correcting my paper
hem-99
5.pdf
Child Labor Issues and Directions for the
World Bank
Peter Fallon and Zafiris Tzannatos
Social Protection Human Development Network
The World Bank Washington, D.C.
© 1998 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433
All rights reserved Manufactured in the United States of America First printing February 1998 (Revised post-conference document)
The authors are grateful for the advice and support given by Robert Holzmann, Director, Social Protection, and for comments received from World Bank staff and others.
Cover design by Glenn McGrath, American Writing Corporation.
Library of Congress Cataloging-in-Publication Data
Fallon, Peter, 1946– Child labor : issues and directions for the World Bank / Peter
Fallon, Zafiris Tzannatos. p. cm.
Includes bibliographical references. ISBN 0-8213-4183-9 1. Children—Employment. 2. Child welfare. 3. World Bank.
I. Tzannatos, Zafiris, 1953– . II. Title HD62331 1998 331.3'1—dc21 98-5030
CIP
i i
Contents
Abbreviations and acronyms iv
Foreword v
Executive Summary vi
Child labor 1
I. Background 2
II. Ways to reduce child labor 9
III. The World Bank’s role 13
IV. Summary 18
Annex 20
References 22
iii
Abbreviations and acronyms
i v
CAS Country assistance strategy DAC Development Assistance Committee of the OECD ESW Economic and sector work GDP Gross domestic product HD Human Development ILO International Labour Organisation IPEC International Program for Elimination of Child Labor LAC Latin America and the Caribbean region NGO Nongovernmental organization OECD Organisation for Economic Co-operation and Development PFP Policy framework paper SAS South Asia region UNESCO United Nations Educational, Scientific and
Cultural Organization UN United Nations UNICEF United Nations Children’s Fund WHO World Health Organization WTO World Trade Organization
Foreword
Child labor is one of the most devastating consequences of persistent poverty. The World Bank is committed to helping reduce child labor through its ongoing poverty reduction efforts in its member countries and through new initiatives aimed specifically at combating the most harmful and exploitative forms of child labor.
Other organizations—the ILO, UNICEF, and NGOs—are in the fore- front of global activities to attack the problem, and appropriately so. Their contributions provide a compelling basis for the Bank to build on as it now steps up its efforts to do its own part in support of others.
Reducing poverty through economic development, and promoting other improvements such as changes in basic education, are essential elements of effective strategies to attack child labor. But such actions take a long time to have a significant impact. Additional faster-acting measures are needed. This paper outlines the specific steps the Bank will take to strengthen its efforts, incrementally gearing up step-by-step.
International concern—as well as interest in action—is growing, as evidenced by the Amsterdam Child Labor Conference in February 1997 and the Oslo Conference in October 1997. The time is right for all con- cerned organizations, including the Bank, to do more.
David de Ferranti Caio K. Koch-Weser Vice President Managing Director Head, Human Development Network
October, 1997
v
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N Kv i
Executive Summary
Children should not have to work, but an estimated 250 million chil- dren are working worldwide. This paper proposes that the Bank take stronger action—and outlines what to do and how—to help address the issue more pro-actively, partnering with others who share a similar con- cern about the damage to children, their development, and their society’s development.
Child labor is a complex subject. While often harmful, it is not always so, especially where the alternative is deeper poverty for children and their families. Missteps in trying to stamp it out can make matters worse, for example, if legislation is unevenly enforced and results in pushing children into worse situations. Solutions are needed to reach beyond conventional thinking. For example, although improving primary edu- cation is key, it is also critical to adapt schools (for example, their times and schedules) to accommodate children who otherwise could not at- tend and increase real access (such as by reducing the costs to families of having their children go to school).
Vital to achieving progress against harmful child labor are (i) effec- tive efforts to reduce poverty generally and (ii) the economic and social policies, programs, and results that are the underpinning for success in poverty reduction. But these broad measures, while important, take time and are not sufficient by themselves. Additional actions focused specifi- cally on child labor per se are also needed.
Others—the ILO, UNICEF, NGOs, and some governments—take the leading roles on child labor issues and are stepping up their efforts. The World Bank, working with other institutions, also has a contribution to make, consistent with its focus on development and poverty reduction.
Specifically, the Bank can help in several ways. Within the context of its country assistance strategies (CAS), it can systematically integrate child labor considerations into its programs. The Human Development Network is already bringing child labor into its CAS reviews and other operational documents. In addition, the Bank should undertake the fol- lowing actions on child labor.
v i
vii
In its lending activities, it is suggested that the Bank should:
• Design and introduce new projects or project components to re- duce the harmful effects of child work. For example, these could identify locations where harmful child activities are taking place and supply services to children and income-earning opportunities to their parents. A start has been made in an Urban Sector Develop- ment project in Brazil. Further pilots should be developed in coun- tries where harmful child labor is present.
• Emphasize child labor issues more in other projects. For example, in education projects, increase the flexibility of school hours with a view to maintaining the contact of child laborers with the school system. Already there are signs of progress in existing projects in India, and future projects will build on this experience.
• Introduce child labor concerns in social assessments for projects where relevant. This is starting in projects in South Asia.
In its nonlending activities, it is suggested that the Bank should:
• Bring child labor issues into the policy dialogue in countries where harmful child labor is a serious problem.
• Undertake more ESW and research on child labor issues. Child la- bor is already featured in ESW on Thailand and in research on Côte d’Ivoire and Ghana.
• Increase staff awareness of child labor. A module on child labor was included in HD Training Week (in the autumn of 1997; twenty- nine partner institutions contributed). A symposium on child labor in Africa was held in May 1997.
• Strengthen partnerships. The Bank is inexperienced in operational matters regarding child labor and must learn from others. A high- level team from the ILO visited the Bank and has indicated strong willingness to share its experience on child labor. The Bank’s SAS Region has formed an informal consultative group with a number of concerned NGOs.
In addition, to ensure that Bank-financed projects do not contribute to the problem, the Bank’s operational practices should allow that, in coun- tries where there is a serious risk, Bank loan agreements can include appropriate safeguards. The SAS Region has instructed task managers
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N Kviii
to identify and report any cases of children working on Bank-financed projects.
No institution acting alone can solve the problem, given the massive resources required. The Bank, with its finite resources and special man- date, is no exception. Partnerships are therefore essential. Working to- gether, a global effort to eliminate child labor is feasible and would be a major contribution to world development.
1
1
Child Labor Issues and Directions for the World
Bank
Child labor is a serious global issue. In its most harmful forms, it im- pairs the physical and mental development of children. A number of questions have been raised about the Bank’s position, both on child la- bor in general and in relation to the possible use of child labor in Bank- financed projects. While primary responsibility for child labor questions and policies internationally resides in other organizations (the ILO, UNESCO, UNICEF, WHO), the Bank is concerned about it from both a development and humanitarian point of view. The evidence is compel- ling that the long-run solution to child labor problems comes from re- ducing poverty, improving the situation of women, and increasing ac- cess to and the quality of education.
Promoting improvements in these areas is central to the Bank’s mis- sion. Nevertheless, abusive forms of child labor are, and will continue to be, too common in a large number of countries until poverty is sub- stantially reduced. The Bank addressed some of the broad concerns about child labor in the 1995 World Development Report. However, in view of recent questions that have been raised directly to the Bank and else- where, Management has decided to summarize in this paper the cur- rent situation and data on child labor and to clarify the position of the Bank. One aim of this report is to sensitize clients and staff to these is- sues and to indicate ways we can help reduce harmful child labor in the course of our operations.
The basic facts and issues about child labor—its extent and nature, relevant laws and the positions of the international organizations—are described in Section I. This is followed by a brief discussion of major approaches taken toward child labor (Section II) and a discussion of the Bank’s position (Section III) with reference to some initiatives now un- der way or planned.
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K2
I. Background
Extent, composition, and trends in child labor
The ILO estimates the number of working children aged between 5 and 14 years to be about 250 million in the developing countries, of whom at least 120 million are working full time. Of these, 61 percent are in Asia, 32 percent in Africa, and 7 percent in Latin America. Relatively few chil- dren work in developed countries. Earlier ILO estimates suggested about 80 million child workers worldwide (see annex table), of whom about 73 million were between 10 and 14 years of age. These earlier estimates are believed to be on the low side, as they imply, for example, a labor force participation rate among children aged 10-14 of only 14 percent, which compares with much higher figures for the percentage of chil- dren not attending school.
There are major differences in the incidence of child labor across re- gions and sectors. As noted above, most child workers are found in Asia. But the proportion of children working is highest in Africa where, on average, one child in three is engaged in some form of economic activ- ity, mostly in agriculture. In general, child labor force participation rates are much higher in rural than in urban areas, and three-quarters of work- ing children work in the family enterprise. Ninety percent of working children in rural areas are engaged in agricultural or similar activities, while their urban counterparts are found mainly in trade and services, with fewer in manufacturing and construction. Although urban street children have received considerable attention, far larger numbers are employed in agriculture and domestic service. Similarly, child workers in export industries (such as textile, clothing, carpets, and footwear) are relatively few compared with those employed in activities geared to domestic consumption (ILO 1996a). Probably less than 5 percent of child laborers are employed in the export manufacturing or mining sectors, and only 1 to 2 percent are employed in export-oriented agriculture.
The incidence of child labor has been falling on a global basis, but there are different trends across regions. The proportion of children who work has been declining in Southeast Asia—given rising per capita in- comes, the spread of basic education, and a reduction in family size— but rising in many countries in Africa. There is also evidence that the incidence of child labor increased in some Latin America and Caribbean (LAC) countries and in Eastern Europe (ILO 1996b).
There does not seem to be much difference in child participation rates by gender. Differences in child labor between boys and girls are masked by measurement problems, as boys are commonly in more visible types of employment while girls work in unpaid household work. For example,
3
in six countries studied under IPEC, the ratio of boys to girls in the labor force was found to be 3:2, but if domestic activities are included, the numbers become similar. In countries where data exist, girls appear to work more hours, which is consistent with gender differences in educa- tional enrollment rates.
Despite data problems, some generalizations can be made regarding differences in the incidence of child labor between countries.
First, both the incidence of child labor and children as a proportion of the total labor force (see figure) decline with per capita GDP. The labor force participation rate of children aged 10-14 years is extremely high at 30-60 percent in countries with per capita income of $500 or less (at 1987 prices). But it declines quite rapidly to 10-30 percent in countries with incomes between $500 and $1,000. This negative relationship between income and child work becomes less marked in the more affluent devel- oping countries (in the $1,000 to $4,000 income range). It is not clear whether the apparent lack of association at higher incomes is due to statistical or other reasons, although the relation between child labor incidence and per capita income across countries is likely to be affected by cultural differences. The best “predictor” of child labor seems related to the structure of production: the higher the share of agriculture in GDP, the higher the incidence of child labor.
0
2
4
6
8
10
12
200 300 400 500 650 825 1200 1700 2600 5300
GDP per capita (in 1987 US$)
P er
ce n
t o
f w
o rk
er s
w h
o a
re c
h il
d re
n
Proportion of child workers declines with rise in GDP per capita
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K4
Second, the incidence of child labor declines with educational enroll- ment, though with large cross-country variation. This relationship is somewhat blurred because some children may work to finance the costs of their education or the education of younger siblings. There is also some evidence that child labor force participation rates are negatively associated with teacher/pupil ratios. This may reflect quality issues— from shortages of teachers to overcrowding—that result in low expecta- tions of value added by education and low expected private returns to a child’s education.
Characteristics of child labor
Child labor can be of widely varying kinds, some harmful and some not. Indicators that should be taken into account to determine whether working has a negative impact on children’s development include work- ing conditions (for example, hours of work, exposure to physical haz- ards, etc.) and other risks and abuse to which working children may be exposed (for example, psychological and social adjustment risks, pros- titution, and child slavery).
Long working hours are often responsible for fatigue that can cause accidents, and impair intellectual development. Some studies consider 20 weekly hours of work as a critical threshold beyond which education starts being significantly affected. In Ecuador, 70 percent of child work- ers work more than 20 hours. In rural Bangladesh, by the age of 13, non- schoolgoing children already work hours as long or even longer than adults. In Thailand, working boys and girls aged 11 to 15 years have average weekly hours of work of 50 hours, but for girls in services the average exceeds 65 hours.
Many children work under exploitative conditions that, apart from totally precluding schooling, have harmful effects on their physical con- dition and mental health. The working conditions of child garbage pick- ers in the Philippines clearly increase the risk of diseases and disability through exposure to lead and mercury, heavy lifting, and the presence of parasites. Children in agriculture are more likely to be adversely af- fected than adults by climatic exposure, heavy work, toxic chemicals, and accidents from sharpened tools and motorized equipment.
Domestic service is primarily undertaken by girls and is one of the occupations that can cause serious psychological and social adjustment problems. Such child workers typically live away from home and may routinely work long hours, often in almost total isolation from family and friends. In Colombia, domestic servants comprise the majority of all child workers laboring over 60 hours per week. WHO 1987 reports
5
that psychological stress, premature aging, depression, and low self-es- teem are common symptoms among young household helpers.
At the extreme, traditional forms of child slavery such as bonded la- bor exist in South Asia and in East Africa. Instances of slavery have also been reported in two LAC countries (ILO 1996a). Slave labor is more common in agriculture, domestic help, the sex industry, carpet and tex- tile industries, and quarrying and brick making.
There are powerful economic arguments for measures to reduce child labor. Premature and extensive engagement in work prevents children from accumulating human capital and having higher earnings in later life, while economic growth is adversely affected by lower rates of pro- ductivity growth. In many instances, child work is the result of capital market failures: when households cannot afford education for their chil- dren and cannot borrow for this purpose, although the long-term ben- efits would be high.
Not all child labor is harmful. Many working children are within a stable and nurturing environment with their parents or under the pro- tection of a guardian and can benefit in terms of socialization and from informal education and training. Between 50 and 70 percent of working children, depending on the country in question, are studying at the same time, although this may reduce their chances of completing primary edu- cation. There are, however, many children for whom work does unques- tionable harm, the main culprits being long hours and physical and mental damage.
Legal framework
Most countries have national legislation that establishes a minimum working age and regulates the working conditions of young persons. In addition, many have ratified one or more ILO Conventions and the UN Convention on the Rights of the Child (1989), which means that the con- ventions that are all in force are binding upon the states party to them. Children are further protected against child labor by peremptory norms of international law binding on all subjects of such law, for example, states and international organizations. Such norms include rules pro- hibiting slave trading and slavery.
As far back as 1919, ILO Convention No. 5 prohibited the work of persons below the age of 14 in industrial establishments. Nine sectoral conventions on the minimum working age were subsequently adopted until the Minimum Age Convention No. 138 (1973) came into force in 1976 along with the accompanying Recommendation No. 146. Today 90 countries are bound by the provisions of one or more of these conven-
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K6
tions. A further 51 countries have ratified Convention No. 138, while 18 additional ones have begun or are in the process of ratifying it, and an- other 14 countries are contemplating doing so. In all, 141 of the 174 ILO member states have now ratified at least one of the ILO conventions concerning the minimum working age. However, as of mid-1997, only 21 developing countries have ratified Convention No. 138, and none is in Asia, where over half of working children are found.
In addition, all except two of the ILO member states have ratified the Convention on the Rights of the Child adopted in 1989 by the UN Gen- eral Assembly. In all, 187 countries have ratified this convention, cover- ing 97 percent of the world’s children. The convention has the force of an international treaty, and its provisions (including those on child la- bor) are binding on the parties to it. Since the scope of the child’s right to be protected from economic exploitation is defined by reference to “the relevant provisions of other international instruments,” it may be con- sidered that any work performed by children in conditions inferior to those laid down by UN or ILO conventions is to be deemed as economic exploitation (ILO 1996b).
Intolerable forms of child labor are partly covered by the Forced La- bor Convention (1930) that has been ratified by 136 countries. Child la- bor practices that amount to forced labor come within the meaning of the convention—that is, “all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily.” Bonded labor and trafficking of children for employment or prostitution are covered by the convention, but not other forms of harmful child labor.
There are likely to be further developments in international law in the next few years. The Governing Body of the ILO has decided to in- clude child labor in the agenda for the 1998 session of the International Labor Conference, with a view to adopting new international standards. The objective is to create a binding instrument banning the most intoler- able forms of child labor (under slavery or forced or bonded conditions; prostitution and other illegal sex practices; and for drug trafficking or in the production of pornography). There are also activities that expose children to particularly grave hazards to their safety or health or pre- vent them from attending school normally.
National legislation often exempts agriculture and household employ- ment and is generally more enforceable in the formal sector. This ex- plains why only a small proportion of working children are found in formal employment. While legislation and adequate enforcement can play a significant role, enforcement beyond the formal sector is often impractical and is not a cost-effective means of protecting working chil- dren most in need.
7
Position of the international organizations
The mandate of other specialized international organizations is narrower than that of the World Bank, and this allows them to have a more fo- cused approach to child labor or child issues in general. For example, UNESCO deals with education; UNICEF has global responsibility to im- prove the situation of children; and ILO deals with employment and broader labor issues on a tripartite basis, while the Bank’s purpose is to assist its borrowing members in their reconstruction and economic and social development. In addition, while the Bank has to take an economic approach because of the requirements of its Articles of Agreement, other organizations may be motivated by a general human rights perspective. However, there is now consensus that partnership at the international and country level is a prerequisite for combating child labor, and coop- eration with the Bank is often explicitly sought. As the following discus- sion indicates, there is increasing international convergence of views on child labor.
Child labor was discussed in World Development Report 1995, which noted the limitations of legislation and the importance of poverty as a determinant of child labor. WDR 1995 called for a multifaceted approach with programs that increase income security, reduce education costs, and improve the quality of schooling. It concluded: “as the incidence of poverty falls and education improves, child labor will decline. That in turn will make enforcement of legislated bans easier, starting with such universally abhorred forms of child labor as prostitution and hazardous work.”
There is significant agreement about why children supply labor and on the fact that child labor cannot be seen in isolation from education and population issues. For example, the Bank’s emphasis on poverty reduction is very much echoed in the following passage:
“Poverty is the greatest single force which creates the flow of children into the workplace ... [Acute need] makes it nearly impossible for households to invest in their children’s education [and] the price of education can be very high [as] most ‘free’ public education is in fact very expensive for a poor family ... Poor households tend to have more children [and large family size] has been statis- tically shown to be associated not only with higher likeli- hood that children work, but also lower school attendance and completion” (ILO 1996b).
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K8
The effects of growth and poverty reduction, education expansion, and control of population growth upon child labor are, therefore, clearly acknowledged, and these are areas where the Bank makes a significant contribution. In particular, “the single most effective way to stem the flow of school-age children into abusive forms of employment is to ex- tend and improve schooling so that it will attract and retain them” (ILO 1996b). There is a considerable body of research that shows that increas- ing the relevance of education improves children’s future employment prospects and makes schools more attractive. And although school at- tendance cannot always remove children from work, there is evidence that children in school are less likely to be involved in bonded labor or in particularly detrimental forms of work, most of which are inconsis- tent with regular school attendance.
In the same vein, the Development Cooperation Report of DAC sets out an integrated approach for reducing child labor that includes an overriding focus on poverty alleviation; strong investment in education; an enhanced participation of women in economic development; private sector development; and greater involvement of civil society.
UNICEF has prepared a “check list” of strategic options that could meet the “best interest of the child.” With respect to child labor and other forms of economic exploitation of children, it includes: expanding education opportunities either by “time off” to attend regular school or by providing schooling in the work place; providing support services to parents, especially mothers; promoting stricter law enforcement against traffickers and those who “bond” child labor; providing services for children working on the streets; raising the age of marriage; and chang- ing cultural values and social norms tolerant of the economic exploita- tion of children.
The ILO proposes actions at the country level and improved interna- tional cooperation. At the country level, it calls for a national strategy and specific measures for combating child labor. A national strategy should have several elements, including: improving the data bases on child labor to give better knowledge and diagnosis of the situation; cre- ating a national plan of action against child labor; raising awareness of the child labor problem; creating a broad social alliance against child labor; and creating institutional capacity to deal with the child labor problem. The most recent ILO report on child labor (ILO 1996a) calls for countries that are signatories to relevant ILO conventions or the UN Convention on the Rights of the Child to adopt time-bound action pro- grams to eliminate child labor.
Specific measures suggested by the ILO include improving child la- bor legislation and enforcement measures; extending and improving
9
schooling for the poor; and using “positive” economic incentives to re- duce the supply of child labor. Interestingly, ILO cautions against the use of “negative” incentives (such as trade sanctions) and their unin- tended effects on child workers. It calls for further examination of the suggestion that “instead of imposing or threatening to impose trade sanc- tions against poorer countries resorting to child labor in their export- related activities, developed countries should assist these poorer coun- tries in solving their child labor problems, both in export-oriented ac- tivities and in domestic economic sectors, including special rewards in the trade field when it is proven that they have achieved good progress in their fight against this scourge” (ILO 1996b).
With respect to international activities and its own role, ILO sees fu- ture activities in the area of new standards and instruments; dissemina- tion of information and undertaking action-oriented research; increased technical cooperation; and increased collaboration with the other inter- national organizations.
In its last ministerial conference (December 1996), the WTO discussed whether core labor standards, including those relating to child labor, should be part of its concerns. After considerable debate, agreement was reached on four points:
• all WTO members oppose abusive workplace practices through their approval of the UN’s Universal Declaration of Human Rights;
• the ILO holds primary responsibility for labor issues; • trade sanctions should not be used to deal with disputes over labor
standards; • member states agree that the comparative advantages of low wage
countries should not be compromised.
II. Ways to reduce child labor
A number of approaches have been suggested to combat child labor. The major thrusts are: reducing poverty; educating children; providing support services for working children; raising public awareness; legis- lating and regulating child labor; and promoting elimination of abusive child labor through international measures. These approaches are, of course, not mutually exclusive and are adopted in various combinations in child labor reduction strategies.
Not all of these approaches can be adopted by the Bank, however. In particular, legislation and regulation are the concerns of governments, al- though the Bank can discuss this in the country dialogue, while trade sanc- tions and other international measures lie outside the Bank’s mandate.
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K1 0
Reducing poverty
It is generally agreed that widespread poverty is the major cause of harm- ful child labor in developing countries. In poor households children may contribute a significant proportion of household income, which means that because such households spend the bulk of their income on food, income from child labor may be critical to survival. Programs that improve the labor earnings of the poor, address capital market imperfections, and pro- vide safety nets may be particularly relevant here. There is widespread consensus that poverty reduction is the most powerful long-term approach. But this is a lengthy process that, even when successful, will in practice tend to raise the incomes of the poor unevenly, thus leaving room for a substantial incidence of child labor for some time to come.
Educating children
Interventions aimed at increasing participation in primary education also have wide support, particularly if embedded in a strategy of pov- erty reduction. These foster human capital development with positive economic and social spin-offs. The impact on child labor is likely to de- pend, however, on the nature of the intervention. There are three ap- proaches: making basic education compulsory; making it easier for chil- dren to attend both work and school; and reducing the cost to house- holds of school attendance.
Making basic education compulsory, while highly desirable on other grounds, may not do much to reduce child labor if the pressures keep- ing children out of work are sufficiently strong, and enforcement weak. Strong enforcement may reduce child labor but also endanger the wel- fare of poorer households who depend heavily on this income source. Compulsory education is an important defense against child slavery and the most exploitative forms of child labor, as children have to be re- leased to attend school, but it also runs the risk of overloading children who work in non-school hours.
Making it easier for children to attend work and school is a desirable option and seems particularly relevant in rural areas if schooling can be scheduled not to conflict with the peak agricultural season. This prin- ciple was applied successfully in Kerala (India) and Vietnam. This can, however, make the school year too short— for example, closure of up to 200 days a year in Kerala. An alternative in rural and urban areas is to run night classes, although it is difficult to target these only for working children.
Reducing educational costs is another option because this gives an incentive for households to divert child labor supply into education.
1 1
This may take the form of educational subsidies such as scholarships, provision of educational goods, school feeding schemes, and the provi- sion of kindergartens so that girls, for example, do not have to look after younger siblings. The practical problems with this approach include costs and sustainability. There is also a tendency to have an impact only on households already considering sending their children to school while leaving the behavior of other households unchanged.
Providing support services for working children
Most of these initiatives have tried to help children working in the ur- ban informal sector, although there have also been initiatives in rural areas and with child domestics. Urban projects include feeding schemes, night shelters, literacy programs, and other ventures. Some projects also try to help the families of working children. While such projects offer scope for improvement in the lives of child workers, they tend to con- centrate on those working visibly on the street and to give overempha- sis to feeding programs.
Raising public awareness
This covers a wide spectrum of activities including: improving child knowledge of work hazards; raising parental awareness of the human capital loss that may be associated with child labor; and changing the emphasis of policymakers. Improved awareness is more likely to lead to cooperation between local communities, NGOs, and governments to put pressure on employers. Although there is widespread agreement with this approach, it is important that it be part of a wider strategy. Putting pressure on employers alone does not solve the problem unless alternative sources of income are created for affected families. Similarly, increased awareness in developed countries of child labor problems in the Third World can lead to demands for inappropriate action against countries in which child labor is a serious problem without sufficient attention to the consequences for the child workers themselves.
Legislation and regulation
As noted earlier, most countries have child labor laws and regula- tions, but enforcement is often weak. Increasingly, it is recognized that in the circumstances of many developing countries, stricter across-the-board enforcement may damage those whom it intends to protect by reducing the incomes of poor families and forcing work-
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K1 2
ing children into more dangerous hidden forms of employment. If the salaries of labor inspectors are low, then the potential for bribery and discretionary enforcement becomes a real danger. Legislation works best when it fits in with an ongoing trend. In Europe the em- ployment of children in factories started falling when employers found it in their interest to switch to adults some 50 years before the introduction of legislation.
The alternative is to focus enforcement only upon the most intoler- able aspects of the problem, such as children working in hazardous con- ditions, child prostitution, and bonded labor. This requires sensitization of inspectors and police toward the nature of child work and a willing- ness to prosecute culpable employers. The ILO intends that its next con- vention will target the elimination of the more harmful kinds of child labor and encourage a more discretionary approach to the enforcement of legislation and regulations.
Promoting elimination through nonregulatory means
Regulatory measures aside, the main mechanism suggested to elimi- nate child labor is through trade-related measures such as trade sanc- tions, consumer boycotts, social clauses, and certification or label- ing schemes. There are a number of problems with this approach. First, if, as is usually the case, exports produced in the formal sector are the products most hit by trade measures, the effect will be to force child workers into the informal sector where working condi- tions are worse. Second, smaller countries are more likely to be tar- geted, given international concerns about the global effects of sanc- tions on large countries. Third, the imposition of sanctions may be little more than a cover for those who wish to introduce trade pro- tectionist measures in developed countries and may be implemented in a discretionary way that has little to do with child labor. Fourth, the use of child labor in countries party to social clauses or labeling schemes may simply be non-monitorable, thus rendering these mea- sures ineffective.
Other measures aimed at the direct elimination of child labor include rescue by NGOs of children in the worst conditions and the establishment of motivational camps in which children leave work and attend informal education and recreational facilities on a residential basis for a period of time. While these are worthy ventures, it is important that they be com- bined with schemes to give alternative income sources to the children and their families.
1 3
III. The World Bank’s role
There are a number of ways that the Bank can assist and perhaps accel- erate the reduction of child labor, including the use of lending and nonlending activities and through modifications in some operational practices. These will be addressed in the context of the CAS in countries where abusive child labor constitutes a significant problem. The Bank can help by giving more prominence to its concern in its dialogue with governments, drawing particular attention to ways to mitigate the con- ditions that permit or encourage child labor as well as encouraging en- forcement against the worst child labor practices. Because poverty is a major cause of child labor, the Bank’s continuing emphasis on poverty reduction contributes to conditions that mitigate the pressures that lead to child labor. At the same time, helping children engaged in harmful work activities is, in itself, a useful further step toward assisting the poor. Many Bank efforts in education, health, and population also help to reduce the incidence and the harmful effects of child labor more di- rectly, while activities such as the Bank’s Children’s Initiative, which currently monitors activities regarding children in health and educa- tion, can be reinforced by extending them to cover child labor issues. Raising its staff’s awareness of the extent of the problem and of ways to combat child labor is critical to the Bank’s efforts in this area. A number of initiatives are under way or being developed to address child labor problems.
Lending initiatives
Lending programs in affected countries can be refocused to have more impact on reducing the incidence of harmful child labor. This is most relevant in countries where the objective of universal primary educa- tion will take many years to achieve. The refocusing includes redesign of conventional lending activities as well as new projects or project com- ponents specifically aimed at reducing the incentives for harmful child labor. District Primary Education Projects in India are being extended to target working children (see box 1).
Greater emphasis is being given to child labor issues in relevant tra- ditional lending activities. For example, in educational projects, an in- crease in school flexibility (such as the third night shift introduced in Peruvian schools or school hours that vary depending on the season in agricultural areas as used in Colombia) helps prevent children from missing out on education due to less than full-time demands upon their
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K1 4
labor. Such measures increase school enrollment but also reduce child labor to the extent that they allow children to keep contact with the school and prevent them from becoming complete dropouts and full-time (of- ten underutilized) members of the labor force. Integrated Child Devel- opment Projects in India are strengthening preschool education to in- crease the likelihood that the children will continue in school, and a fol- low-up women and child development project is being expanded to in- clude measures to address child labor (see box 1). In upcoming projects, components are being considered to assist families with working chil- dren and to identify means to reduce harmful child labor directly.
Nonlending activities
There are also a number of areas where the Bank can increase its atten- tion to child labor in its nonlending activities, particularly in those coun- tries where child labor is seen as a serious problem, for example, consti- tuting a significant proportion of the total labor supply or representing a significant proportion of the child population.
More focus will be given to child labor issues in the Bank’s dialogue, helping governments explore and understand their economic and so- cial implications. The dialogue will aim to raise awareness among gov- ernment officials and politicians as well as other important actors re- garding the hazardous and abusive employment of children. This is important because the effectiveness of interventions will ultimately de- pend upon the existence of a social alliance against harmful child labor. The Bank’s SAS Region has created an informal consultative group on child labor, bringing together Bank staff, NGOs, and others, and has formed a partnership with UNICEF on this issue.
Programs aimed at improving the welfare of working children can- not be successful unless they are based on relatively accurate knowl- edge of the magnitude, nature, and effects of child labor, and the needs, constraints, and opportunities of the target group (children themselves as well as their families). In some cases, information can be relatively informal; for example, teachers can act as child labor “monitors” if they can report absences. The Bank can advise and support countries in set- ting and improving their information bases on child labor.
The Bank can further assist by offering ESW and, where appropriate, promoting research to identify options for reducing the harmful aspects of child labor. There is a considerable need for rigorous analysis on this issue, and the Bank is well placed to make an important contribution. Options are being explored to expand the Bank’s work in this area in collaboration with relevant organizations such as the ILO, UNICEF, and informed NGOs.
1 5
Box 1 Child labor initiatives in projects in India
District Primary Education Projects: These projects support primary school devel- opment by building national, state, and district managerial and professional capac- ity. They are already targeted to disadvantaged children, including girls and sched- uled caste and scheduled tribe students, and the government has recently directed program managers to make working children an additional target group. In addi- tion, the World Bank’s SAS Region is exploring with the government specific steps that can be taken by the projects to get children out of work and into schools while taking into account the need to help families raise their incomes to offset the loss of the children’s wages.
Integrated Child Development Scheme (ICDS): The three early childhood development projects aim to improve the nutrition and health status of children and can address child labor concerns by strengthening preschool education work and enhancing the chances that participants in ICDS would attend primary school. A Women and Child Development project will build on the on-going ICDS projects and during preparation would identify measures that could be taken to address selected child labor issues.
Proposed District Poverty Initiatives Project: Some of the analysis conducted as part of the social assessment for the project has examined child labor issues, and the problem is being explored further in the fieldwork that is currently under way.
Rural Women’s Development and Empowerment Project: This project aims to strengthen processes that promote economic development of women and create an environment for social change. One of its components supports income earning opportunities for poor women, which will raise their incomes and thus expand the opportunities for children to leave work and attend school. Specified ways in which this project could be used as a vehicle for directly assisting the families of working children will be explored.
In some cases, projects or project components can be designed to help child workers by reducing some of the harmful effects of their employment. Such projects would operate by defining areas or specific places where child labor was present and then supply services such as supplemental feeding programs or continuing education to the affected children and greater income opportunities for their par- ents. This approach in effect uses the existence of child labor as a targeting device for poverty alleviation and aims to enhance opportunities for the affected youth until the conditions that lead to child labor are improved. There are, however, some difficulties. First, the provision of such facilities would risk increasing the supply of child labor because child labor attracts these incentives. Second, such projects or components, unless carefully handled, could violate existing child labor laws in many countries if they supplied facilities on the premises in which children were employed. In some cases, child worker benefits could be targeted through intermediaries, such as local NGOs, and services provided elsewhere. The Bank has already taken the first steps in this direction in an urban development project in Brazil that supports a state program to provide training to children in low-income families, many of whom seek employment in the informal sector economy as “street children.”
Other pilot projects are being considered in several countries with substantial child labor. The Bank and the ILO have agreed to cooperate closely in designing such projects. Such lending interventions would be consistent with poverty alleviation, and new approaches need to be tested, balancing the urgency of reducing child labor with the legal and practical difficulties of designing effective interventions. One im- portant tool will be more explicit inclusion of child labor concerns in social assess- ments for projects where it is likely to be relevant.
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K1 6
In countries where child labor is seen as a problem, a discussion of relevant issues would also be specifically included in the CAS and in the PFPs. The Bank’s HD Network has identified child labor as one of the topics to receive particular attention in the CAS reviews. Where agreed with the government, the Bank can help design a strategy that integrates actions to discourage harmful child labor with other comple- mentary interventions, such as initiatives in education, health, family planning, and labor legislation and its enforcement, as well as with other policies designed to reduce poverty, such as employment creation and income generation programs.
Greater partnership with other concerned international organizations would also help. This would increase the breadth of the Bank’s knowl- edge and raise public exposure. There is already substantial coopera- tion and joint work with the ILO, and interest has been expressed by the latter in working with the Bank on harmful child labor and other labor standards. The two institutions are currently exploring the possibilities. There are also opportunities to work with other international organiza- tions, such as UNICEF, as well as calls for cooperation from concerned NGOs, such as Forgotten Children. The Bank is increasing its participa- tion in conferences on the topic—most recently at the Child Labor Con- ference in Amsterdam—to play a role in the formulation of the emerg- ing consensus on policies regarding child labor.
Partnership with NGOs may also yield gains for the Bank in borrow- ing countries. Apart from the potential for informal feedback on child labor practices in general, this could also give warning signals of the possibility of children being employed in Bank-financed projects. At present, such information, whether true or false, can be used as a means of discrediting the Bank. In a more cooperative context, it could be used by the Bank to act before problems develop.
Raising the awareness and sensitivity of Bank staff will be an essen- tial prerequisite of these initiatives. This is being achieved through semi- nars, workshops, etc. The Bank’s Africa Region held a child labor sym- posium in May 1997, and the SAS Region is planning one as well. The HD Week, held in March, with attendance of representatives from 29 partner institutions, included a module on child labor issues. These ac- tivities inform staff of the facts and major research on child labor and stimulate discussion on approaches that the Bank could take. Aware- ness should rise automatically if more focus on child labor is brought to bear in ESW and other Bank documents. The work in Thailand (see box 2), Côte d’Ivoire, and Ghana has led the way. Further work with UNICEF is planned as a basis for a project in Yemen.
1 7
Box 2. The case of Thailand
Child labor issues were explicitly addressed in a recent World Bank report on Thailand. The report noted the recent rapid decline in child labor, in which the labor force participation rate of those aged 13-14 years almost halved in the first half of the 1990s. This decline was attributed to a reduction in poverty re- lated to fast growth; an expansion of public education; and the maturity of suc- cessful family planning policies that reduced fertility from 6 children to just over 2 children per woman in the last 30 years.
However, the report noted that child labor remains significant and that more than 1.6 million children below the age of 15 are out of school. Most of these are from poor families. Many work under harsh conditions that adversely affect their mental and physical development and can lock them into poverty, thus perpetuating a vicious cycle. In fact, despite the growth of the Thai economy, the unemployment rates of the illiterates and those with less than primary edu- cation increased over time, though corresponding rates declined for all other education groups.
The report recommended that at the aggregate level policies for growth and further public support for basic education should continue. It also argued that targeted labor policies would be beneficial with respect to the improvement of employment conditions for children and that prohibition of harmful child labor should be vigorously enforced. Neither of these recommendations is surprising or unique to Thailand. However, where the report utilized cost/benefit analy- sis, specific recommendations emerged.
Using survey data and econometric analysis, the report argued that at an early age (below 12 years) dropping out of school is not synonymous with child labor. Children are pulled out of school because of direct costs of education rather than the need for work. At older ages (12 to 15 years) the need for income, that is, the opportunity cost of work, became more important. Based on these “income” and “price” effects, the report examined whether there is a market failure—in the sense that private cost/benefit considerations led to a premature exit of children from school. The results indicated that this was the case, and additional analysis established the extent to which market failure could be cor- rected with targeted subsidies or scholarships to poor households in urban ar- eas or in areas where enrollment was particularly low and child labor particu- larly prevalent.
Source: World Bank 1996 and Tzannatos 1996.
Changing operational practices
Outside groups have raised the possibility that the Bank use lending conditionalities to pressure governments to do more to curb child labor through stricter legislation, more stringent enforcement, or other means.
Issues of conditionality must be addressed in the light of the Bank’s mandate to promote economic and social development. The Bank can- not assume the role of an enforcement agency for the requirements of
C H I L D L A B O R : I S S U E S A N D D I R E C T I O N S F O R T H E W O R L D B A N K1 8
national or international law, which have no bearing on its specific op- erations. Requiring its member countries as a general proposition to enforce certain labor law standards regardless of their relevance to Bank operations would raise broader issues with respect to the Bank’s man- date and cannot be limited only to the subject of child labor. Therefore, the Bank can only impose conditionalities in this area to the extent that an absence of consistency with child labor standards undermines the execution or the developmental objective of its specific programs and projects. This would mean that a provision that the borrower would undertake to enforce its laws would be included in loan agreements where there are good reasons to believe that exploitative child labor with negative developmental effects may occur. Another approach in projects where it is felt there is a clear danger of child labor being employed would be to finance the appointment of an inspectorate. This would, however, add to project costs and could raise questions of jurisdiction if the inspectors were not recruited by the borrower itself.
The Bank can also offer other safeguards against the use of harmful child labor in Bank-financed projects. For example, in the SAS Region, task managers have been instructed to report any indication of child labor in Bank projects or related activities. The HD Network will dis- seminate the practices adopted by the SAS Region to other regions and will also monitor regional activities on child labor issues.
IV. Summary
Harmful child labor is a serious problem in a number of countries. It is primarily due to the conditions of poverty in which large segments of the population live and to a lack of access to quality education among the poor. Most countries have legislation governing the employment of children and are signatories to one or more international treaties on the subject, but enforcement is often lacking. In the longer run, programs to alleviate poverty and improve education and human services are essen- tial to reducing harmful child labor, and this is entirely consistent with the Bank’s central mandate on poverty reduction. However, reducing poverty is only likely to make significant inroads on child labor over the long term. The simple imperatives of widespread poverty mean that child labor will remain common in many countries for quite some time.
The Bank has identified a number of initiatives where actions to re- duce harmful child labor practices may be incorporated into its activi- ties, often in conjunction with other partners. It also recognizes the im- portance of including more on the topic in its training programs to sen- sitize staff to these concerns so that they can better incorporate activities to reduce child labor into their work. The HD Network will include at-
1 9
tention to child labor problems in its work program, and the issue will be addressed in CAS reviews where appropriate. Project components that address child labor issues will be highlighted in project appraisal documents and in presentations to the Board of Executive Directors.
A n
n ex
D is
tr ib
u ti
o n
a n
d l
ab o
r fo
rc e
p ar
ti ci
p at
io n
r at
es o
f ch
il d
re n
( 10
-1 4
y ea
rs )
b y
r eg
io n
R eg
io n
W o
rk in
g c
h il
d re
n R
eg io
n P
ar ti
ci p
at io
n ra
te s
D is
tr ib
u ti
o n
,0 00
(% )
C u
m u
la ti
v e
E as
te rn
A si
a 22
44 8
31 .6
% 31
.6 %
E as
te rn
A fr
ic a
32 .9
So u
th er
n A
si a
20 14
3 28
.4 %
60 .0
% W
es te
rn A
fr ic
a 24
.2 E
as te
rn A
fr ic
a 79
65 11
.2 %
71 .3
% M
el an
es ia
23 .9
W es
te rn
A fr
ic a
57 85
8. 2%
79 .4
% M
id d
le A
fr ic
a 21
.6 So
u th
-E as
te rn
A si
a 55
87 7.
9% 87
.3 %
E as
te rn
A si
a 20
.0 So
u th
A m
er ic
a 34
85 4.
9% 92
.2 %
So u
th er
n A
si a
14 .0
M id
d le
A fr
ic a
18 48
2. 6%
94 .8
% So
u th
A m
er ic
a 12
.8 W
es te
rn A
si a
11 09
1. 6%
96 .4
% So
u th
-E as
te rn
A si
a 11
.1 C
en tr
al A
m er
ic a
10 22
1. 4%
97 .8
% C
en tr
al A
m er
ic a
10 .3
N o
rt h
er n
A fr
ic a
98 2
1. 4%
99 .2
% C
ar ib
b ea
n 6.
8 C
ar ib
b ea
n 21
6 0.
3% 99
.5 %
W es
te rn
A si
a 6.
7 M
el an
es ia
14 7
0. 2%
99 .7
% N
o rt
h er
n A
fr ic
a 5.
8 So
u th
er n
A fr
ic a
10 0
0. 1%
99 .8
% P
o ly
n es
ia 4.
8 So
u th
er n
E u
ro p
e 84
0. 1%
99 .9
% So
u th
er n
A fr
ic a
4. 6
E as
te rn
E u
ro p
e 4
0. 0%
10 0.
0% So
u th
er n
E u
ro p
e 0.
8 P
o ly
n es
ia 1
0. 0%
10 0.
0% E
as te
rn E
u ro
p e
0. 1
N o
rt h
er n
E u
ro p
e 1
0. 0%
10 0.
0% N
o rt
h er
n E
u ro
p e
0. 0
A u
s- N
Z 0
0. 0%
10 0.
0% A
u s-
N Z
0. 0
W es
te rn
E u
ro p
e 0
0. 0%
10 0.
0% W
es te
rn E
u ro
p e
0. 0
M ic
ro n
es ia
0 0.
0% 10
0. 0%
M ic
ro n
es ia
0. 0
N o
rt h
A m
er ic
a 0
0. 0%
10 0.
0% N
o rt
h A
m er
ic a
0. 0
A si
a 49
28 7
69 .5
% 69
.5 %
A si
a 15
.3 A
fr ic
a 16
68 0
23 .5
% 93
.0 %
A fr
ic a
22 .0
A m
er ic
as 47
23 6.
7% 99
.7 %
A m
er ic
as 7.
9 O
ce an
ia 14
8 0.
2% 99
.9 %
O ce
an ia
6. 9
E u
ro p
e 89
0. 1%
10 0.
0% E
u ro
p e
0. 3
W o
rl d
70 92
7 10
0. 0%
W o
rl d
13 .7
S ou
rc e:
A d
ap te
d f
ro m
G ro
o ta
er t
an d
K an
b u
r 19
95 a
n d
b as
ed o
n I
L O
1 99
3: (
i) r
et u
rn o
f a
sp ec
ia l I
L O
q u
es ti
o n
n ai
re s
en t
to m
o re
th an
2 00
c o
u n
tr ie
s an
d t
er ri
to ri
es in
A p
ri l 1
99 2;
( ii
) L
A B
O R
ST A
T d
at ab
as e;
( ii
i) p
re li
m in
ar y
I L
O e
st im
at es
a n
d p
ro je
ct io
n s
o f
ec o
n o
m ic
al ly
a ct
iv e
p o
p u
la ti
o n
; an
d (
iv )
U n
it ed
N at
io n
s P
o p
u la
ti o
n D
iv is
io n
, “ Se
x an
d A
g e
D is
tr ib
u ti
o n
o f
th e
W o
rl d
’s P
o p
u la
ti o
n s:
T h
e 19
92 R
ev is
io n
( 19
50 -2
02 5)
.” N
ew Y
o rk
.
REFERENCES
Canagarajah, R.S., and H. Columbe. 1997. “Child Labor in Ghana.” Africa Technical Family: Human Development 3. World Bank, Washington, D.C.
Grootaert, Christiaan. 1997. “Child Labor in Côte d’Ivoire: Incidents and Approaches.” Social Policy and Resettlement Division, Environment Dept. World Bank, Washington, D.C.
Grootaert, Christiaan, and Ravi Kanbur. “Child Labor: A Review.” World Bank Policy Research Working Paper 1454. Washington, D.C.
International Labor Office 1996a. “Child Labour: Targeting the Intolerable.” International Labor Conference. 86th Session 1996. Geneva.
________. 1996b. Child Labour: What is to be Done? Geneva. Melchior, Arne. 1996. “Child Labour and Trade Policy.” In Grimsrud, B. and A.
Melchior, eds. Child Labour and International Trade Policy. Oslo: Institute for Applied Social Science and Norwegian Institute of International Affairs.
Save the Children. 1996. Small Hands: Children in the Working World. London. Shihata, Ibrahim. 1996. “The World Bank’s Protection and Promotion of
Children’s Rights.” International Journal of Children’s Rights. vol. 4. pp. 383- 405.
Siddiqi, Faraaz, and Harry Patrinos. 1995. “Child Labor: Issues, Causes and Interventions.” Human Resources Development Working Paper No. 56. World Bank, Washington, D.C.
Tzannatos, Zafiris. 1996. “Child Labor and School Enrollment in Thailand in the 1990s.” Social Protection, Human Development Network. World Bank, Washington, D.C.
UNICEF. 1993. The Progress of Nations. Geneva. WHO. 1987. Children at Work: Special Health Risks. Geneva. World Bank. 1995. World Development Report 1995. Workers in an Integrating World.
New York: Oxford University Press. World Bank. 1996. “Thailand: Growth, Poverty and Income Distribution: An
Economic Report”. East Asia and Pacific Region, Report No. 15689-TH. Washington, D.C.
World Bank. 1997. “Social Development and Results on the Ground.” Task Force Report. Washington, D.C.
2 2
