43761 1.2003 Child labour indicators used by the April,s UCW Project: Serie an explanatory note Papergn A.Cigno Workit L. Guarcello Projec Y. Noguchi S. Lyon Work F. C. Rosati Children's April 2003 Understanding Child labour indicators used by the UCW Project: an explanatory note A. Cigno* L. Guarcello** Y. Noguchi*** S. Lyon** F. C. Rosati** Working Paper April 2003 Understanding Children's Work (UCW) Project University of Rome "Tor Vergata" Faculty of Economics V. Columbia 2 00133 Rome Tor Vergata Tel: +39 06.7259.5618 Fax: +39 06.2020.687 Email: info@ucw-project.org As part of broader efforts toward durable solutions to child labor, the International Labour Organization (ILO), the United Nations Children's Fund (UNICEF), and the World Bank initiated the interagency Understanding Children's Work (UCW) project in December 2000. The project is guided by the Oslo Agenda for Action, which laid out the priorities for the international community in the fight against child labor. Through a variety of data collection, research, and assessment activities, the UCW project is broadly directed toward improving understanding of child labor, its causes and effects, how it can be measured, and effective policies for addressing it. For further information, see the project website at www.ucw-project.org. This paper is part of the research carried out within UCW (Understanding Children's Work), a joint ILO, World Bank and UNICEF project. The views expressed here are those of the authors' and should not be attributed to the ILO, the World Bank, UNICEF or any of these agencies' member countries. * University of Florence ** UCW-Project and University of Rome "Tor Vergata" ***IPEC Child labour indicators used by the UCW Project: an explanatory note Working Paper April 2003 ABSTRACT This document briefly explains and illustrates the set of indicators developed for the Country Statistics found on the UCW project website. The indicators are not the theoretical "best" of the many that could be constructed on child work, but rather those that are feasible given the available data. As such, the indicators are unable to fully capture the various dimensions of the child work phenomenon. They do, however, provide at least partial answers to the following key questions: 1) What is child work? 2) How widespread is it? 3) What are its characteristics? 4) How damaging is it for the child's health and future earning capacity? 5) Which household, local or national characteristics make it more likely that a child will work? The data used to develop the indicators are from household surveys conducted mainly by the World Bank, under its Living Standards Measurement Study (LSMS) programme, the ILO International Programme on the Elimination of Child Labour (IPEC), under its Statistical Information and Monitoring Programme on Child Labour (SIMPOC), and UNICEF, under its Multiple Indicator Cluster Survey (MICS) programme. These surveys yield a wide variety of data in areas such as education, employment, health, expenditure, and consumption that relate to child work. They do not, however, provide information on unconditional worst forms of child work, such as child prostitution and child slavery, for which different data collection methodologies are required. Child labour indicators used by the UCW Project: an explanatory note Working Paper April 2003 CONTENTS 1. Introduction ...........................................................................................................................1 2. What is child labour?.............................................................................................................4 3. How widespread is child labour? ........................................................................................4 4. What are the characteristics of children's work? ..............................................................5 5. How damaging is work for the child's health and future earning capacity? ................5 6. What makes it more likely that a child will work?............................................................7 1 UCW WORKING PAPER SERIES, APRIL 2003 1. INTRODUCTION 1. This note briefly explains and illustrates the set of indicators developed for the Country Statistics found on the UCW project website. All these indicators are meant to contribute to the action against child labour by a better understanding of work carried out by children. First of all, it is necessary to clarify one point: not all work done by children (i.e. girls and boys under 18 years of age) is equivalent to child labour that needs elimination. Certain forms and situations of work performed by children or adolescents that does not affect their health and personal development or interfere with their schooling, may be regarded as some-thing positive. Such work would contribute to children's development and to the welfare of their families; it helps children to acquire skills and prepare to be productive members of society during their adult life. In no way can such activities be equated with child labour that needs to be eliminated. 2. There has been considerable debate for instance in development circles in the English language over the use of terms: "child work" and "child labour", and the distinctions between the two. A consensus is gradually emerging, however, that `child work' or `children's work' is seen as a general term covering the entire spectrum of work and related tasks performed by children, and `child labour' as a subset of children's work that is injurious to children and that should be targeted for elimination. There is also growing recognition that there are certain intolerable, or `unconditionally worst', forms of child labour that constitute especially serious violations of children's rights, and that should be targeted as a priority for immediate action. 3. It is less easy, however, to reach agreement concerning where the line between benign forms of work, on one side, and child labour for elimination, on the other, is drawn. This question is by no means merely academic, as underlying it is the more basic question of where precisely exists child labour that should be eliminated. Seeking answer to this question is the beginning of action to tackle child labour and should be an integral part of the comprehensive measures against it. Maybe there is no single answer that fits all countries and all circumstances. International instruments have been adopted as an effort to guide this quest. 4. The UN Convention on the Rights of the Child (CRC) recognises the children's right to be protected from forms of work that are likely to be hazardous or to interfere with the child's education, or to be harmful to the child's health or physical, mental, spiritual, moral or social development. In order to achieve this goal, the CRC also calls on States parties to set minimum ages for admission to employment, having regard to other international instruments. The ILO Conventions No. 138 (Minimum Age) and No. 182 (Worst Forms of Child Labour) are no doubt the most important of them. 5. Figure 1 illustrates the basic distinctions embodied in Conventions Nos. 138 and 182 [ILO Global Report 2002, Figure 2 and Paragraphs 25]. It shows that it is the interaction between the type of work and the age of the child involved that defines the boundaries of child labour for effective abolition. Thus, our search for information should target as child labour 1) all forms of work carried out by children below a minimum cut-off age (13 years in general, and 12 years in less developed countries), 2) all forms except `light work' carried out by children below a second higher cut-off CHILD LABOUR INDICATORS USED BY UCW PROJECT: AN 2 EXPLANATORY NOTE Note: The size of the areas in the above figure does not represent quantitative proportions. Source: Global Report under the Follow-up to the ILO declaration on fundamental principle and rights at Work, ILO, Geneva, 2002 age ( this is usually the minimum working age, at least 15 years, but may be14 years in less developed countries), and 3) all `worst forms' of child labour, including hazardous types of work, carried out by children of any age under 18. 6. The above explanation shows that it is not such an easy task to develop an indicator that depicts "child labour" either in a given country or globally. The ILO's Global Estimate [Every Child Counts, ILO, Geneva, 2002] for instance used a set of cut-off criteria for all countries so as to arrive at estimates for the whole world, even though the line drawn to mark the limit of child labour for elimination varies from one country to another. Even on a country level of statistics, the available data tend to be on work performed by children rather than exactly on child labour, because the latter depends on how the line is drawn between acceptable and prohibited work for children, often based on the age of the child, types of work as well as the conditions in which the work is carried out. In fact, a better understanding of the circumstances in which children work may shed a new light to the way the line is drawn and contribute to a revision of legislative provisions. 7. All these details might give an impression that child labour is not scientifically measurable. We should point out that the previous estimate made by the ILO that 250 million children were working (taking 5-14 age group in economic activity) was not so far off from the new global estimate established after the above conceptual differences in considerations. After all, children under the minimum working age should not, in principle, be at work but be in school. Their work may be allowed as exceptions to the rule. In turn, children who are over that minimum age but still under 18 are allowed to work, except for in hazardous work or other worst forms. 3 UCW WORKING PAPER SERIES, APRIL 2003 Therefore, if you take the figure of working children under the minimum working age (usually 14, 15 or 16), it needs adjustment both ways: subtracting the exceptionally permissible work by children under the minimum age, and adding the prohibited forms of work by those over that age. If that figure is to be taken as a proxy of the "child labour" it should be an informed decision pending further refinement according to the age and work, which could become possible with more investment in the effort to gather more information. 8. It is the hope of UCW and its participating organisations that by producing more information on work performed by children and its conditions, characteristics and context, a better picture will be obtained on child labour that needs to be eliminated. Especially by presenting the information in a form of objective and comparable indicators, it is hoped to contribute to following up the evolution over time, and also allow cross-national comparison. 9. The indicators developed provide at least partial answers to the following key questions: 1. What is child labour? 2. How widespread is it? 3. What are its characteristics? 4. How damaging is it for the child's health and future earning capacity? 5. Which household, local or national characteristics make it more likely that a child will work? 10. The data used to develop the indicators are from household surveys conducted by the World Bank, under its Living Standards Measurement Study (LSMS) programme, the ILO International Programme on the Elimination of Child Labour (IPEC), under its Statistical Information and Monitoring Programme on Child Labour (SIMPOC), by UNICEF, under its Multiple Indicator Cluster Survey (MICS) programme and by the National Statistical Office. These surveys yield a wide variety of data in areas such as education, employment, health, expenditure, and consumption that relate to child work. They do not, however, provide information on unconditional worst forms of children's work, such as child commercial sexual exploitation and child slavery, for which different data collection methodologies are required. 11. Information on how the data were collected and some indication of their reliability can be found on the relevant websites of the three agencies (http://ilo.org/childlabour/simpoc; http://child info.org; and http://www.worldbank.org/ lsms). A discussion of these issues can also be found in B. Grimsrud, Developing New Strategies for Understanding Children's Work and its Impact: A Comparison of Survey Instruments for Collecting Data on Child Work. CHILD LABOUR INDICATORS USED BY UCW PROJECT: AN 4 EXPLANATORY NOTE 2. WHAT IS CHILD LABOUR? 12. The Country Statistics employ the broader concept of children's work rather than the narrower one of child labour, because of, as explained in the introduction, the difficulties in drawing a statistical line between acceptable forms of work, on the one hand, and harmful forms that need to be eliminated, on the other. 13. Children's work can be thought of as any form of economic activity performed by children. Economic activity, in turn, as defined by the UN System of National Accounts (1993 Rev. 3), is a broad concept that encompasses most productive activities by children, including unpaid and illegal work, work in the informal sector, and production of goods for own use. 14. Not all children's work is equivalent to child labour. Child labour refers only to negative or undesirable forms of work that should be eliminated in accordance with ILO Conventions No. 138 (Minimum Age) and No. 182 (Elimination of Worst Forms) and the UN Convention on the Rights of the Child (CRC). 15. The Country Statistics cover the 5-14 years age range. They do not include the 15-17 years age group, because regular work is acceptable for these children, and because the household data offer little information about forms of work that constitute child labour for this group. 16. For each country, the indicators give the total number of children (Total number of children, by sex and age), the proportion of them who work (Percentage of children economically active only, or economically active and studying, by sex and age), and the proportion who study (Percentage of children attending school, by sex and age). As data become available, tables showing a measure of children's work that includes, beside economic activity, also household chores above a given threshold will be added. 17. It should not be forgotten that "working" does not include helping in the home, which is a non-economic activity, and therefore outside the `production boundary', according the UN System of National Accounts (1993 Rev. 3). However, the performance of domestic chores also has a direct bearing on child welfare, and therefore is also included in the Country Statistics. These activities may in fact conflict with formal education just as much as, or in the case of girls even more than, working in the fields or helping in the family shop. But since questions about domestic chores are not put in the same way in all questionnaires, and domestic chores is typically not an exhaustive activity category, indicators relating to this type of activity are kept separate from information about child work (Percentage of children who carry out household chores, by sex and age; Percentage of children who carry out household chores, by sex and type of activity; Percentage of children who carry out household chores, by sex, type of activity and household chore). 3. HOW WIDESPREAD IS CHILD LABOUR? 18. Since working and attending school are not mutually exclusive activities, and do not exhaust the range of possibilities, children are grouped in four non-overlapping activity categories: those who work only, those who study only, those who do both, and those who do neither (Percentage of children economically active only, by sex and age; Percentage of children studying only, by sex and age; Percentage of children economically active and studying, by sex and age; Percentage of children involved in no activity, by sex and age). In many countries, the proportion of children falling into each these four activity categories varies among regions and between rural and urban 5 UCW WORKING PAPER SERIES, APRIL 2003 areas (Percentage of children by sex, type of activity, and regions; Percentage of children by sex, type of activity, and areas). 19. The number of children forming part of the last category, that of children reported by their parents as doing nothing, can be quite substantial, and varies a great deal from country to country (from around eight percent of the total in Brazil, to over a quarter in India). To some extent, this is a measure of our ignorance. Some of these children may be actually doing nothing, perhaps because their parents are too poor to send them school, and there is nothing else that the children can usefully do. But this hypothesis is not very plausible where children of 13-14, or girls of any age with younger siblings, are concerned. In some cases, these reportedly idle children may be actually helping in the home, and either this information is not reported because the questionnaire did not ask for it, or it is kept separate from estimations of child work prevalence. In other cases, it may be that the child is engaged in seasonal work, but the questionnaire asked only what the child was doing at the time of the interview. Finally, parents may be deliberately concealing the fact that their child is working, either because child work is illegal, or because their child is engaged in some illegal form of work. These are the children most at risk. 4. WHAT ARE THE CHARACTERISTICS OF CHILDREN'S WORK? 20. For children who do work, the indicators say whether they are engaged in agriculture, manufacturing, etc. (Percentage of children, by sex and branch of activity), and whether they are unpaid workers, wage workers, or self-employed (Percentage of children, by sex and modality of employment). In many countries, child work is mostly unpaid, but there are exceptions (in Venezuela, for example, 44 percent of all working children between the ages of 10 and 14, nearly half if male, receive a wage.) 21. For a limited number of countries, the indicators tell us how many hours children spend working or helping in the home (Children working only: average working hours by sex and age; Children economically active and studying: average working hours by sex and age; Time spent by children performing household chores). This information is very important to determining the intensity of work. A little light work is not necessarily bad for the child's health, and need not interfere with formal education. Long working hours, on the other hand, are likely to have more serious health and developmental consequences on the child. The evidence suggests that the latter case is more common. In nine of the 13 countries where data are available, children who work (and do not attend school) put in an average of at least 40 hours of work per week, i.e., as much or more as a full-time adult worker in the industrial world. In two of the countries, children's average workweek was 55 hours long. 5. HOW DAMAGING IS WORK FOR THE CHILD'S HEALTH AND FUTURE EARNING CAPACITY? 22. The health of children is important for its own sake, and also because it conditions their future ability to earn a living. A series of tables provide information on child illness, one important measure of child health. The tables show how illness incidence varies with children's activity status (Percentage of children with health problems, by sex and type of activity; Percentage of children with health problems, by sex, type of activity and duration of illness), and, among working children, how illness incidence varies by different forms of employment (Percentage of children CHILD LABOUR INDICATORS USED BY UCW PROJECT: AN 6 EXPLANATORY NOTE economically active only with health problems, by sex and modality of employment; Percentage of children economically active only with health problems, by sex, modality of employment and duration of illness; Percentage of children economically active and studying with health problems, by sex and modality of employment). 23. These tables do not suggest a clear negative relationship between child work and child health. Indeed, in 13 of the 20 countries where data are available, the rate of reported illness is lower among children who only work compared to those who only attend school. Rather, the tables suggest that it is certain types of work, and not work per se, that are bad for the child's health. In Brazil, for example, reported illness among total working children (18 percent) is lower, but reported illness among the subset of working children who work for wages (28 percent) is higher, than among children who attend school (21 percent). In Tajikistan, while the rate of reported illness is only eight percent among all working children, it rises to 58 percent among children who are self-employed. In Zambia, health problems are reported for 20 percent of total working children, but for 80 percent of children working at home on a piecework basis. 24. But the relationship between child work and health is complex, and often difficult to disentangle empirically, and therefore the tables on child work and health must be read with caution. The negative impact of child work on health, for example, may be obscured by the selection of the healthiest children for work. Health perceptions may also differ across population groups, and levels of reported illness among working children and non-working children may be affected by different levels of consciousness of illness. Much of the relationship between child health and work is dynamic (i.e., current health is affected by past as much as present work, and current work affects future as much as present health), a fact not captured by measuring reported illness over a short period. 25. Nutritional status is another important indicator of child health. Various anthropometric indices, based on simple height and weight measurement (height for age, weight for age, weight by height squared, etc.), can be used to gage a child's nutritional status. The tables Body mass distribution, by sex and type of activity and Body mass distribution, by age, type of activity and sex show how BMI (the body- mass index, equal to weight in kilograms, divided by height in meters squared) varies with the child's activity status. Judging by these tables, the nutritional status of working children appears no worse than that of children attending school, though data on child nutrition and work are available for only a limited number of countries and, like data on child illness and work, must be interpreted with caution. Tables showing one additional indicator of nutrition, BMI-for-age (BMI-for-age, by sex and type of activity), will be provided in the near future. 26. School attendance and educational attainment also bear an obvious link to a child's future earning capacity. Child work, not surprisingly, negatively affects both. The table Percentage of Children economically active and studying, by sex and age indicates that work frequently interferes with school attendance. The attendance rate of working children is 50 percent or less in 16 of the 31 countries where data are available. And in five Sub Saharan African countries (Burundi, Central African Republic, Gambia, Ghana and Guinea) less than five percent of working children attend school. Tables on educational attainment, to be provided shortly, show that the amount of formal schooling received by working children is much less than that of their non-working counterparts. 7 UCW WORKING PAPER SERIES, APRIL 2003 6. WHAT MAKES IT MORE LIKELY THAT A CHILD WILL WORK? 27. Summary indicators cannot establish causalities, but can help detect important correlates of child work, and should at least permit meaningful comparisons across countries. Wage rates and household income are obviously important correlates of child work, but government policy is also very relevant. 28. Cross-tabulating income quintiles with child activities (Percentage of children, by household income quintile and sex; Percentage of children by household income quintile, sex and type of activity) reveals a broad tendency for the proportion of children who work to decrease, and that of children who attend school to increase, as household income increases. In the case of Brazil, for example, 20 per cent of children in the bottom income quintile work only, or work and study, as against only five percent in the top quintile. The percentage of children who study only rises from 65 in the bottom income quintile, to 81 in the top one. By contrast, nutrition (Body mass distribution by household income quintile, type of activity and sex) and health status (Percentage of children with health problems by household income quintile, sex and type of activity) do not appear to improve systematically with household income. This is coherent with the observation that working children have better nutritional and health status than children who only attend school, and confirms our earlier remarks about inequality of distribution within the family. 29. Finally, the educational attainment of the parents may be a factor in determining whether a child will work, or go to school. In particular, there is a substantial literature claiming that the mother's education has a positive effect on the probability that a child will study, and generally on the share of family resources that is expended on children. There is also evidence, however, that the daughters of better educated women are more likely to stay at home to look after younger siblings, because their mothers are more likely to find outside employment. 30. A series of tables relating the child's activity to the educational attainment of the household's head, usually the father or both (Percentage of children economically active only, by household head's level of education and children's age; Percentage of children studying only, by household head's level of education and children's age; Percentage of children economically active and studying, by household head's level of education and children's age; Percentage of children that carry out household chores, by household head's level of education and children's age; Percentage of children involved in no activities, by household head's level of education and children's age) show that the children of better educated parents are by and large more likely to go to school, and less likely to work.