence of a factor price differential between urban and rural sectors. The model is used to analyse the effects of imposition of various forms of trade restrictions on the unemployment of adult workers on the one hand and on the incidence of child labour on the other. Alternative trade policy prescriptions to combat child labour as well as adult market unemployment problem have been analysed. An interesting result obtained from the exercise of trade policy suggests that supply of child labour is shown to vary inversely with the level of unemployment in the adult labour market. In part three, we focus on an important non-trade policy, namely education subsidy to find its effects on the incidence of child labour.

(i) Biswajit Chatterjee and Runa Ray (2008), “A Three Sector Model With Child Labour And Adult Market Unemployment: Impact Of Trade Restriction On National Welfare and Trade Balance” Artha Beekshan, vol. 17, No. 3, December 2008. II. The notational and structural similarity follows from the pioneering papers on General Equilibrium Analysis by Ronald W. Jones (1965, 1971), which have been used widely in the literature on trade and development.

6.1 Child Labour in a Three Sector General Equilibrium Framework …

89

Part I

6.1 Child Labour in a Three-Sector General Equilibrium Framework Without Any Adult Unemployment This part of the present chapter considers a 3 × 3 × 3 competitive general equilibrium framework of a small open less developed economy which exports product produced by child labour. In this chapter, we abstract from unemployment to begin with and assume that adult labour market maintains full employment. Our objectives have been to investigate the effectiveness of trade and non-trade policies in the presence of child labour problem and also to examine the impacts on national welfare and trade balance of such an economy. In particular, we have examined the effectiveness of three different kinds of policies on the incidence of child labour. Among the three, one is trade policy and the rest are non-trade policies. This chapter reveals that economic expansion via increase in adult labour endowment is ineffective in curtailing the incidence of child labour. The other two policies seem to be efficient in this respect. However, any policy designed to control child labour incidence must have an impact on welfare and trade balance situation of the economy. None of the policies can surely lead to an improvement in the trade balance situation of the economy. Trade balance situation worsens in case of economic expansion via increase in domestic capital stock. However, the other non-trade policy leaves it as unchanged. But the impact is ambiguous for the case of trade policy. Again, welfare impact is positive only in case of economic expansion via increase in adult labour endowment but ambiguous in the other two cases. Hence, from the above analysis, we observe that none of the policies simultaneously can have favourable impact on child labour supply, trade balance and welfare of the small open economy. The Model The economy we consider is a small, open less developed economy with the following characteristics: (1) There are three sectors in the economy: (a) Among the three, two are informal sectors. The informal sector is comprised of ‘plucky’ micro-entrepreneurs who choose to operate informally in order to avoid costs and effort of formal registration. (i) One of the informal sectors is the export sector producing agricultural product (X). (ii) The service sector attempts to incorporate the other informal sector into the model. The service sector produces non-traded final commodity (Z). (b) The third sector is the formal sector and it is the import competing sector producing a manufacturing commodity (Y).

90

6 Effectiveness of Trade and Non-trade Policies …

(2) Three kinds of inputs are available in the economy—adult labour, child labour and capital. (3) X and Z are produced using adult labour and child labour but Y is produced using adult labour and capital. Thus in our model, adult labour is perfectly mobile between informal and formal sectors but child labour is perfectly mobile within informal sectors and capital is specific factor of production. (4) Owing to effective wage legislation and unionization of labour, the adult wage rate in the formal sector is fixed at W* , which is greater than the competitive informal sector adult wage rate, W. (5) The adult labour allocation mechanism is of the following type: Adult workers first try to get employment in the formal manufacturing sector, and those who are unable to find employment in the said sector are automatically absorbed in any one of the informal sectors as there is complete wage flexibility in the latter sector. (6) Following Basu and Van (1998), we make the assumption of ‘substitution’ in the informal sector, which suggests that adult labour is a substitute for child labour. Gupta (2000) has done theoretical work in determining the wage rate of a child worker. In our model, it is assumed that an adult labour is equivalent to β units of child labour where β > 1. The adult labour in the informal sectors earns W, while a child labour earns Wβ . (7) Production functions satisfy constant return to scale with positive but diminishing return to each factor. (8) All inputs are fully employed. Stock of capital and adult labour force are exogenously given. (9) Owing to the small open economy, assumption prices of X and Y are given internationally. Since Z is produced and consumed domestically its price is determined within the country by demand and supply forces. (10) We assume each working family comprises one adult and certain number of child worker. The following symbols are used in the formal presentation of the model: aLa i aLc i aKY Pi Pz W W β

W* r LA LC S K

Adult labour-output ratio in the ith sector i = X, Y, Z Child labour-output ratio in the ith sector, i = X, Z Capital-output ratio in the formal sector (Y) World Price of the ith good, i = X, Y Domestically determined price of Z Adult wage rate in the informal sectors (X, Z) Child wage rate in the informal sectors (X, Z) Institutionally given wage rate in formal sector (Y ) Rate of return to capital Total adult labour endowment Aggregate supply of child labour Fixed stock of domestic capital in the economy

6.1 Child Labour in a Three Sector General Equilibrium Framework …

L1 ∧ X Y

91

Number of adult workers engaged in the informal sectors Proportionate change Domestic production of exportable Domestic production of importable.

The General Equilibrium Analysis1 A general equilibrium of the system is represented by the following set of equations: aLa X W + aLC X

W = Px β

aLa Y W ∗ + aKY r = PY (1 + t) aLa Z W + aLc Z

W = Pz β

(6.1) (6.2) (6.3)

aLa X X + aLa Y Y + aLa Z Z = LA

(6.4)

aKY Y = K

(6.5)

aLa Y W ∗ Y + rK = I

(6.6)

Z(Pz , I ) = Z with

∂Z ∂Z >0 < 0, ∂PZ ∂I

aLc X X + aLc Z D(Pz , I ) = LC S (W, L1 )with

∂LC S ∂LC S < 0, >0 ∂W ∂L1

(6.7) (6.8)

where L1 = aLa X X + aLa Z Z

(6.9)

Equations (6.1)–(6.3) are the usual price unit cost equality conditions in the three sectors of the economy. Equations (6.4) and (6.5) are the endowment equations of adult labour and capital. Total income of the formal sector consists of wage income of the formal sector workers and the rental income of the capital owners. It is shown in Eq. (6.6). The demand for non-traded final commodity (Eq. 6.7) produced by child and adult workers comes from the richer section of the society. The demand function obeys the usual price and income effect. Finally, child labour market equilibrium condition is shown by Eq. (6.8). The left-hand side reveals the demand for child labour which depends on output of X and Z. The aggregate supply function is assumed to depend on adult wage rate as well as on number of people engaged in the informal sector. If wage rate of the adult worker rises, they send less children to the job market. On the other hand, the supply of child labour varies positively with the number of

1 Refer

to Ray and Chatterjee (2010a), Ray (2014)

92

6 Effectiveness of Trade and Non-trade Policies …

the informal sector adult workers because these workers earn low wage income and send their children to work. The working of the model is described as follows: There are eight endogenous variables in the system and eight independent equations. The endogenous variables are W, r, PZ, X, Y, Z, I, LC S . The parameters in this model are: PX , PY , β, L A , K, W * . We should note that the system possesses the decomposition property since the two unknown input prices W (hence Wβ ) and r can be determined from the price system alone, independent of the output system. Once the factor prices are known, factor coefficients aij ’s are also known. PZ is to be obtained from Eq. (6.3). Since Z is non-traded, its price is determined by the demand—supply mechanism—the supply of Z is constrained by the demand for Z in the formal sector. Y can be obtained from Eq. (6.5) and substituting the value of Y and r in (6.6) we get I. Similarly, once PZ and I are determined, Z can also be obtained from (6.7). X will be solved from (6.4) and finally, from (6.8) LC S will be derived.

6.2 Comparative Static Exercises In this section, we investigate the impact of trade and non-trade policies on different macroeconomic variables. The trade policy undertaken by the domestic economy is protectionism in the country’s import competing sector, whereas the non-trade policy constitutes economic expansion or growth. (A) Impact of Trade Policy (Protectionism in Import Competing Sector) on Different Macroeconomic Variables Much of the recent policy debate and controversy surrounding globalization and the WTO has been focused on the issue of child labour in poor countries. On the one hand, opponents of market integration argue that globalization may increase the wages paid to the working children or increase the earning opportunities of children in poor economies, thereby increasing child labour. Some further suggest that rich countries should restrict the sale of goods from developing countries that lack or do not enforce child labour laws. Theoretical models by Maskus (1997) and Ranjan (2001) show that trade sanctions or import tariffs against countries that use child labour do not necessarily reduce the incidence of child labour. Alternatively, increase in household income and increased availability of schooling opportunities in low-income countries could help in reducing child labour (Basu 1999). In this chapter, we want to investigate whether trade restriction will have a perverse impact on child labour supply. Here, we are considering the policy of protectionism in import competing sector. In our model,   ∂LC S ∂LC S  > 0 (6.10) < 0,  LC S = LC S W, aLa X X + aLa X Z with ∂W aLa X X + aLa Z Z

6.2 Comparative Static Exercises

93

  ∂LsC ∂LsC dLSC dX dZ   aLa X = dW + + aLa Z ∴ dt ∂W dt dt ∂ aLa X X + aLa Z Z

(6.11)

To find out the impact of increased protectionism on child labour supply, we will first examine the impact on factor prices and composition of output. Due to protectionism, there will be no change in adult as well as child wage rate as revealed by Eqs. (6.1) and (6.2). However, rental rate will change. By differentiation and substitution, we have:   W ∗ d aLa X + aKY dr + rdaKY = PY (dt) or, θLa Y aˆ La Y + θKY rˆ + θKY aˆ KY = dt or, θKY rˆ = dt

(6.12)

[Since from the condition of cost minimization θLa Y aˆ La Y + θKY aˆ KY = 0] or, dt >0 θKY

rˆ =

(6.13)

Proposition 1 A rise in price of import competing good produces a Stolper–Samuelson effect (Stolper and Samuelson 1941) and raises the rental rate leaving the wage rates unchanged. Proposition 2 Since there is no change in adult and child wage rate, there will be no change in price of non-traded good. Now, we try to find out the impact on output of import competing sector. Differentiating (6.5) we get Yˆ = −ˆaKY In specific factor model (in our model capital is specific to import competing sector), we know that 



∗





∗

Y = −ˆaKY = −θLa Y σY (W − rˆ ) = θLa Y σY r (∵ W = 0) dt >0 = −θLaY σY θKY

(6.14)

94

6 Effectiveness of Trade and Non-trade Policies …

Proposition 3 Due to increased protectionism in import competing sector, formal sector of the economy must expand. The intuitive explanation can be stated as follows: As price of import competing good has increased, profit-maximizing producers will be interested to produce more units of Y leading to an expansion of the formal sector. Alternatively, producers will curtail the use of capital in per unit Y production as aˆ KY = −θLa Y σY rˆ < 0. To ensure full employment of capital, output of import competing sector must expand. Taking total differentiation of (6.6), we get   ∗ ∗ W Y daLa y + W aLa y dY + Kdr = dI or, 

 ∗   ∗ W YaLa y aˆ La Y + W YaLa Y Y + Krˆr = dI 

or, 

∗

W aLa Y Y

  aˆ La Y + Y + Krˆr = dI 

or, 

  ∗ W aLa Y Y aˆ La Y − aˆ KY + Krˆr = dI

or, 

   ∗ W aLa Y Y σY rˆ + Krˆr = dI ∵ aˆ La Y = θKY σY rˆ and aˆ KY = −θLa Y σY rˆ

∗ W aLa Y Y σY + Kr dt dI = >0 (6.15) θKY

Thus the income of the richer section of the society increases. Proposition 4 Due to increase in rental rate, income of formal sector increases. Differentiating (6.7), we get ∗

Z1 W aLa Y Y σY + Kr dt dZ = Z1 dI = >0 θKY

(6.16)

Proposition 5 The output of non-traded final good producing sector goes up due to operation of income effect. Finally, to find out the impact on export sector, we differentiate Eq. (6.8) and get

6.2 Comparative Static Exercises

aLa X

95

dX dY dZ daLa Y + aLa Y +Y + aLa Z =0 dt dt dt dt

or,      aLa Y Y θLa Y σY Y aLa Y θKY σY dX aL Z ZI  ∗ =− W aLa Y σY Y + Kr − − a aLa X dt θKY θKY θKY or, aLa X

  aL Z ZI  dX Y σY  = −aLa Y θLa Y + θKY − a W ∗ aLa Y σY Y + Kr dt θKY θKY

or,  aL Y Y σY aLa Z ZI  ∗ dX =− a W aLa Y σY Y + Kr − dt aLa X θKY aLa X θKY aL Y Y σY aL Z ZI Kr =− a [1 + aLa Z ZI W ∗ ] − a aLa X θKY aLa X θKY     

1 aLa Y Y σY 1 + aLa Z ZI W ∗ + aLa Z ZI Kr < 0 =− aLa X θKY

(6.17)

Thus the export sector will contract. Proposition 6 Due to protectionism in import competing sector, one of the informal sector (nontraded final good producing sector) will expand and export sector (another informal sector) will contract. The intuitive explanation is quite straight forward: as the income of the richer section of the population increases due to expansion of formal sector, the demand for non-traded good boosts up leading to an expansion of that sector. Since adult labour endowment is fixed and adult labour is used in the production of all three commodities, adult labour has to be released from export sector to the other two expanded sectors. At the same time, child labour will also be switched from this sector to the other expanded informal sector. Hence, there will be contraction of the export sector. To find out the aggregate impact on the informal sector of the small open economy, we proceed as follows: Let m = x · z, such that log m = log x + log z or, M =X +Z ∴ dM = dX + dZ =−

  ZI dt  ∗     dt aLa Y σY Y 1 + aLa Z ZI W ∗ + aLa Z ZI Kr + W aLa Y σY Y + Kr aLa X θKY θKY

96

6 Effectiveness of Trade and Non-trade Policies …         aLa Y σY Y dt 1 + aLa Z ZI W ∗ aL Z ZI Krdt 1− a − W ∗ ZI + θKY aLa X θKY aLa X        aLa Y σY Y dt 1 a Z a Krdt L Y I L Z =− − W ∗ ZI 1 − a + 1− a θKY aLa X aLa X θKY aLa X 

=−

(6.18)

Suppose aLa Z a aLC Z aLC X

(6.23)

Then

i.e., Z is adult-labour intensive commodity. In this case, informal sector will unambiguously contract. Proposition 7 Due to protectionism in import competing sector, informal sector must contract when exported product is child labour intensive in nature and the result becomes ambiguous when exported product is adult labour intensive in nature. Now, we are interested in investigating the effect on child labour supply. From 4.10, we have obtained   ∂LC S ∂LC S dX dZ dLC S  aLa X = dW +  + aLa Z dt ∂W dt dt ∂ aLa X X + aLa Z Z

6.2 Comparative Static Exercises

97

  dX dZ ∂LC S  aLa X + aLa Z =  (∵ d W = 0) dt dt ∂ aLa X X + aLa Z Z Now,   dX dZ aL a X + aL a Z dt dt  

aL Z ZI  ∗  aLa X  aLa Y σY Y 1 + aLa Z ZI W ∗ + aLa Z ZI Kr + a W aLa Y σY Y + Kr =− aLa X θKY θKY

(6.24)

aL Y σY Y =− a 0 θKY

(6.31)



(6.32)

Now, consumption demand function for exportable is given by XD = XD {Px , PY (1 + t), N }

(6.33)

It is well known from the theory of consumer behaviour that demand function is homogeneous of degree zero in all prices and income. We choose PY (1 + t) as numeraire, and thus the demand function for exportable is reduced to the form:   Px N , 1, XD = XD PY (1 + t) PY (1 + t)   ∂XD N Px ∂XD  > 0 (6.34) < 0,  ,p = , = XD p, N PY (1 + t) PY (1 + t) ∂p ∂ PY (1+t) Similarly, the demand function for importable is   ∂YD N ∂YD  >0 > 0,  YD = YD p, , N PY (1 + t) ∂p ∂ PY (1+t)

Differentiating Eq. (6.30), we get d (TB) = Px dX 0 − PY (1 + t)dY 0 − PY Y0 dt

(6.35)

100

6 Effectiveness of Trade and Non-trade Policies …

= Px [dX − dX D ] − PY (1 + t)[dY D − dY ] − PY Y0 dt = Px dX + PY (1 + t)dY − Px dX D − PY (1 + t)dY D − PY Y0 dt

(6.36)

 

−Px  aLa Y Y σY 1 + aLa Z ZI W ∗ + aLa Z ZI Kr dt aLa X θKY ⎤ ⎡   PY (1 + t)θLa Y Y σY dt N ∂X ∂XD ⎦ d − Px ⎣ D dp +  + N θKY ∂p PY (1 + t) ∂

∴ d (TB) =

PY (1+t)

⎤  ∂YD N ∂YD ⎦ − PY Y0 dt ⎣ d − PY (1 + t) dp +  N ∂p PY (1 + t) ∂ P (1+t) ⎡



Y

 

P (1 + t)Y σ Y dt PY (1 + t)θLa Y Y σ Y dt −Px  aLa Y Y σ Y 1 + aLa Z ZI W ∗ + aLa Z ZI Kr + Y + aLa X θKY θKY θKY ⎤ ⎡     ∂YD ∂XD N N ⎦ d d + PY (1 + t)  − ⎣Px  N N PY (1 + t) PY (1 + t) ∂ ∂

=

PY (1+t)

PY (1+t)

∂Y ∂X − Px D dp − PY (1 + t) D dp − PY Y0 dt ∂p ∂p   

PY (1 + t)θLa Y Y σ Y dt Px aLa Y Y σ Y 1 + aLa Z ZI W ∗ + aLa Z ZI Kr + =− aLa X θKY θKY ⎤ ⎡     ∂YD ∂XD N N ⎦ d d + PY (1 + t)  − ⎣Px  N N PY (1 + t) PY (1 + t) ∂ ∂ PY (1+t)

PY (1+t)

Px dt Px dt ∂Y ∂X + Px D + PY (1 + t) D − PY Y0 dt ≷ 0 ∂p PY (1 + t)2 ∂p PY (1 + t)2

(6.37)

Thus impact on trade balance is ambiguous. Proposition 10 Impact of trade restrictive policy on trade balance is ambiguous. To explain the impact on trade balance, we will separately study the effect on value of export and value of import. Both of these will be changed due to operation of price effect and quantity effect. Let us first concentrate on the value of export. Since price of exportable remains unchanged, the value of exportable will be changed due to operation of only quantity effect. To find out the impact of quantity effect, we have to concentrate on net export for the country. Production of exportable goes down leading to a fall in net export. But quantity demanded for exportable will be changed due to operation of substitution and income effect. Due to substitution effect demand for exportable will rise. But the income effect becomes ambiguous due to ambiguous impact on national welfare. Hence, impact on value of export becomes ambiguous. Due to increase in relative price of importable, value of import goes up. Concentrating on the quantity effect, we find that production of importable goes up and substitution effect will lead to a fall in import demand. Both of these will lead to fall in value of importable. However, income effect is ambiguous leading to ambiguous impact on volume of import. Hence, we conclude that stringent protectionism will not necessarily improve the trade balance situation of the economy.

6.2 Comparative Static Exercises

101

Thus protectionism in the import competing country is not expected to reduce the incidence of child labour and it will not improve trade balance either. In so far as national welfare is concerned, such a policy will have ambiguous effect, i.e., not necessarily immiserizing as the standard literature on protectionism in a small open economy tend to suggest. We now move to the examination of non-traded policies on different macroeconomic variables in the same three-sector general equilibrium framework. Non-trade policy in our model is captured by economic expansion. For similar works, one may refer to Gupta (2003) and Swaminathan (1998). (B) Impact of Non-trade Policy on Different Macroeconomic Variables in a Three-Sector General Equilibrium Framework With No Adult Unemployment2 Economic expansion in our model is captured in two different ways: (a) by an increase in domestic capital stock (b) by an increase in adult labour endowment Since import restriction in the form of tariff is ruled out by assumption, in this case, Eq. (6.2) of our model is rewritten in the form aLa Y W ∗ + aKY r = PY

(6.38)

There will be no change in any other equations. Case 1: Economic Expansion by Increase in Domestic Capital Stock As the size of the capital stock of the economy swells up, the factor prices remain unaffected due to the decomposition property of the system. Proposition 11 Economic expansion by increase in domestic capital stock will not lead to any change in factor prices. There will be no change in the price of the non-traded good since there will be no change in factor prices. Proposition 12 Economic expansion by increase in domestic capital stock will not lead to any change in price of non-traded good. To find out the impact on output of different sectors, taking total differentiation of Eq. (6.5), we get aKY dY = dK or, dY = 2 Refer

1 dK > 0 aKY

to Ray (2014), Ray and Chatterjee (2010a), Chatterjee and Ray (2013, 2008) for details.

102

6 Effectiveness of Trade and Non-trade Policies …

or, dY dK = Y K

  K ∵ aKY = Y

or, 



Y =K >0

(6.39)

Proposition 13 Capital expansion will lead to enhance the size of the import competing sector. Capital is treated as specific factor in import competing sector of theeconomy and  there is no change in requirement of capital in per unit Y production ∵ aˆ KY = 0 . Hence, full utilization of the expanded capital stock must lead to an expansion of the import competing sector. Now, we want to find out the impact on output of non-traded good producing sector. Taking total differentiation of (6.6), we get dI = aLa Y W ∗ dY + rdK (∵ daLa Y = 0) or, aLA Y W ∗ dK + rdK a  KY ∗  aLa Y W = + r dK > 0 aKY

dI =

(6.40)

Taking total differentiation of (6.7), we get ∂Z dI ∂I    aLa Y W ∗ ZI = ZI + r dK = PY dK > 0 aKY aKY

dZ =

Taking total differentiation of (6.4), we get aLa X dX + aLa Y dY + aLa Z dZ = 0 or,     aLa Z ZI aLa Y dK − aLa X dX = − PY dK aKY aKY or,

(6.41)

6.2 Comparative Static Exercises

103

     aLa Z ZI aLa Y + PY dK < 0 dX = − aKY aLa X aKY aLa X

(6.42)

Proposition 14 Expansion of domestic capital stock of the economy will adversely affect the export sector. However, the non-traded good producing sector will expand. Thus, one of the informal sectors will expand and the other will contract. The intuitive explanation can be given as follows: Expansion of the formal sector will lead to increase wage income of the sector. Hence, demand for non-traded good must go up. Since total adult labour supply is assumed to be constant, export sector will contract. Now, we want to find out the impact on child labour supply. In our model, we have assumed that LC S = LC S (W, aLaX X + aLaZ Z) with

∂LC S ∂LC S < 0, >0 ∂W aLa X X + aLa Z Z

The first determinant of child labour supply is unaffected due to economic expansion. Hence,   dLSC ∂LC S dX dZ = aLa X + aLa Z dK ∂(aLa X X + aLa Z Z) dK dK

(6.43)

Now,   dX dZ aLa X + aLa Z dK dK   aLa Y aL Z ZI PY aLa Z ZI = −aLa X + PY + a aKY aLa X aKY aLa X aKY aLa Y aLa Z ZI PY aLa Z ZI PY =− − + aKY aKY aKY aLa Y =− aKY   −∂LC S aLa Y dLSC = 0 aLa x

(6.46)

Proposition 20 Expansion of adult labour endowment will expand the export sector of the economy. We find that entire increase in supply of adult labour will be absorbed in export sector since none of the other two sectors expand. With unchanged input coefficient, increased supply of adult labour increases the output of the export sector. Now, we shall find out the impact on child labour supply.   ∂LC S ∂LC S  >0 < 0,  LC S = LC S W, aLa x X + aLa Z Z with ∂W ∂ aLa x X + aLa Z Z Since there will be no change in adult wage rate ∂LC S  aLa x dX > 0 ∴ dLSC =  ∂ aLa x X + aLa Z Z

(6.47)

Proposition 21 Since the number of adult workers engaged in informal sector of the economy goes up, supply of child labour of the economy also increases. Thus, two sources of economic expansion lead to different impacts on child labour supply. Hence, we conclude that increase of domestic capital stock can be thought of as an efficient instrument for curtailing child labour incidence of an economy. B. Economic Expansion and Impact on National Welfare We now focus on the effects of economic expansion or growth on national welfare. Again we distinguish between two sources of expansion: capital accumulation or increase in the size of domestic capital stock, and exogenous increase in supply of adult labour force (say by the demographic dividend type of phenomenon). Case 1: Economic Expansion by Increase in Domestic Capital Stock Now, we want to analyse the impact of economic expansion on the welfare of the small open economy. From (6.28), we see that

106

6 Effectiveness of Trade and Non-trade Policies …

  W X + aLa Y W ∗ Y + rK + PZ Z (N ) = aLa X W + aLC X β or,   W dN = aLa X W + aLC X dX + aLa Y W ∗ dY + rdK + PZ dZ β or, dN = PX dX + aLa Y W ∗ dY + rdK + PZ dZ or,   aLa Y aL Y W ∗ aLa Z ZI PZ ZI PY dN = −PX + PY + a +r+ dK aKY aLa X aLa X aKY aKY aKY   aLa Y PY aLa Z ZI Pz Z I PY = −PX + PY + + aKY aLa X aLa X aKY aKY aKY

PY PX aLa Y + aLa Z ZI PY + =− (Pz Z I + 1) aKY aLa X aKY   PX aLa Y PY aL Z ZI PX Pz Z I + 1 − a =− + aKY aLa X aKY aLa X Thus, national welfare will improve only when        PY    PX aLa Y PY aLa Z ZI    PX   a (Pz Z I + 1) >  a a + a aLa X KY La X KY KY

(6.48)

(6.49)

If condition (6.49) is not satisfied, then there will be immiserizing kind growth in the small open economy. Proposition 22 Under economic expansion, welfare of the economy may go down. The case of ‘immiserizing growth’ (Bhagwati 1968) implies a situation where growth (due to technical progress and/or factor accumulation) leads to a sufficiently acute deterioration in the terms of trade which imposes a loss of real income outweighing the primary gain in real income due to growth itself. Harry Johnson (1965) has shown that the phenomenon of immiserizing growth, involving reduction in social welfare below the initial pre-growth level can also arise in the case of a small country without any monopoly power in trade, if technical progress occurs in a tariff protected import competing industry or if the factor in whose use this industry is intensive is augmented. However, in our model, capital is specific factor of production and no technical progress occurs. Hence, the above conditions are not satisfied in our model but we get ambiguous result on welfare.

6.2 Comparative Static Exercises

107

In our model, owing to changes in the output composition, labour reallocation between the sectors of the economy takes places, thereby affecting the aggregate wage income of the workers. Two forces operate on aggregate wage income. Since some of the low-paid workers from the export sector (one of the informal sector) will be switched to the higher paid formal sector in view of its expansion, wage income of the workers will go up implying a gain on welfare side. On the other hand, some of the workers will be recruited in non-traded final good producing sector (another informal sector) after being released from export sector. In this case, aggregate wage income of the workers will be affected not by change in wage rate since the wage rate of the two informal sectors are same, but by change in the output composition. Since one of the informal sector contracts and the other informal sector expands, the aggregate wage income of the workers will be increased due to labour reallocation effect if aggregate informal sector expands. Hence, in this case, both forces of labour reallocation effect will increase the welfare of the economy measured by the national income at world price. But if the aggregate informal sector contracts, then two opposite forces will operate on aggregate wage income and labour reallocation effect may increase or decrease the national income. Thus, the possibility of immiserizing growth is observed in case of economic expansion in a small economy. Case 2: Economic Expansion by Increase in Adult Labour Endowment Due to economic expansion, there is change in X only. Hence,   W dX = Px dX > 0 dN = aLa x W + aLC x β

(6.50)

Proposition 23 Welfare of the small open economy unambiguously expands due to economic expansion as aggregate wage income of the informal sector workers goes up. C. Effect of Economic Expansion on Trade Balance Case 1: Economic Expansion by Increase in Domestic Capital Stock Trade balance of the economy is given by TB = PX X0 − PY Y0

(6.30a)

Now, consumption demand function for exportable is given by XD = XD {Px , PY , N }

(6.51)

It is well known from the theory of consumer behaviour that demand function is homogeneous of degree zero in all prices and income. We choose PY as numeraire and thus the demand function for exportable is reduced to the form

108

6 Effectiveness of Trade and Non-trade Policies …



 Px N , 1, PY PY   N Px ∂XD ∂XD ,p = < 0,   > 0 = XD p, , PY PY ∂p ∂ N

XD = XD

(6.52)

PY

Similarly, the demand function for importable is   ∂YD N ∂YD > 0,   > 0 YD = YD p, , PY ∂p ∂ PNY

(6.53)

d (TB) = PX (dX − dX D ) − PY (dY D − dY ) ⎡ ⎡ ⎤ ⎤     N ⎦ N ⎦ ∂XD ∂YD ∂XD ∂YD ⎣ ⎣ − PY dp +   d dp +   d = PX dX + PY dY − PX ∂p PY ∂p PY ∂ PNY ∂ PNY

(6.53a) In case of economic expansion there will be no change in relative price of any commodity. Thus, substitution effect will be zero. Hence, we can write ⎡ ⎡ ⎤ ⎤ ∂Y PX ⎣ ∂XD P Y⎣ D   dN ⎦ −   dN ⎦ d (TB) = PX dX + PY dY − PY ∂ N PY ∂ N PY PY ⎡ ⎤ ∂XD ∂YD = PX dX + PY dY − ⎣p   +   ⎦dN (6.54) N ∂ PY ∂ PNY Now substituting the value of dX, dY and dN, we get 

  aLa Z ZI dK PY dK + PY aKY aLa X aKY ⎡ ⎤     aLa Z ZI PX PX aLa Y ∂X P ∂Y − ⎣p  D  +  D  ⎦dK − + Y (PZ ZI + 1) − N N a a a a La X KY KY La X ∂ ∂

d (TB) = −PX

aLa Y aKY aLa X

PY





+

PY

⎫⎤ ⎫⎤ ⎧ ⎧ ⎡ ⎡ ⎨ ∂X ⎨ ∂X PX aLa Y Px aLa Z ZI PY dK ∂YD ⎬⎦ ∂YD ⎬⎦ D D ⎣ ⎣ +   +   =− dK − 1− p  1− p  ⎩ ∂ N ⎩ ∂ N aLa X aKY aKY aLa X ∂ PN ⎭ ∂ PN ⎭ PY PY Y Y ⎫⎤ ⎫ ⎧ ⎧ ⎡ ⎬ ⎬ ⎨ ∂X ⎨ ∂X P dK ⎣ ∂Y P dK ∂Y + Y PZ Z I p  D  +  D  (6.55) 1− p  D +  D ⎦− Y ⎭ ⎭ ⎩ ∂ N ⎩ N N N aKY a KY ∂ P ∂ P ∂ P P Y

Y

Now, ⎤   ∂Y Px ∂XD PY ∂YD ∂X D D + 1 − ⎣p   +   ⎦ = 1 − ∂N ∂N ∂ PN ∂ PN ⎡

Y

Y

Y

Y

6.2 Comparative Static Exercises

109  =1−

Px X D ∂XD N PY YD ∂YD N + N ∂N XD N ∂N YD

d (TB) = −

 =1−1=0

PY dK PZ ZI < 0 aKY

(6.56)

Hence trade balance of the economy will deteriorate. Proposition 24 Due to economic expansion of the small open economy there will be adverse impact on the trade balance of the small open economy. Trade balance will deteriorate for decrease in volume of production of exportable and improve for increase in volume of production for import competing good. In income effect, we find that two opposite forces are in operation. Our exercise shows that ultimately part of depressing force will persist in income effect and all other opposite forces will cancel each other. Hence, trade balance of the economy will deteriorate. Case 2: Economic Expansion by Increase in Adult Labour Endowment From the trade balance Eq. (6.30a), we can write d (TB) = PX (dX − dX D ) − PY (dY D − dY ) ⎡ ⎡ ⎤ ⎤     ∂X ∂Y N N ∂X ∂Y D D D D ⎦ − PY ⎣ ⎦ = PX dX + PY dY − PX ⎣ dp +   d dp +   d ∂p PY ∂p PY ∂ PN ∂ PN Y

Y

In case of economic expansion, there will be no change in relative price of any commodity. Thus, substitution effect will be zero. ⎡ ⎡ ⎤ ⎤ ∂Y PX ⎣ ∂XD P Y⎣ D   dN ⎦ −   dN ⎦ ∴ d (TB) = PX dX + PY dY − PY ∂ N PY ∂ N PY PY ⎡ ⎤ ∂XD ∂YD = PX dX + PY dY − ⎣p   +   ⎦dN N ∂ PY ∂ PNY In this case, since there is no change in output of import competing sector ⎡ ⎤ ∂X ∂Y D D d (TB) = Px dX − ⎣p   +   ⎦dN N N ∂ PY ∂ PY

! ! D D Px X D ∂X PY YD ∂Y ∂N ∂N = Px dX 1 − − (6.57) XD YD N N N N = Px dX [1 − 1] = 0

(6.58)

110

6 Effectiveness of Trade and Non-trade Policies …

Proposition 25 Due to economic expansion by increase in adult labour endowment, there will be no change in trade balance of the economy. To explain the impact on trade balance, we will separately study the effect on value of export and value of import. Both of these will be changed due to operation of price effect and quantity effect. In case of economic expansion, price effect will be zero. Hence, demand for the commodities will be changed due to operation of only income effect. Now in this case production of exportable goes up. Also there is increase in demand for exportable due to operation of income effect. The first force will increase the volume of export and the later one will reduce it. To study the effect on value of import, we say that production of importable is unchanged but demand for importable goes up due to operation of income effect. Hence, value of import will be increased. Thus two opposite forces are operating on trade balance—Trade balance will improve for increase in production of exportable but increase in demand for exportable and importable will create an opposite force on it. Change in value of production exportable = Px dX

(6.59)

Change in value of demand for (exportable + importable) ⎡ ⎤     ∂X ∂Y N N D D ⎦ + PY   d = ⎣Px   d PY PY ∂ PNY ∂ PNY

! ! D D Px X D ∂X PY YD ∂Y ∂N ∂N = + dN = dN XD YD N N N N = Px dX

(6.60)

Hence, there will be no change in trade balance.

6.3 Comparison Between Alternative Policies In this chapter, we have examined the effectiveness of three different kinds of policies on the incidence of child labour. Among the three, one is trade policy and the rest are non-trade policies. This chapter reveals that economic expansion via increase in adult labour endowment is ineffective in curtailing the incidence of child labour. The other two policies seem to be efficient in this ground. However, any policy designed to control child labour incidence must have an impact on welfare and trade balance situation of the economy. None of the policies can surely lead to an improvement in the trade balance situation of the economy. Trade balance situation worsens in case of economic expansion via increase in domestic capital stock. However, the other

6.3 Comparison Between Alternative Policies

111

non-trade policy leaves it as unchanged. But the impact is ambiguous for the case of trade policy. Again, welfare impact is positive only in case of economic expansion via increase in adult labour endowment but ambiguous in the other two cases. Hence, from the above analysis, we observe that none of the policies simultaneously can have favourable impact on child labour supply, trade balance and welfare of the small open economy. Part II3

6.4 Child Labour in a Three-Sector General Equilibrium Framework in the Presence of Urban Adult Unemployment In partial modification of the 3 × 3× 3 analysed in part I of the chapter, we now extend the analysis by introducing adult unemployment. In particular, in this part of the chapter, we consider a three-sector competitive general equilibrium model of a small open, less developed economy suffering from unemployment problem in the adult labour market on one hand, and from the existence of a child labour market on the other hand. The economy is divided into one rural and two urban sub-sectors. Four inputs are used in the model among which three are specific in nature. The representative adult worker in this model not only supplies his own labour but also sends his children out to work. Factor market distortion in this model is captured by the existence of a factor price differential between urban and rural sectors.

6.4.1 The Model We consider a small open dual economy, which is broadly divided into rural and urban sector. The rural sector (sector X) produces its product (the export commodity of the economy) with adult labour and child labour. In our model, adult labour is perfectly mobile in all the three sectors and capital and child labour are specific factors of production. The adult labour in rural sector earns W A while a child labour earns β1 W A . The urban sector is further subdivided into two sub-sectors so that on the whole we have three sectors in the economy. The first sector (Y sector) within the urban sector produces the non-traded intermediary with the help of adult labour and capital. The second sector (sector Z) within the urban sector is the tariff protected import competing sector that produces its product using adult labour and a non-traded intermediary input.

3 Refer

to Ray (2014), Ray and Chatterjee (2010b), Chatterjee and Ray (2008, 2013), for details.

112

6 Effectiveness of Trade and Non-trade Policies …

In our model intermediate goods are produced and they are not directly traded, but they are required in production of import competing good. However, the wage rate in the two urban sectors is W * . It is exogenously given with W * > W A . In our model, we consider the case of factor market imperfection. The kind of factor market imperfection analysed here concerns the existence of a factor price differential between sectors. This wage differential leads to migration of labour from the rural to urban sector. There is the presence of Harris—Todaro type of urban adult unemployment. Capital and child labour are assumed to be fully employed. The rural and urban wage rates are related by the migration equilibrium condition. The complete flexibility of rural adult and child wage rates lead to the equality between the expected urban income for a prospective rural migrant family, and their actual family income in the rural sector. Unemployment in the rural sector is ruled out due to complete flexibility of wage rate. Production functions reflect CRS with diminishing marginal productivity to each other. Product markets are competitive owing to the small open economy assumption, prices of the traded goods X and Z are given internationally. Since Y is non-traded, its price is endogenously determined by the demand–supply mechanism. The supply of Y is constrained by the demand for Y. Finally, we assume that the rest of the world imposes tariff on the exported product of the small open economy, so there is a decline in effective producer’s price. Exporter of the small open economy has to pay tariff at a rate t before entering into the rest of the world. The following symbols will be used in the formal presentation of the model: Adult labour-output ratio in the ith sector i = X, Y, Z Child labour-output ratio in the X sector Capital-output ratio in the Y sector Non-traded intermediary input–output ratio in Z sector Distributive share of the ith input in the jth commodity, i = L A , L C , K, Y, j = X, Z World price of the ith commodity, i = X, Z Pi t Ad-valorem tariff rate on the export of X Domestically determined price of Y PY WA  A  Adult wage rate in the rural sector C Child wage rate in the rural sector W = Wβ ∗ W Adult wage rate in the urban sectors r Rate of return to capital K Capital stock of the economy L Total adult labour endowment Level of urban adult labour unemployment in the economy LU Aggregate supply of child labour LC ∧ Proportional change X Domestic production of exportable Z Domestic production of importable. aLa i aLC X aKY aYz θij

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

113

A general equilibrium of the system is given by the following set of equations:  aLa X W + aLC X A

WA β

 = PX (1 − t)

(6.61)

aLa Y W ∗ + aKY r = PY

(6.62)

aLa Z W ∗ + aYZ PY = PZ

(6.63)

aKY Y = K

(6.64)

aLa X X + aLa Y Y + aLa Z Z + LU = L

(6.65)

  ∂LC S ∂LC S >0 < 0, aLC X X = LC S W A , aLA X X , ∂WA ∂aLa X X

(6.66)

aLa Y W ∗ Y + aLa Z W ∗ Z WA = WA + aLa Y Y + aLa Z Z + LU β

(6.67)

aYZ Z = Y

(6.68)

Equations (6.61)–(6.63) are the usual price-unit cost equality conditions in the three sectors of the economy. Equation (6.61) shows that the rest of the world has decided to impose a tariff on the exported product of the small open economy which is produced using child labour. This policy suggests that the exporter of the small open economy has to pay tariff t per unit on their export before entering into the rest of the world. Equation (6.64) is the full employment condition of capital and Eq. (6.65) is the endowment equation of adult labour. Equation (6.66) gives the child labour market equilibrium condition. The left-hand side of Eq. (6.66) gives the demand for child labour in the economy and the right-hand side reflects the aggregate supply of child labour in the economy (Harris and Todaro 1970). Child labour supply depends on adult wage rate (Basu 2000) as well as on number of adult workers engaged in informal sector as these people due to poverty send their kids to the job market. Higher the adult wage rate, lower will be child labour supply as some adults will then decide to send their children to school instead of job market. Moreover, higher the number of people engaged in rural sector, higher will be child labour supply. The Harris–Todaro migration equilibrium condition is given by Eq. (6.67). The total income of a working family in the rural sector [the right-hand side of (6.67)] must be equal to the expected total income of the family in the urban sector (the left-hand side) in migration equilibrium. The demand–supply equilibrium condition for the intermediate input is given by Eq. (6.68). There are eight endogenous variables in the system: WA , r, PY , X , Y , Z, LC S , LU and eight independent equations. Hence, the system is solvable. The parameters in the system are Px , Pz , t, W ∗ , L, K, β. We should note that the system possess decomposition property. The three unknown factor prices W A , r and PY are determined

114

6 Effectiveness of Trade and Non-trade Policies …

from Eqs. (6.61)–(6.63) independent of the factor endowments. Once factor prices are known, factor coefficient is also determined. From (6.61), given Px , t and β, W A is determined. From (6.63) given W ∗ and PZ , PY is obtained and substituting the value of PY in (6.62), r can be solved. From (6.64) given K, output of Y is determined. Once Y is obtained, Z is automatically obtained from (6.68). The level of urban adult unemployment is then obtained from (6.67). Substituting the value of Y, Z and LU in (6.65), we shall solve for X. Finally, LC S can be solved from Eq. (6.66).

6.4.2 Effect of Imposition of Stringent Trade Restriction by Rest of the World on Exported Product of the Small Open Economy Following Maskus (1997), the impact of stringent trade restriction is captured by an increase in tariff rate that the rest of the world imposes on the exported product of the small open economy. (A) Effect on Factor Prices For this purpose, taking total differentiation of (6.61), we obtain  A  A   W W + daLC X = −Px dt aLa X d W A + W A daLa X + aLC X d β β

(6.69)

or, A

A





θLa X W + θLa X aLa X + θLC X W + θLC X aˆ LC X = −dt or, A



A



θLa X W + θLC X W = −dt (since from condition of cost minimization it follows that θLa X aˆ La X + θLC X aˆ LC X = 0) or, A



W = −dt < 0

Since W C =



WA β

  ∵ θLa X + θLC X = 1



∴

dWc 1 dWA = c W β WC

(6.70)

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

115

or, 

c

A



W = W = −dt < 0

(6.71)

Hence, both adult and child wage rate will decline due to the imposition of trade restriction on the exported product of the small open economy by the rest of the world. Now, we will try to investigate the effect on the price of non-traded intermediary good. Taking total differentiation, we get aLa Z d W ∗ + W ∗ daLa Z + aYZ dP Y + PY daYZ = 0

(6.72)

or, 

θLa Z aˆ La Z + θYZ P Y + θYZ aˆ YZ = 0 or, 

θYZ P Y = 0 (since from condition of cost minimization, we get θLa Z aˆ La Z + θYZ aˆ YZ = 0) 

⇒ PY = 0

(6.73)

Thus, there will be no effects on the price of non-traded intermediary product. To find out the effect on rental, taking total differentiation of (6.62), we get aLa Y d W ∗ + W ∗ daLa Y + aKY dr + rdaKY = 0

(6.74)

or, 

∗



θLa Y W + θLa Y aˆ La Y + θKY r + θKY aˆ KY = 0

(6.75)

(∵ from the condition of cost minimization, it follows that θLa Y aˆ LaY +θKY aˆ KY = 0) 

⇒r=0

(6.76)

Hence, the rental rate on capital will be unaffected. Proposition 26 Due to imposition of trade restriction on the exported product of the small open economy, only labourers (both adult and child) will be worse off. However, there will be no change in any other factor price.

116

6 Effectiveness of Trade and Non-trade Policies …

The intuitive explanation is straightforward: since effective producer’s price for the exported product goes down, so the profit of the producer will decline. Thus, the profit-maximizing producers in order to compensate their loss will cut down the wage bill of the workers. This is possible since the wage rate in the rural sector is assumed to be flexible. Since PZ is exogenously given and the urban wage rate is institutionally determined, there is no change in the price of non-traded product. Hence, the rental rate on capital also remains unaffected. (B) Effect on Composition of Output and Urban Unemployment Level To see how the product mix changes, we differentiate Eq. (6.64) and get aKY dY + YdaKY = dK

(6.77)

dY = 0

(6.78)

or,

Hence, there will be no change in output of non-traded good. Differentiating (6.68), we get aYZ dZ + ZdaYZ = dY

(6.79)

dZ = 0

(6.80)

or,

Thus, the output of the import competing good will also remain unchanged. Now to find out the change in output of export sector, we first have to investigate the impact on adult unemployment rate. For this, we rewrite Eq. (6.67) in the following way:    WA  aLa Y Y + aLa Z Z + LU aLa Y W ∗ Y + aLa Z W ∗ Z = W A + β

(6.81)

Differentiating it, we get         1 WA d W A aLa Y Y + aLa Z Z + LU + W A + dLU 0 = dWA + β β or,       1  1 A d W aLa Y Y + aLa Z Z + LU = − 1 + W A dLU 1+ β β or,

(6.82)

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

  dLU = (dt) aLa Y Y + aLa Z Z + LU > 0

117

(6.83)

Hence, unemployment in adult labour market will increase which is an undesirable outcome. Proposition 27 Imposition of trade restriction on exported product of small open economy, produced by child labour, aggravates the unemployment problem in the adult labour market. Due to increased wage differential between rural and urban sector, more rural workers will migrate to urban sector. However, none of the urban sub-sectors has expanded. As a result, urban unemployment magnitude increases. Now taking total differentiation of (6.65), we write aLa X dX + XdaLa X = −dLU

(6.84)

or, 

λLa X X + λLa X aˆ La X = −

dLU L

or,   1 dLU X =− L λ La X



or,      dt L − aLa X X aL X X −dt dt  1− a = =− X =− 1 − λ La X < 0 λ La X L λ La X L λ La X



(6.85)

Thus, the output of the export sector will fall. Proposition 28 Due to imposition of trade restriction on exported product of the small open economy, output of the export sector will contract. On the other hand, there will be no change in output of any urban sub-sector. Since effective producer’s price has decreased, producers will be less willing to produce the commodity. Moreover, as the wage differential between the urban and rural sector grows up, more and more rural workers will get an incentive to migrate in urban sector. Hence, the export sector will shrink. Regarding non-traded intermediary producing sector, we may argue that expansion of this sector is constrained by the endowment of capital which acts as a specific factor in Y production. Since there is no change in either the supply of capital or capital-output ratio, the output of this sector remains static. On the other hand, the demand for import competing good is derived demand in the sense that demand for it depends on how much Y is produced. As long there is no change in the production of import competing good along with

118

6 Effectiveness of Trade and Non-trade Policies …

no change in input requirement coefficient, there will not be any variation in output of Z. Rural sector contracts as rural workers migrate to urban sectors. (C) Effect on Child Labour Supply In our model, there are two determinants of child labour supply: (a) Adult wage rate—higher is the income of adult workers engaged in rural sector, lower will be their incentive to send their children to job market. (b) Number of adult workers engaged in rural sector—As rural workers send their children to job market, higher is the number of adult workers engaged in rural sector, higher is the supply of child labour. Thus, in our model, child labour supply function in functional form can be written as   ∂LC S ∂LC S   >0 LC S = LC S W A , aLa X X , < 0, ∂WA ∂ aLa X X

(6.86)

Differentiating (6.86), we derive dLSC =

  ∂LC S ∂LC S A   d aLa X X d W + ∂WA ∂ aLa X X

=

∂LC S A A ∂LC S  aLa X dX + XdaLa X W W +  A ∂W ∂ aLa X X

(6.87)

dLSC =

  ∂LC S A A ∂LC S    + a a W W + X X L X L X a a ∂WA ∂ aLa X X

(6.88)

 ∂LC S A A ∂LC S   aLa X X X ≷ 0 W W +  A ∂W ∂ aLa X X

(6.89)



or, 



or, dLSC =





Proposition 29 Due to imposition of trade restriction on exported product of the small open economy which uses child labour, impact on child labour supply becomes ambiguous. Since adult wage rate has declined, adult workers in rural sector will become worse off and they will send more children to work to compensate their income loss. On the other hand, as number of adult labourers engaged in rural sector goes down, it will lower child labour supply. Thus, two opposite forces are operating on child labour supply. Hence, the impact on aggregate supply of child labour becomes ambiguous.

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

119

6.4.3 Alternative Trade Policy Option—Effect of Protectionism in Non-traded Intermediary Producing Sector With protectionism in non-traded intermediary producing sector, Eqs. (6.61)–(6.63) of the model are rewritten in the following form: aLa X W A + aLC X

WA = Px β

(6.90)

aLa Y W ∗ + aKY r = PY (1 + t)

(6.91)

aLa Z W ∗ + aYZ PY (1 + t) = PZ

(6.92)

All symbols in the model have their earlier meanings. Here, t denotes the rate of protection provided to the non-traded intermediate producing sector. (A) Effect on Factor Prices From (6.90), we can conclude that as (Px ) will remain unaffected due to protectionism in import competing sector, adult wage rate as well as child wage rate will remain unaffected. From (6.92), we can conclude that as W ∗ and PZ remain unaffected, rise in PY will be adjusted by a fall in aYZ (see Appendix 6.2 for detail derivation). ∗





aˆ YZ = θLa Z σZ [W − P Y (1 + t)] 

= −θLa Z σZ P Y (1 + t) < 0 Differentiating (6.97), we will find out the change in rental rate. Differentiating (6.97), we get W ∗ daLa Y + aKY dr + rdaKY = PY dt or, 

θLa Y aˆ La Y + θKY r + θKY aˆ KY = dt or, 

θKY r = dt (since from condition of cost minimization, it follows that θLa Y aˆ La Y +θKY aˆ KY = 0) or, 

r=

dt >0 θKY

(6.93)

120

6 Effectiveness of Trade and Non-trade Policies …

Proposition 30 Due to protectionism in the sector producing non-traded intermediate input, only capital owners will be benefited. There will be no change in adult or child wage rate. The result is similar to Stolper–Samuelson effect. The exact Stolper–Samuelson (Stolper and Samuelson, 1941) effect does not work because of sector specificity of factors of production. Capital is specific in Y production and wage rate of adult labourers are institutionally fixed in urban areas. Hence, the gain from increase in price in this sector goes entirely to the capital owners. (B) Effect on Composition of Output and Urban Unemployment Level Differentiating (6.64), we can write aKY dY + YdaKY = 0 or,     ∗ ∗ Y = −ˆaKY = −θLa Y σY W − r = θLa Y σY r ∵ W = 0 



=







θLa Y σY dt >0 θKY

(6.94)

Thus, the protective sector will expand (see Appendix 6.1 for detail derivation). To obtain the impact on import competing sector, we differentiate Eq. (6.66). aYZ dZ + ZdaYZ = dY or 

aYZ Z(Z + aˆ YZ ) = dY or 



Z = Y − aˆ YZ = −ˆaKY − aˆ YZ =

θLa Y σY dt + θLa Z σZ P Y (1 + t) > 0 θKY 

(6.95)

Thus, the import competing sector will expand. To find out the effect on export sector, we differentiate Eq. (6.65) and get

! 



 WA aLa Y Y + aLa Z Z + LU W ∗ aLa Y dY + YdaLa Y + W ∗ aLa Z dZ + ZdaLa Z = d W A + d β !  WA  aLa Y dY + YdaLa Y + aLa Z dZ + ZdaLa Z + dLU (6.96) + WA + β

or

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

121

      WA A ∗ W aLa Y Y Y + aLa Y + W aLa Z Z Z + aˆ La Z = W + β "    #  aLa Y Y Y + aˆ La Y + aLa Z Z Z + aˆ La Z + dLU ∗









or  

W ∗ aLa Y Y aˆ La Y − aˆ KY + W ∗ aLa Z Z Z + aLa Z    #   W A " A aLa Y Y aˆ La Y − aˆ KY + aLa Z Z(Z + aLa Z ) + dLU = W + β 



or, 

W ∗ aLa Y Y σ Y r + W ∗ aLa Z Z[(Y − aˆ YZ ) + aLa Z ]    WA  aLa Y Y σ Y r + aLa Z Z{(Y − aˆ YZ ) + aˆ LaZ } + dLU = WA + β 





or,        WA WA aLa Y Y σ Y r + W ∗ − W A + W∗ − WA + aLa Z Z θLa Y σY r + σZ P Y (1 + t) β β   WA = dLU W A + β ⎡ ⎤ ⎡ ⎤   ∗ ∗ W W − 1⎦aLa Z Z θLa Y σY r + σZ P Y (1 + t) − 1⎦aLaY Y σ Y r + ⎣ ∴ dLU = ⎣ WA WA A A W + β W + β 













(6.97)  Now, we have to find out the relationship between W ∗ and W A + From migration equilibrium condition (6.65), we can write     W ∗ aLa Y Y + aLa Z Z WA   = WA + β aLa Y Y + aLa Z Z + LU

WA β



or, 

W∗ WA +

WA β

 >1

(6.98)

Hence, from (5.38) dLU > 0

(6.99)

Thus adult unemployment rate will rise. To find out the change in output of export sector, we differentiate Eq. (6.63).

122

6 Effectiveness of Trade and Non-trade Policies …

aLa X dX + aLa Y dY + YdaLa Y + aLa Z dZ + ZdaLa Z + dLU = 0 or,   aLa X dX + aLa Y Y Y + aˆ La Y + aLa Z Z(Z + aˆ La Z ) + dLU = 0 



or,     aLa X X X + aLa Y Y σ Y r + aLa Z Z (Y − aˆ YZ + aˆ La Z ) + dLU = 0 





or,        aLa X X X = − aLa Y Y σY r + aLa Z Z (Y − aˆ YZ + aˆ La Z ) + dLU 





or,      aLa Y Y σY r + aLa Z Z (Y − aˆ YZ + aˆ LaZ ) + dLU 



X =−



aLa X X

0 (C) Impact on Child Labour Supply.   CS < 0, In our model, LC S = LC S W A , aLa X X , ∂L ∂WA ∴ dLSC =

(6.103)

∂LC S ∂ (aLa X X )

>0

  ∂LC S ∂LC S A   d aLa X X d W + A ∂W ∂ aLa X X

 ∂LC S   aLa X dX < 0 =  ∂ aLa X X Proposition 33 Protective trade policy will lead to a decline in child labour supply.

(6.104)

124

6 Effectiveness of Trade and Non-trade Policies …

Child labour supply in this case will be affected only for variation in number of people engaged in rural sector. Due to shrinkage of export sector, child labour supply will be reduced.

6.4.4 Effect of Protectionism in Import Competing Sector With protectionism in import competing sector, Eqs. (6.1) and (6.3) are rewritten in the following manner: aLa X W A + aLC X

WA = Px β

aLa Z W ∗ + aYZ PY = PZ (1 + t)

(6.1b) (6.3b)

All other equations will remain unaltered. All symbols in the model have their same meanings as expressed in Sect. 6.4.1. Only t here denotes the rate of protection provided to import competing sector of the economy. (A) Effect on Factor Prices From (6.1b), we can conclude that as (Px ) will remain unaffected due to protectionism so adult and child wage rate will remain unaffected. From (6.3b), we can conclude that as adult wage rate in urban sector is fixed so protectionism must raise the price of non-traded intermediate good. Differentiating (6.3b), we get aLa Z d W ∗ + W ∗ daLa Z + aYZ dP Y + PY daYZ = PZ dt

(6.105)

or, 

θLa Z aˆ La Z + θYZ P Y + θYZ aˆ YZ = dt or, 

θYZ P Y = dt (∵ from condition of cost minimization θLa Z aˆ La Z + θYZ aˆ YZ = 0) 

∴ PY =

dt >0 θYZ

To see the effect on rental, differentiating Eq. (6.2), we obtain that aLa Y d W ∗ + W ∗ daLa Y + aKY dr + rdaKY = dP Y or

(6.106)

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

125





θLa Y aˆ La Y + θKY r + θKY aˆ KY = P Y or 



θKY r = P Y =

dt >0 θYZ

or 



r=

dt PY = >0 θKY θYZ θKY

(6.107)

Proposition 34 Due to protectionism in import competing sector capital owners will be benefited but there will be no change in adult as well as child wage rate. However, price of non-traded intermediate input will be increased. The initial impact of increased price of import competing good falls on the nontraded intermediary producing sector. Since adult wage rate is exogeneously given in the urban sector, PY must rise due to increase in protectionism. The next obvious consequence is gain of capital owners. (B) Effect on Composition of Output and Urban Unemployment Level Differentiating (6.4), we get  ∗  Y = −ˆaKY = −θLa Y σY W − r = θLa Y σY r > 0 







Differentiating (6.8), we get aYZ dZ + ZdaYZ = dY or, 



Z + aˆ YZ = Y

  ∗ Now, aˆ YZ = θLa Z σZ W − P Y = −θLa Z σZ P Y = − θLaθZYZσZ dt < 0 









Z = Y − aˆ YZ 

= θLa Y σY r + θLa Z σZ = θLa Y σY

dt σYZ

dt dt + θLa Z σZ θYZ θKY σYZ

(6.107)

126

6 Effectiveness of Trade and Non-trade Policies …

=

dt θYZ



θLa Y σY + θLaZ σZ θKY

 >0

(6.108)

From (6.7)      WA  A aLa Y Y + aLa Z Z + LU W aLa Y Y + aLa Z Z = W + β       WA WA = LU W A + ⇒ aLa Y Y + aLa Z Z W ∗ − W A − β β ∗

(6.109)

Differentiating (6.109), we get ! !   WA WA d aLa Y Y + aLa Z Z = dLU W A + β β ! ! A   W WA aLa Y dY + YdaLa Y + aLa Z dZ + ZdaLa Z = dLU W A + W∗ − WA − β β ! ! "    # A W WA W∗ − WA − aLa Y Y Y + aˆ La Y + aLa Z Z Z + aˆ La Z = dLU W A + β β ! !  

  W A  WA aLa Y Y aˆ La Y − aˆ KY + aLa Z Z aˆ La Z − aˆ YZ − aˆ KY = dLU W A + W∗ − WA − β β ! ! #  A A "    W W aLa Y Y σY r + aLa Z Z σZ P Y + θLa Y σY r = dLU W A + W∗ − WA − β β ! !  A  Zθ σ dt a dt dt W WA L Z La Y Y aLa Y Y σ Y W∗ − WA − = dLU W A + + aLa Z ZσZ + a β θKY θYZ θYZ θKY θYZ β   A W ∗ A   W −W − β aL Z ZθLa Y σY dt aLa Y Y σY   + aLa Z ZσZ + a (6.120) >0 dLU = A θYZ θKY θKY W A + Wβ

W∗ − WA − ⇒



⇒ ⇒ ⇒ ⇒

⇒









To find out the impact on export sector, differentiating (6.6), we get aLa X dX + XdaLa X + aLa Y dY + YdaLa Y + aLa Z dZ + ZdaLa Z + dLU = 0

(6.121)

dLU =0 L     dLU ⇒ λLa X X + λLa Y Y + aˆ La Y + λLa Z Z + aˆ La Z = − L     σZ dt dLU θL Y σY dt =− + a ⇒ λLa X X + λLa Y σY r + λLa Z θYZ θYZ θKY L   dLU dt λLa Y σY λL Z θL Y σY =− + λLa Z σZ + a a ⇒ λ La X X + θYZ θKY θKY L   λLa Y σY dLU −dt λLa Z θLa Y σY − + λLa Z σZ + 0, there must be an increase in adult unemployment rate. (C) Effect on Child Labour Supply   In our model, LC S = LC S W A , aLa X X , Differentiating it, we get dLSC =

∂LC S ∂WA

< 0,

∂LC S ∂ (aLa X X )

>0

  ∂LC S ∂LC S A   d aLa X X d W + ∂WA ∂ aLa X X

128

6 Effectiveness of Trade and Non-trade Policies …

∂LC S  aLa X dX =  ∂ aLa X X  ∂LC S   aLa X X X < 0 =  ∂ aLa X X 

(6.126)

Proposition 37 Child labour supply will also be reduced due to protectionism in import competing sector as number of adult labourers engaged in rural sector of the economy goes down.

6.4.5 Effects of Protectionism on National Welfare Impact on National Welfare Two general considerations merit detailed analysis—they are: impact of restrictive trade policies on national welfare, and the effects of such protectionist policies aimed at curbing the incidence of child labour on the trade balance of the economy. These two considerations, apart from being important for the macro perspective of the policy effectiveness of trade restrictions, are also crucial since some conventional propositions associated with such types of policy changes get revised in the process. The first task is to test for the impact on national welfare. Our measure of welfare in the small open economy is national income (N) and it is expressed as follows:     WA A X + aLa Y Y + aLa Z Z W ∗ + rK N = aLa X W + aLC X β

(6.127)

  A In Eq. (6.127), aLa X W A + aLC X Wβ X gives the wage income of adult and child   workers in the rural sector. On the other hand, aLa Y Y + aLa Z Z W ∗ gives the wage income of adult workers in the urban sector. rK is the rental income from capital. Case 1: Impact of Protectionism in the Exported Sector of the economy on Welfare of the Economy Differentiating (6.127), we get ! $ ! WA WA dX + aLa X d W A + W A daLa X + aLC X d + β β     + aLa Y W ∗ dY + (W ∗ Y )daLa Y + aLa Z W ∗ dZ + (W ∗ Z)daLa Z + Kdr

dN = aLa X W A + aLC X

! % WA daLC X X β

or,        A WA WA WA A A X ∧ dX + aLa X W X W + aLC X dN = aLa X W + aLC X β β β 

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

129



= PX (1 − t)X X − PX (1 − t)Xdt < 0

(6.128)

Proposition 38 Due to imposition of trade restriction on the exported product of small open economy, welfare will unambiguously decline. Case 2: Impact of Protectionism in the Non-traded Intermediate Sector on Welfare of the Economy Rewrite national income from Eq. (6.127):   WA A X + aLa Y W ∗ Y + aLa Z W ∗ Z + rK N = aLa X W + aLC X β

(6.127)

Differentiating it, we have dN = PX dX + W ∗ aLa Y dY + W ∗ YdaLa Y + W ∗ aLa Z dZ + W ∗ ZdaLa Z + Kdr   dN = (PX X )X + W ∗ aLa Y Y Y + aˆ LaY + W ∗ aLa Z Z(Z + aˆ La Z )    = (PX X )X + W ∗ aLa Y Y aˆ La Y − aˆ KY + W ∗ aLaZ Z(Y − aˆ YZ + aˆ La Z ) + Krˆr        θLa Y σY dt = (PX X )X + W ∗ aLa Y Y σY rˆ + W ∗ aLa Z Z + σZ P Y (1 + t) + Krr θKY  ∗  W aLa Y Y + W ∗ aLa Z ZθLa Y σY dt Krdt = (PX X )X + + W ∗ aLa Z ZσZ P Y (1 + t) + θKY θKY   θLa Y PY Y + W ∗ aLa Z Z σY dt Krdt ∗ = (PX X )X + + W aLa Z ZσZ P Y (1 + t) + θKY θKY

      aLa Y Y σY rˆ + aLa Z Z(Y − aYZ + aLa Z ) + dLU θLa Y PY Y + W ∗ aLa Z Z σY dt = −(PX X ) + aLa X X θKY 



























+ W ∗ aLa Z ZσZ P Y (1 + t) +

Krdt θKY



  aLa Y Y σY dt aL Z ZθLa Y σY dt + a + aLa Z ZσZ P Y (1 + t) θKY θKY ⎞ ⎛   ∗ aLa Y Y σ Y dt aL Z ZθLa Y σY dt W P X ⎝ − 1⎠ + a + aLa Z ZσZ P Y (1 + t) − X aLa X X W A + W A θKY θKY β    θLa Y PY Y + W ∗ aLa Z Z σY dt Krdt + W ∗ aLa Z ZσZ Pˆ Y (1 + t) + ≷ 0 + θKY θKY

=

−(PX X ) aLa X X





(6.129)

Hence, impact on welfare of small open economy is ambiguous due to increase in protectionism in non-traded intermediary producing sector. In our model, we have derived that for a small open economy, higher tariff rate is not necessarily inferior to a lower tariff rate. In our model, result of Kemp’s (1962, 1969) theorem becomes invalid, since in our model, there is wage differential between the urban and rural sectors which represents factor market distortion. Hence, we get ambiguous impact on welfare of the small open economy following protectionist trade policy.

130

6 Effectiveness of Trade and Non-trade Policies …

Proposition 39 Stricter protectionism in non-traded intermediary producing sector may or may not be immiserizing in nature. Welfare of the small open economy will improve if       dt λLa Y σY dLU  θLa Y σY dt (PX X ) + + λ σ + La Z Z  λLa X θYZ θKY θYZ θKY LλLa X        W ∗a  ∗  θLa Y σY dt Krdt  σZ dt La Y σY dt  0 = XD p, , (6.135) PZ PZ ∂p ∂ PNZ

132

6 Effectiveness of Trade and Non-trade Policies …

Similarly, the demand function for importable is   N ∂ZD ∂ZD , ZD = ZD p, > 0,   > 0 PZ ∂p ∂ PNZ

(6.136)

Therefore d (TB) = −PX X0 dt + PX (1 − t)dX 0 − PZ dZ 0 = −PX X 0dt + PX (1 − t)[dX − dX D ] − PZ [dZ D − dZ] ⎡ ⎤     1 − λLa X ∂XD ∂XD N ⎦ ⎣ dp +   d − PX (1 − t) = −PX X0 dt − PX (1 − t)Xdt λLa X ∂p PZ ∂ PN Z ⎡ ⎤   ∂ZD N ⎦ ∂ZD dp +   d − PZ ⎣ ∂p PZ ∂ N PZ



 1 − λLa X λLa X ⎤ ⎡     ∂XD ∂ZD N N ⎦ ⎣ + PZ   d − PX (1 − t)   d ≷ 0 PZ PZ ∂ N ∂ N

= PX X0 dt − PX (1 − t)Xdt

PZ

(6.137)

PZ

Proposition 41 Due to trade restriction by the rest of the world on the small country’s exported product, we get an ambiguous impact on trade balance of that economy. To investigate the impact on trade balance, we will separately study the effect on value of export and value of import. Value of importable will be reduced only for operation of income effect. Value of exportable will fall due to fall in production of exportable. But demand for exportable will be reduced following a fall in national income. This will create an opposite impact on value of exportable. Finally, impact on trade balance becomes ambiguous. Case 2: Protectionism in Non-traded Intermediate Producing Sector and Impact on Trade Balance Trade balance of an economy is given by TB = PxX0 − PZ Z0 The effect on domestic production of X and Z is given by      aLa Y Y σY rˆ + aLa Z Z (Y − aˆ YZ + aˆ La Z ) + dLU 0

(6.140)

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

133

Consumption demand function for exportable is given by XD = XD {Px , PZ , N }

(6.141)

Since demand function is homogeneous of degree zero in all prices and income, we choose PZ as numeraire and thus the demand function for exportable is reduced in the form   Px N XD = XD , 1, PZ PZ   N Px ∂XD ∂XD ,p = < 0,   > 0 = XD p, , (6.142) PZ PZ ∂p ∂ PNZ Similarly, the demand function for importable is   N ∂ZD ∂ZD , > 0,   > 0 ZD = ZD p, PZ ∂p ∂ PNZ

(6.143)

Whence d (TB) = Px dX 0 − PZ dZ 0 = Px (dX − dX D ) − PZ (dZ D − dZ) ⎡ ⎤ ⎤ ⎡     ∂Z ∂X N ⎦ N ⎦ ∂X ∂Z = Px X X + PZ ZZ − Px ⎣ D dp +  D  d −P Z ⎣ D dp +  D  d ∂p PZ ∂p PZ ∂ PN ∂ PN 



Z

In this case, price effect is zero. ⎡



∂XD N ∴ d (TB) = Px X X + Pz ZZ − ⎣Px   d N P Z ∂ PZ 



(6.144)

Z



⎤  ∂ZD N ⎦ + PZ   d ≷0 N P Z ∂ 

PZ

(6.145) The sign of first term is negative and second term is positive. Now, the impact on national income due to stricter protectionism in the sector producing non-traded intermediate product is ambiguous. Hence, the impact on trade balance is also ambiguous. Proposition 42 Due to protectionism in non-traded intermediary producing sector, we get an ambiguous impact on the trade balance of the economy. A fall in export production will lead to reduce value of exportable. On the other hand, increase in import production will reduce value of importable. However, impact on quantity demanded for both exportable and importable remains ambiguous as protectionism leads to ambiguous impact on national income. Hence, we cannot surely infer the consequence of such protectionism on trade balance.

134

6 Effectiveness of Trade and Non-trade Policies …

Case 3: Protectionism in Import Competing Sector and Impact on Trade Balance In this case, the trade balance of an economy is given by TB = Px X0 − PZ (1 + t)Z0

(6.146)

We have already found the impact of protectionism on export and import competing sector and they are as follows:   λLa Y σY dLU dt λL Z θL Y σY − + λLa Z σZ + a a 0 θYZ θKY 



Now, consumption demand function for exportable is given by XD = XD {Px , PZ (1 + t), N }

(6.149)

As demand function is homogeneous of degree zero in all price and income, we choose PZ (1+t) as numeraire and thus the demand function for exportable is reduced in the form   Px N XD = XD , 1, (6.150) PZ (1 + t) PZ (1 + t)   N Px ∂XD ∂XD  >0 ,p = , < 0,  = XD p, (6.151) N PZ (1 + t) PZ (1 + t) ∂p ∂ PZ (1+t) Similarly, the demand function for importable is   N ∂ZD ∂ZD  >0 , > 0,  ZD = ZD p, N PZ (1 + t) ∂p ∂

(6.152)

PZ (1+t)

Differentiating Eq. (6.146), we get d (TB) = Px dX 0 − PZ (1 + t)dZ 0 − PZ Z0dt = Px [dX − dX D ] − PZ (1 + t)[dZ D − dZ] − PZ Z0dt = Px dX + PZ (1 + t)dZ − Px dX D − PZ (1 + t)dZ D − PZ Z0 dt ∴ d (TB) = Px X Xˆ + PZ (1 + t)Z Zˆ ⎤ ⎡     ∂YD ∂XD N N ⎦ d d − ⎣ Px  + PZ (1 + t)  N N PZ (1 + t) PZ (1 + t) ∂ P (1+t) ∂ P (1+t) Z

− PZ dZ 0 dt

Z

(6.153)

6.4 Child Labour in a Three-Sector General Equilibrium Framework …

135

Hence, impact on trade balance also becomes ambiguous. Proposition 43 Due to protectionism in import competing sector, impact on trade balance is ambiguous.

6.4.7 Comparison of Alternative Policies We have considered in this part of the chapter, a three-sector competitive general equilibrium model of a small open, less developed economy suffering from unemployment problem in the adult labour market on one hand, and from the existence of a child labour market on the other hand. The economy is divided into one rural and two urban sub-sectors. Four inputs are used in the model among which three are specific in nature. The representative adult worker in this model not only supplies his own labour but also sends his children out to work. Factor market distortion in this model is captured by the existence of a factor price differential between urban and rural sectors. The model is used to analyse the effects of imposition of various forms of trade restrictions on the unemployment of adult workers on the one hand and on the incidence of child labour on the other. Alternative trade policy prescriptions to combat child labour as well as adult market unemployment problem have been analysed. An interesting result obtained from the exercise of trade policy suggests that supply of child labour is shown to vary inversely with the level of unemployment in the adult labour market. This goes against the existing view that unemployment in the adult labour market is one of the important causes for the existence of child labour market. The result is also interesting in the context of the problems raised by Basu (2000) who thinks that imposing a minimum wage in the adult labour market causes unemployment to rise and may aggravate the problem of child labour. This problem should not be serious in our model as the supply of child labour is shown to vary inversely with the level of adult unemployment. We find that all the three trade policies have similar impacts on child labour market as well as on adult labour market—all the trade policies are effective in curbing the incidence of child labour but none of them could effectively tackle the adult labour unemployment problem. Hence, we conclude that none of the trade policy discussed in this section is able to combat both the incidence of child labour and adult unemployment. The effect on trade balance is ambiguous, but the effects on national income or welfare may be immiserizing in cases, depending on sufficient conditions. Part III4

4 See

Ray and Chatterjee (2010c) for details.

136

6 Effectiveness of Trade and Non-trade Policies …

6.5 Education Subsidy, Adult Unemployment and The Incidence of Child Labour—A Three-Sector General Equilibrium Analysis We now move to non-trade policies, and one such non-trade policy suggested to curb or contain the incidence of child labour in a less developed economy is education subsidy to children. There is no denying the fact that schooling of children raises their future earnings, yet one finds relatively low enrolments in schools amongst children belonging to currently poor families. A common explanation is that schooling competes with labour-intensive jobs for children (wage labour, employment in family enterprises, or collection activities). By this view, the low current income of their families keeps poor children out of school and thus perpetuates their poverty into the next generation. Households that are very poor are much more likely to send their children to work, and child labour contributes to poverty in the next generation by reducing schooling attainment. This circular constellation of link between poverty and child labour may lead to a vicious cycle of poverty. If schooling attainments can be improved, then the next generation’s income could be higher and their children can in turn become yet better educated. The most effective way to draw children out of damaging work is to encourage school attendance. Establishment of compulsory education is a necessary condition for the reduction and abolition of child labour. Without compulsory education, government would not have been able to enforce child labour laws. If the school-leaving age is lower than the age of admission to employment, children are likely illegally to seek employment, and the enforcement of child labour laws is rendered more difficult. It is administratively easier to monitor school attendance than to monitor children in the work place, especially when there are thousands and tens of thousands of employees, and easier to force parents to send their children to school than to force employers not to hire children. But legally making education for children would not work unless the vicious circle of poverty and child labour are broken. In this part of the present chapter, we develop a three-sector general equilibrium model of a small open economy to examine the effectiveness of education subsidy on school enrolment or reduction in the incidence of child unemployment. In this model, we do not impose any trade restriction on the export of the small open economy but introduce child schooling function.

6.5.1 Assumptions The present paper considers (a) a competitive general equilibrium model of a small open less developed economy suffering from unemployment problem in the adult labour market on the one hand, and from the existence of child labour on the other. (b) There are three-sectors in the model with four inputs.

6.5 Education Subsidy, Adult Unemployment and The Incidence …

137

(c) The rural sector of the economy produces exportable commodity using adult and child labour. (d) One of the urban sub-sectors produces non-traded intermediary using adult labour and capital. (e) The other urban sub-sector is the tariff protected import competing sector of the economy that produces its product using adult labour and a non-traded intermediary input. (f) There is presence of urban adult employment but the other inputs are fully employed.

6.5.2 The Model The following symbols will be used in the formal presentation of the model: Adult labour-output ratio in the ith sector i = X, Y, Z Child labour-output ratio in the X sector Capital-output ratio in the Y sector Non-traded intermediary input–output ratio in Z sector Distributive share of the ith input in the jth commodity, i = L A , L C , K, Y, j = X, Y, Z World price of the ith commodity, i = X, Z Pi Domestically determined price of Y PY WA  A  Adult wage rate in the rural sector C Child wage rate in the rural sector W = Wβ ∗ W Adult wage rate in the urban sectors r Rate of return to capital K Capital stock of the economy L Total adult labour endowment Level of urban adult labour unemployment in economy LU Aggregate supply of child labour LC S ∧ Proportional change X Domestic production of exportable Z Domestic production of importable Number of schoolgoing children LS E Expenditure on education incurred by parents Access to school σ1 Adult literacy parameter. σ2 aLa i aLC X aKY aYZ θij

A general equilibrium of the system is given by the following set of equations: aLa X W A + aLC X

WA = PX β

aLa Y W ∗ + aKY r = PY

(6.154) (6.155)

138

6 Effectiveness of Trade and Non-trade Policies …

aLa Z W ∗ + aYZ PY = PZ

(6.156)

aKY Y = K

(6.157)

aYZ Z = Y

(6.158)

aLa X X + aLa Y Y + aLa Z Z + LU = L

(6.159)

aLC X X = aLa X X − LS = LC S

(6.160)

 ∂f  ∂f ∂f ∂f < 0, LS = f W A , E,σ1 ,σ2 , > 0, > 0, > 0, E = F ∂WA ∂E ∂σ1 ∂σ2   aLa Y Y W ∗ + aLa Z Z W ∗ WA = WA + aLa Y Y + aLa Z Z + LU β

(6.161) (6.162)

Equations (6.154)–(6.156) are the usual price unit cost equality conditions in the three sectors of the economy. X, Y, and Z sectors respectively denote the exportable, non-traded intermediary and the import competing sector of the economy. Equation (6.157) gives the full employment condition of capital. The demand–supply equilibrium condition for the intermediate good Y is given by Eq. (6.158). Equation (6.159) gives the adult labour market equilibrium condition. Equation (6.160) gives the child labour market equilibrium condition. The left-hand side gives the demand for child labour in the economy and the right-hand side gives the supply of child labour for the whole economy. In our model, parents engaged in rural sector send their children to the job market. Adult rural workers send their children to labour market due to poverty which acts as a deterrent. Two major ways in which poverty can lower the children’s enrolment rate in school are (a) a reduced capability to learn in school because of malnourishment, poor health and frequent illness, and (b) sporadic attendance at schools because of the value of child’s labour for the family. There are two costs of sending children to school—one is direct costs and another is indirect cost. Tuition fee, expenditure on books, stationery, cloths and related items and even cash payments in the form of examination fees, sports fee, hostel fee, private tuition fee contribute to the direct cost of schooling. Indirect cost of schooling can be thought of as the opportunity cost of losing income from child labour’s engagement in work as opposed to going to school. The opportunity cost is assumed to be very high as children make valuable contributions to household economy. Hence, sending them to school, instead of using them as household help or wage earners is not an economically feasible option to the parents. In our model we assume that each adult worker has one child labour at his disposal. So, the number of adult worker engaged in rural sector gives the number of children at their disposal. Let LS denote the number of children are going to school from rural sector. Hence, the number of children coming to the job or effective supply of child

6.5 Education Subsidy, Adult Unemployment and The Incidence …

139

labour in the small open economy is aLa X X − LS . In Eq. (6.161), child schooling function has been specified. We have incorporated four determinants of this function: A > Adult wage rate: Higher is the rural adult wage rate, less will be incidence of poverty and higher will be enrolment of rural children in school. B > Expenditure of education: Hike in expenditure on education will curtail school enrolment rate. Initially, suppose parents have to incur a fixed cost F if they want to provide education to their children. This is independent of parents’ choice. C > Access to school: Shortage of number of schools and inadequate infrastructure in schools or distance of schools from remote rural residence will curtail the school enrolment rate. D > Adult literacy rate: Socio-economic status is a crucial determinant of school education. Illiteracy and parents’ lack of interest in education is one major cause for low enrolment rate. Educated parents will always be willing to send their children to the job market. Finally, the Harris–Todaro migration equilibrium condition is given by Eq. (6.162). The right-hand side of (6.162) is the total income of a working family in the rural sector, which must be equal to the expected total income of the family in the urban sector (the left-hand side) in migration equilibrium. There are nine endogenous variables in the system: W A , r, PY , X , Y , Z, LC S , LS , LU and nine independent equations. Hence the system is solvable. The parameters in the system are PX , PZ , W ∗ , L, K, β, E. We should note that the system possesses the decomposition property. The three unknown factor prices W A , r, PY are determined from Eqs. (6.154)–(6.156) independent of the factor endowments. Once factor prices are known factor coefficients are also determined. From (6.156), given W ∗ and PZ , PY is obtained and substituting the value of PY in (6.155), r can be solved. From A (6.154), given PX , W A (hence Wβ ) is determined. From (6.157) given K, output of Y is determined. Once Y is obtained, Z is automatically obtained from (6.158). Again, from (6.159), LU can be obtained by substituting value of Y and Z. Substituting the value of Y, Z, LU in (6.160), X is determined. LS is determined from (6.161) and substituting the value of LS , LC S is obtained.

6.5.3 Comparative Static Exercises Case 1: Government Subsidy in School Education Initially, we assume parents have to incur a fixed cost F on schooling. Now, we assume that government provides subsidy at a rate s to each student, so that effective cost of schooling is now reduced to F(1 − s). Now, we shall investigate its effect on various aspects. The model is rewritten in the following way: aLa X W A + aLC X

WA = PX β

(6.163)

140

6 Effectiveness of Trade and Non-trade Policies …

aLa Y W ∗ + aKY r = PY

(6.164)

aLa Z W ∗ + aYZ PY = PZ

(6.165)

aKY Y = K

(6.166)

aYZ Z = Y

(6.167)

aLa X X + aLa Y Y + aLa Z Z + LU = L

(6.168)

aLC X X = aLa X X − LS = LC S

(6.169)



 ∂f ∂f ∂f ∂f LS = f W A , E,σ1 ,σ2 , > 0, > 0, > 0, E = F(1 − s), 0 < s < 1 < 0, ∂E ∂σ1 ∂σ2 ∂WA ∗

∗



aLa Y Y W + aLa Z Z W W = WA + aLa Y Y + aLa Z Z + LU β

A

(6.170) (6.171)

We observe that in this modified model, only Eq. (6.170) is changed. A. Effect on Factor prices Since there will be no change of Eqs. (6.154)–(6.156) which determine factor prices, no factor prices will be affected. Proposition 44 Government subsidy to school education will not alter factor prices in the system. B. Effect on Composition of Output Since there will be no change of Eqs. (6.159)–(6.160), output of rural as well as urban sector will be unaffected. Proposition 45 Government subsidy to school education will not affect output of rural or either urban sub-sectors. C. Effect on Urban Adult Unemployment Since the output of any urban sub-sectors or rural sector is not affected, there will be no change in urban adult unemployment. Proposition 46 Government subsidy to school education will not aggravate or reduce urban adult unemployment problem. D. Effect on Number of School going Children

6.5 Education Subsidy, Adult Unemployment and The Incidence …

141

Due to imposition of subsidy, expenditure on school education is curtailed. Expenditure on education is E = F(1 − s), 0 < s < 1

(6.172)

dE = −Fds < 0

(6.173)

Differentiating, it we get

Differentiating (6.170), we get dLS =

∂f dE > 0 ∂E

(6.174)

Proposition 47 Government subsidy to education will increase the school enrolment rate by reducing expenditure on education. E. Effect on Child Labour Supply Differentiating (6.160), we get dLSC = aLa X dX − dLs = −dLs < 0 Proposition 48 Government subsidy to education will be effective in controlling child labour supply of the economy. Hence, we may conclude that government encouragement to school education will affect only two variables in the system and the policy is effective in eradicating child labour incidence. Case 2: Education Cess on Urban Workers to Finance Subsidy on Education and Impact on Different Macroeconomic Variables Suppose government of the small open economy decides to take a fiscal step in curbing child labour supply of the economy. They have announced that all urban adult workers have to pay a proportional tax t on per unit wage income. The tax revenue collected in this connection will be distributed as education subsidy. The framework of the model is given as follows: aLa X W A + aLC X

WA = PX β

(6.175)

aLa Y W ∗ + aKY r = PY

(6.176)

aLa Z W ∗ + aYZ PY = PZ

(6.177)

142

6 Effectiveness of Trade and Non-trade Policies …

aLa X X + aLa Y Y + aLa Z Z + LU = L

(6.178)

aKY Y = K

(6.179)

aYZ Z = Y

(6.180)

aLC X X = aLa X X − LS = LC S

(6.181)

  ∂f ∂f ∂f ∂f < 0, LS = f W A , E,σ1 ,σ2 , > 0, > 0, >0 ∂WA ∂E ∂σ1 ∂σ2   aLa Y Y W ∗ t + aLa Z Z W ∗ t E =F 1− LS   WA aLa Y Y W ∗ (1 − t) + aLa Z Z W ∗ (1 − t) A = W + aLa Y Y + aLa Z Z + LU β

(6.182) (6.183) (6.184)

Expenditure on education, i.e. E is introduced as another explanatory variable in this system. It is solved from (6.183). Now, we shall investigate the effect of education cess on various macroeconomic variables. Since all the equations except (6.183) and (6.184) are not affected, we conclude that there will be no change in (i) any factor price (ii) output of rural sector (iii) output of any sub-sector To find out the effect on child labour supply, we rewrite Eq. (6.183) in the following way:   ELS = FLS − F aLa Y Y W ∗ + aLa Z Z W ∗ t

(6.185)

Differentiating it, we get   (F − E)dLS − LS dE = F aLa Y Y W ∗ + aLa Z Z W ∗ dt   ∂f (F − E) dE − LS dE = F aLa Y Y W ∗ + aLa Z Z W ∗ dt ∂E     ∂f − LS dE = F aLa Y Y W ∗ + aLa Z Z W ∗ dt (F − E) ∂E   F aLa Y Y W ∗ + aLa Z Z W ∗ dE   ⇒ = < 0(F > E) dt (F − E) ∂f − L ∂E

(6.186)

S

dLS ∂f dE Hence = > 0 (∵ W A ,σ1 ,σ2 areunaffected ) dt ∂E dt

(6.187)

6.5 Education Subsidy, Adult Unemployment and The Incidence …

143

Proposition 49 Education cess on urban workers to finance subsidy in education is effective in encouraging school enrolment. Differentiating Eq. (6.181), we get dLs dLSC =− 0  A  aˆ KY = θLaY σY W − rˆ < 0

! aˆ KY − aˆ La Y W here σY = Elasticity of substitution = A W − rˆ 





Proof The zero profit conditions for the economy is shown by the following two equations:  aLC X

WA β

 + aLa X W A = Px

aLA Y W A + aKY r = PY

(6.193)

(6.194)

Cost minimization condition for the producer of Y entails that daKY WA =− daLa Y r

(6.195)

or, rdaKY + W A daLa Y = 0 or,

Now σY =

θKY aˆ KY + θLa Y aˆ La Y = 0

(6.196)

 A  ∴ aˆ KY = σY W − rˆ + aˆ La Y

(6.197)

aˆ KY −ˆaLa Y 

W

A



−r





Substituting the value of aKY KY in (6.196), we get   A θLa Y aˆ LaY + θKY aˆ La Y + θKY σY W − rˆ = 0 

or,

148

6 Effectiveness of Trade and Non-trade Policies …

 A  aˆ La Y = −θKY σY W − rˆ 

∴ From (6.197)  A  aˆ KY = (1 − θKY )σY W − rˆ  A  = θLa Y σY W − rˆ 





A

∵ aˆ LaY − aˆ KY = −σY (W − rˆ )

(6.198) (6.199)

Appendix 6.2 In specific factor model, we will prove that  ∗  aˆ LaZ = −θYZ σZ W − P Y (1 + t)  ∗  aˆ YZ = θLaZ σZ W − P Y (1 + t) 







where σZ = Elasticity of substitution ! aˆ YZ − aˆ LaZ = ∗ W − P Y (1 + t) 



Proof The zero profit conditions for the economy are shown by the following two equations: aLaY W ∗ + aKY r = PY (1 + t)

(6.200)

aLaZ W ∗ + aYZ PY (1 + t) = PZ

(6.201)

Cost minimization condition for the producer of Z entails that daYZ W∗ =− daLaZ PY (1 + t) or, PY (1 + t)daYZ + W ∗ daLaZ = 0 or,

(6.202)

Appendix 6.2

149

θYZ aˆ YZ + θLaZ aˆ LaZ = 0  Now σZ =



aˆ YZ −ˆaLaZ ∗



(6.203)



W −P Y (1+t) ∗





∴ aˆ YZ = σZ [W − P Y (1 + t)] + aˆ LaZ

(6.204)

Substituting the value of aˆ YZ in (6.203), we get 

∗



θYZ σZ [W − P Y (1 + t)] + θYZ aˆ LaZ + θLaZ aˆ LaZ = 0 or,

aˆ LaZ = −θYZ σZ W ∗ − PY (1 + t)

(6.205)

References Basu, K. (1999). Child labour: Cause, consequence, and cure, with remarks on international labour standards. Journal of Economic Literature, 37(3), 1083–1119. Basu, K. (2000). The intriguing relationship between adult minimum wage and child labour. The Economic Journal, 110(462), C50–C61. Basu, K., & Van, P. H. (1998). The economics of child labour. American Economic Review, 88(3), 412–427. Batra, R. N. (1973). Studies in the pure theory of international trade. London and Basingstoke: The Macmillan Press Limited. Bhagwati, J. (1968). Distortions and immiserizing growth: A generalization. Review of Economic Studies, 35. Chatterjee, B., & Ray, R. (2008). A three sector model with child labour and adult market unemployment: Impact of trade restriction on national welfare and trade balance. Artha Beekshan, 17(3). Chatterjee, B., & Ray, R. (2008). Trade policy and impact on child labour supply—A two sector general equilibrium analysis. Journal of Quantitative Economics, New Series, 6(1 and 2). Chatterjee, B., & Ray, R. (2013). Trade restriction, adult unemployment and incidence of child labour: A three sector general equilibrium analysis. Artha Vijnana Journal of The Gokhale Institute of Politics and Economics, 55(3), 239–251. Chaudhuri, S. (2006). Trade liberalization in agriculture in developed nations and incidence of child labour in a developing economy. Bulletin of Economic Research, 4. Gupta, M. R. (2000). Wage determination of a child worker: A theoretical analysis. Review of Development Economics, 4(2), 219–228. Gupta, M. R. (2003). Child labour and economic development: A theoretical analysis. Arthaniti, 2, 97–107. Also reprinted in Arthaniti (2018). Harris, J. R., & Todaro, M. P. (1970). Migration, unemployment and development: A two sector analysis. American Economic Review, 60, 126–142. Johnson, H. G. (1965). Optimal trade intervention in the presence of domestic distortions. In Caves, Johnson, & Kenen (Eds.), Money, trade and growth: Essays in honour of G. Haberler. Rand Mcnally, Chicago. Jones, R. W. (1965). The structure of simple general equilibrium models. The Journal of Political Economy, 73(6), 557–572.

150

6 Effectiveness of Trade and Non-trade Policies …

Jones, R. W. (1971). A three-factor model in theory, trade, and history. In J. Bhagwati et al. (Eds.), Trade, balance of payments and growth: Essays in honour of Charles P. Kindleberger. Kemp, M. C. (1962). The gain from international trade. Economic Journal, 72, 803–819. Kemp, M. C. (1969). The pure theory of international trade and investment. Englewood Cliffs, N.J.: Prentice Hall. Maskus, K.E. (1997). Should core labour standards be imposed through international trade policy? World Bank, Policy Research Working Paper, No. 1817. Ranjan, P. (2001). Credit constraints and the phenomenon of child labour. Journal of Development Economics, 64(1), 81–102. Ray, R. (2014). Chapter 5. Import restrictions, capital accumulation and use of child labour—A general equilibrium analysis. In A. N. Ghosh & A. K. Karmakar (Eds.), Analytical issues in trade, development and finance, essays in honour of Biswajit Chatterjee. Springer, India. Ray, R., & Biswajit, C. (2010). Impact of restrictive trade policy on adult unemployment, welfare and the incidence of child labour—A three sector general equilibrium analysis. Journal of Quantitative Economics, The Indian Econometric Society, 8(1), 148–161. Ray, R., & Chatterjee, B. (2010a). Education subsidy, adult unemployment and the incidence of child labour in an open economy: A three sector general equilibrium analysis. The Indian Economic Journal, 57(4), 37–59. Ray, R., & Chatterjee, B. (2010b). Child labour and restrictive trade policy—A three sector general equilibrium framework. Asian Economic Review, Journal of The Indian Institute of Economics, 52(1), 179–190. Stolper, W. F., & Samuelson, P. A. (1941). Protection and real wages. Review of Economic Studies, 9, 58–73. Swaminathan, M. (1998). Economic growth and the persistence of child labour: Evidence from an Indian city. World Development, 26(8), 1513–1528.

Chapter 7

Conclusion

The practice of sending children to work is prevalent widely throughout the globe, particularly in developing countries. Although there have been diminutions in the incidence of child labour over the recent periods, including in India, the size of child labour employment in various forms, continues to be quite large in such economies. Among the various factors contributing to child labour use, the poverty of parents is considered to be the principal factor that induces parents to send their children to work, rather than sending them to schools. Various policy alternatives have been suggested to combat the incidence of child labour in developing countries. One policy is, of course, administrative or legislative, i.e. ban on the use of child labour. Such a prescription of total ban on child labour use may not be effective, as there are imperfect monitoring, and loss of income from working children may actually compel parents to send their children further to work, rather than to school, and since child labour is cheaper compared to adult labour, the producers would also be interested in increasing the demand for child labour. The other alternative suggested by the WTO in imposing labour standards is to put up tariff wall against the exports of the developing countries, as these products are often child labour intensive in nature. Trade policy alternatives, if implemented, have the possible consequences of generating various forms of domestic distortions, depending upon the structures of the economies, which often adversely affect national welfare, and may not always result in the diminution of the incidence of child labour in such developing economies. The other alternative is non-trade policy package, which includes reorientation of growth strategies towards non-child—labour using production sectors, policy on education subsidy to encourage sending children to school and not towards work, etc. Our studies in the present monograph analyse the implications of non-trade policies visà-vis trade policies towards child labour use in the developing economies in terms of alternative structures of production and labour use in general equilibrium framework. In the two-sector version of our general equilibrium model presented in Chap. 5 of the book (also refer to Ray 2015; Ray and Chatterjee 2010a; Chatterjee and Ray 2007, 2008, 2016a, b), we have sought to investigate the effect of two alternative trade restrictive policies on child labour market. The comparative statics in the model, generate the following results: (1) When the Rest of the World (ROW) imposes © Springer Nature Singapore Pte Ltd. 2019 B. Chatterjee and R. Ray, Economics of Child Labour, https://doi.org/10.1007/978-981-13-8199-7_7

151

152

7 Conclusion

trade restriction on the exported product (which is child labour intensive) of the small open developing country, the size of its export sector contract and that of the import competing sector expands, and capital owners gain while both adult and child labourers face a decline in their wages; (2) In case of additively separable utility function of the adult worker, imposition of trade restriction may or may not have perverse impact on incidence of child labour supply (also Chatterjee and Ray 2016a, b); (3) Protectionism in the country’s import competing sector will have a redistribution effect while only capital owners shall benefit, there shall no change in adult as well as child wage rate, though the output of the import competing sector shall expand at the cost of its export sector; (4) Restrictive trade policy in the form of increase in import tariff will unambiguously increase child labour supply of the economy. Therefore, trade policy instrument in such an economy does not yield the desirable impact of diminution of the incidence of child labour. We have extended the model into a three-sector general equilibrium framework in Chap. 6 of the book (also refer to Chatterjee and Ray 2008, 2010a, b, c, 2016a, b). The trade policy considered here involves protectionism in import competing sector of the economy, whereas non-trade policy is captured in terms of economic expansion (i) by an increase in domestic capital stock and (ii) by an increase in adult labour endowment. In this chapter, we assume three sectors in the economy of which two are informal sectors. The first informal sector produces agricultural product (X), which is exported, whereas the other informal sector produces a nontraded final commodity (Z) (services). The third sector is a formal sector producing a manufacturing commodity (Y) in the import competing sector. Adult labour, child labour and capital are the three inputs, but the specifications are that while Y is produced using adult labour and capital, X and Z are produced using adult labour and child labour. In this model, child labour is perfectly mobile within informal sectors (X and Z), adult labour is perfectly mobile between all sectors but capital is a specific factor of production in the import competing Y sector. The adult wage rate in the formal sector is assumed to be fixed at W* (may be due to unionization or legislative support) exceeding the competitive informal sector adult wage rate, W. The main results of our comparative statics exercises are the following: (1) ‘A rise in price of import competing good produces a Stolper–Samuelson effect and raises the rental rate leaving the wage rates of child and adult workers unchanged and there will be no change in price of non traded good’ (also Ray 2014); (2) Stricter protectionism in import competing sector expands the formal sector, i.e. the import competing sector of the economy and increase in rental rate also increases income of the formal sector; (3) ‘Due to protectionism in import competing sector one of the informal sector (non-traded final good producing sector) will expand and export sector (another informal sector) will contract’ (also Ray 2014); (4) ‘Stringent protectionism in import competing sector must lead to a decline in child labour supply’ (also Ray 2014); (5) ‘Economic expansion by increase in domestic capital stock will not lead to any change in factor prices as well as in price of non-traded good’ (also Ray 2015); (6) ‘Expansion of domestic capital stock of the economy will adversely affect the export sector. However, the non-traded good producing sector will expand and the size of the import competing sector will be enhanced’ (also Ray 2014); (7) Due to

7 Conclusion

153

economic expansion of the above kind, child labour supply in the economy will fall; (8) ‘Economic expansion by increase in adult labour endowment will not make any change in factor prices, will expand the export sector of the economy, but there will be no change in either the import competing sector of the economy or in the other informal sector of the economy’ (also Chatterjee and Ray 2013); (9) ‘Expansion of adult labour endowment of the economy will increase child labour supply of the economy’ (also Chatterjee and Ray 2013). When we introduce the coexistence of adult unemployment and existence of child labour (to accommodate realism prevailing in many less developed economy) and want to investigate the effect of imposition of three alternative trade restrictive policies on child labour supply, as well as on urban adult unemployment problem, we have some interesting results. Our model economy consists of a rural sector (sector X), which produces the export product of the economy with help of intersectorally mobile adult labour and sector-specific child labour, and two urban sectors. Adult labour in our model is assumed to be perfectly mobile between all the three sectors, but capital and child labour are specific factors of production. The first urban sector (Y sector) produces the non-traded intermediate product with the help of sector-specific capital and adult labour. The second urban sector (sector Z) is tariff protected which produces the import competing good using adult labour and a non-traded intermediate input, Z, which is specific to the sector. It is assumed the two urban sectors face an exogenously given W* > WA , in other words, the urban wage rate exceeds the market determined the wage rate in the agricultural sector. We, however, retain the assumption of perfect competition in all the product markets, but we consider the case of factor market distortion in the form of the existence of a factor price differential between sectors. Thus our model resembles a Harris–Todaro type of rural–urban wage differential. It is clear that unemployment in the rural sector is ruled out due to complete flexibility of wage rate. CRS production functions, with diminishing marginal productivity to each factor is assumed. The assumption of small open economy fixes prices of the traded goods X and Z internationally, but the price of the non-traded commodity Y is endogenously determined by the demand–supply mechanism. Finally, we assume that rest of the world imposes a tariff on the exported product of the small open economy, so there is a decline in effective producer’s price. Exporter of the small open economy has to pay tariff at a rate t before entering into the rest of the world. In such a model, our comparative static exercises examine the effect of the imposition of trade restriction (1) by rest of the world on the exported product of the small open economy; and the imposition of stringent protectionism (2) on the production of non-traded intermediate product; and (3) on import competing sector of the economy. We have the following results: (1) Due to the imposition of trade restriction on the exported product of the small open economy, only labourers (both adult and child) will be worse off, but there will be no change in any other factor price; (2) Trade restrictive policy will reduce the output of the export sector of the economy. On the other hand, there will be no change in the output of any urban sub-sector; (3) Imposition of trade restriction on the exported product would aggravate the unemployment problem in the adult labour market; (4) Due to trade restriction by rest of the world on

154

7 Conclusion

the exported product of the small open economy, the impact on child labour supply becomes ambiguous, and the size of child labour market (i.e. actual employment of child labour in equilibrium) will be reduced; (5) Due to protectionism in the sector producing non-traded intermediate input, capital owners will only be benefited, but there will be no effect on adult or child wage rate, the export sector will contract and output of the other two urban sub-sectors will expand; (6) Due to protectionism in non-traded intermediate input producing sector, adult unemployment rate will go up, but child labour supply will contract (vide Ray 2014, 2015; Ray and Chatterjee 2010a, b, c). The above three-sector model has been extended incorporating three changes. In the first place, we drop the assumption of trade restrictions on the export of the small open economy. Second, we incorporate child schooling function in our model. Finally, child labour supply in this model also made dependent on the child schooling function. Here, our objective has been to examine the role of education subsidy in curtailing the incidence of child labour supply. In our extended three-sector version, adult unemployment in the urban sector coexists with the presence of child labour use. Also, the exportable commodity is produced in the rural sector of the economy using both adult and child labour. One of the urban sub-sectors produces non-traded intermediate input, which uses adult labour and capital. The other urban sub-sector is the tariff protected import competing sector of the economy that produces its product using adult labour and a non-traded intermediate input. Here, we have examined the effectiveness of alternative non-trade policies on the incidence of child labour as well as on urban adult unemployment. The alternative non-trade policies considered are: (i) government subsidy in school education; (ii) education cess on urban workers to finance subsidy on education; (iii) tax on urban workers to finance unemployment allowance; (iv) an increase in adult literacy rate. Our main results are the following: (a) Government subsidy to school education will not alter factor prices in the system; (b) Government subsidy to school education will not affect output of rural or either urban sub-sectors, nor will it aggravate or reduce urban adult unemployment problem; (c) Government subsidy to education will increase the school enrolment rate and will be effective in controlling child labour supply of the economy; (d) Education cess on urban workers to finance subsidy in education is effective in encouraging school enrolment, and child labour supply of the economy goes down when the government provides subsidy in education by taxing urban workers; (e) Government subsidy in education by imposing education cess on urban workers is effective in curtailing urban adult unemployment problem; (f) If instead of education subsidy tax revenue is distributed as unemployment allowance to urban unemployed, then there does not occur any change in the system and no macroeconomic variable will be affected, making this policy ineffective in solving either urban adult unemployment problem or problem of child labour; (g) An increase in adult literacy rate by making parents more literate, favourably impact the school enrolment of children, reduce the child labour supply, without any change in other variables in the system (the results contain use of our own research papers, Ray and Chatterjee 2010a, b, cited with permission).

7 Conclusion

155

The above theoretical results of our models indicate that the use of blanket trade restriction policies may not be sufficient to eliminate the incidence of child labour in a small open economy. Rather, non-trade policies like the acceleration of growth in sectors not using child labour may be effective by reducing the adult unemployment and increasing the level of income of parents. It can be said that progress towards the reduction in child labour can be made if there is a revival of sustained economic growth, the pattern of growth is more egalitarian, the modern sector is expanding rapidly, there is universal compulsory schooling, particularly of girl child, and adult literacy programmes are launched, and when the government provides subsidy in education by taxing urban workers through an education cess, it encourage school enrolment, and child labour supply of the economy goes down (ILO 2013, 2017). There may be a time lag in making such non-trade policies effective, but given the elimination of child labour as an objective, such policies, together with complementary support policies, shall influence the incidence of child labour, by tackling the underlying cause. It can be said that a country’s full commitment to the total abolition of child labour should be judged not merely on the basis of official pronouncements but on whether the child labour objective is consciously considered in the above outlined social and economic policy framework (ILO 2013). The 2006 Global Action Plan of ILO (2006a, b, c) called on member States to pursue the effective abolition of child labour by committing themselves to the elimination of all worst forms of child labour by 2016. This target was revised subsequently. The pursuit of trade policies may prove to be counterproductive, non-trade policies of various forms may be used to reduce the incidence of child labour in different parts of the globe. Growth promoting strategies are one kind of initiatives—such policies would directly reduce the incidence of poverty among the adult labour force and ease out pressures on poor parents to send their children to work. Social protection measures are another set of non-trade policies which may provide support to poor families, and facilitate access to education and health care facilities and help them to enrol their children to schools. Such social protection instruments, which are most helpful in combating child labour, include (a) Cash and in-kind transfer programmes that enhance income security for families and; (b) Public employment programmes, which provide jobs for adults to build and improve roads, schools, health centres and the like, helping to ensure that it is adults who are at work and not children; (c) Social health protection measures, which ensure access to health care and financial protection in case of sickness, and can stop households sending children to work when a member of the household falls ill; (d) Unemployment protection, which provides adults with at least partial income replacement would reduce the need to rely on the income of working children when facing job loss (ILO 2013). In other words, a set of social protection measures could address the economic and social vulnerabilities associated with poverty, and deprivation and are likely to impact favourably towards the elimination of child labour. Such social protection policies, which are basically non-trade policies, have the potential to have a diminishing impact on the child labour use and increase social welfare as our study clearly demonstrate. Our general equilibrium framework can accommodate different types of production structures in developing countries and show how growth promoting

156

7 Conclusion

policies or policy like education subsidy can have economic effects of reducing the incidence of child labour. Reliance on trade sanctions, ban on child labour or trade policy interventions in different forms of protectionism in less developed poor open economies seem to bypass the issue of elimination or reduction in the use of child labour.

References Chatterjee, B., & Ray, R. (2007). Growth, national welfare and trade balance in a small open economy: A general equilibrium analysis. Artha Vijnana, XLIX(3&4), 203–222. Chatterjee, B., & Ray, R. (2008). A three sector model with child labour and adult market unemployment: Impact of trade restriction on national welfare and trade balance. Artha Beekshan, 17(3). Chatterjee, B., & Ray, R. (2013). Trade restriction, adult unemployment and incidence of child labour: A three sector general equilibrium analysis. Artha Vijnana—Journal of the Gokhale Institute of Politics and Economics, LV (3), 239–251. Chatterjee, B., & Ray, R. (2016a). Impact of trade vs. non-trade policies on the incidence of child labour: A two-sector general equilibrium framework. Foreign Trade Review, 287–297. Chatterjee, B., & Ray, R. (2016b). Impact of trade restriction on child labour supply and the role of parents’ utility function: A two sector general equilibrium analysis. In M. Roy & S. S. Roy (Eds.), International trade and international finance. Springer. ILO. (2006a). The end of child labour: Within reach, Global Report under the follow-up to the ILO Declaration on Fundamental Principles and Rights at Work. Report of the Director-General, Report I (B), International Labour Conference, 95th Session, Geneva, 2006. ILO. (2006b). The International Programme on the Elimination of Child Labour (IPEC). What it is and what it does. Geneva: ILO. International Labour Organization (2006c). Global child labour trends 2000–2004. Geneva: ILO. International Labour Organization (2013). Child labour and social protection. Accelerating the pace of progress, campaign and advocacy. Geneva: ILO. International Labour Organization (2017). Global estimates of child labour: Results and trends 2012–2016. Geneva: ILO. Ray, R., & Chatterjee, B. (2010a). Impact of restrictive trade policy on adult unemployment, welfare and the incidence of child labour—A three sector general equilibrium analysis. Journal of Quantitative Economics—The Indian Econometric Society, 8(1), 148–161. Ray, R., & Chatterjee, B. (2010b). Education subsidy, adult unemployment and the incidence of child labour in an open economy: A three sector general equilibrium analysis. The Indian Economic Journal, 57(4), 37–59. Ray, R., & Chatterjee, B. (2010c). Child labour and restrictive trade policy—A three sector general equilibrium framework. Asian Economic Review—Journal of the Indian Institute of Economics, 52(1), 179–190. Ray, R. (2014). Import restrictions, capital accumulation and use of child labour—A general equilibrium analysis. In A. N. Ghosh & A. K. Karmark (Eds.), Analytical issues in trade, development and finance. Essays in honour of Biswajit Chatterjee. Springer India (chapter 5). Ray, R. (2015). Trade versus non-trade policy in a two sector general equilibrium framework. Business and Economics Journal, 6(2).