Evolution of the relationship between child labour and education since 2000 Evidence from 19 developing countries

Understanding Children s Work Programme Working Paper Series, March 215 Evolution of the relationship between child labour and education since 2 Evidence from 19 developing countries L Guarcello S Lyon C Valdivia

Evolution of the relationship between child labour and education since 2 Evidence from 19 developing countries L Guarcello * S Lyon * C Valdivia * Working Paper March 215 Understanding Children s Work (UCW) Programme Villa Aldobrandini V Panisperna 28 184 Rome Tel: +39 6434128 Fax: +39 66792197 Email: info@ucw-projectorg As part of broader efforts towards 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) Programme in December 2 The Programme 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 Programme 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 wwwucw-projectorg This paper is part of the research carried out within UCW (Understanding Children's Work), a joint ILO, World Bank and UNICEF Programme 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 * Understanding Children s Work (UCW) Programme and University of Rome Tor Vergata

Evolution of the relationship between child labour and education since 2 Evidence from 19 developing countries Working Paper March 215 ABSTRACT Children s work is one of the main obstacles to Education for All (EFA) Children s work harms children's ability to enter and survive in the school system, and makes it more difficult for children to derive educational benefit from schooling once in the system At the same time, education is a key element in the prevention of children s work Understanding the interplay between education and children s work is therefore critical to achieving both EFA and child labour elimination goals The question of most interest in terms of policy, not captured by the static picture of children s work, is the direction in which countries are moving in terms of children s work and schooling, ie, whether a higher or lower proportion of children are working and/or studying over time The current Report is aimed at documenting the trends in children s time allocation between work and schooling in a sample of 19 developing countries since 2

Evolution of the relationship between child labour and education since 2 Evidence from 19 developing countries CONTENTS 1Introduction 1 2Data sources and description of indicators 1 3Evolution of children s time use patterns 4 4Trends in the time intensity of children s employment 11 5Trends in the status of children s employment 15 6Relation between grade-for-age and children s work 18 Statistical appendix 2 LIST OF FIGURES AND TABLES Figure 1 Recent years have seen a fall in children s employment across all 19 countries 4 Figure 2 The overall decline in children s employment frequently masks very different patterns for children combining employment and school and children working only 6 Figure 3 Children who are excluded from schooling are diminishing as a share of total children in employment 8 Figure 4 The decline in the incidence of children s employment has been accompanied by a reduction in the time intensity of employment in most countries 11 Figure 5 Changes in the time intensity of children s differ between children combining employment and schooling and children only in employment 13 Figure 6 There are important differences between children only in employment and children combining employment and schooling in terms of work setting and work modalities 16

Figure 7 The differences in work arrangements between children only in employment and children combining employment and schooling appear to be growing over time 17 Figure 8 School attendance is consistently much lower for children in employment than for their non-working peers 18 Figure 9 Children in employment also lag behind in terms of grade for age 19 Figure A1 The general decline in children s employment has been accompanied by a general rise in school attendance 2 Figure A2 The decline in the incidence of children s employment has been accompanied by a reduction in the time intensity of employment in most countries 22 Table A1 Changes in the age of children combining employment and schooling and of children only in employment, base year and most recent year, by country(a) sex, residence and age range 23 Table A2 Changes in average weekly working hours in employment, children only in employment and children combining employment and schooling, base year and most recent year, by country(a) sex, residence and age range 24 Figure A3 The overall decline in children s employment frequently masks very different patterns for children combining employment and schooling and children only in employment 25 Figure A4 The sectoral composition of employment also differs between children in employment only and children combining employment and schooling 27 Figure A5 There are no clear patterns in terms of how the sectoral composition of employment is changing for over time for the two groups of children in employment 28

1 UCW WORKING PAPER SERIES, MARCH 215 1 INTRODUCTION 1 Children s work is one of the main obstacles to Education for All (EFA) Children s work harms children's ability to enter and survive in the school system, and makes it more difficult for children to derive educational benefit from schooling once in the system At the same time, education is a key element in the prevention of children s work Understanding the interplay between education and children s work is therefore critical to achieving both EFA and child labour elimination goals 2 The question of most interest in terms of policy, not captured by the static picture of children s work, is the direction in which countries are moving in terms of children s work and schooling, ie, whether a higher or lower proportion of children are working and/or studying over time The current Report is aimed at documenting the trends in children s time allocation between work and schooling in a sample of 19 developing countries since 2 3 The remainder of the Report is structured as follows Section 2 provides an overview of the data sources and indicators utilized in the Report Section 3 reports trends in division of children s time between work and schooling and Section 4 reports trends in the time intensity of children s work Section 5 assesses trends in children s status in employment and how these differ between children only in employment and children combining employment and schooling Finally, Section 6 reports another education indicator, grade-for-age, to show how children s employment impacts on their educational opportunities 2 DATA SOURCES AND DESCRIPTION OF INDICATORS 4 The present Report makes use of datasets from nationally-representative ILO SIMPOC surveys, Demographic and Health (DHS) surveys, UNICEF MICS surveys and national household and labour force surveys The selection of the countries (and data) included in the Report was guided by data considerations and by the objective of broad geographic representation The Report includes countries with two comparable data points in most cases around ten years apart, with the first point referring to the early 2s The 19 countries covered in the Report, and the corresponding data sources and reference period for each, are detailed in Table 1

2 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Table 1 Data sources Countries Base year survey Most recent year survey Data limitations Asia Bangladesh Child Labour Survey (CLS) 22-23 Labour Force Survey (LFS) 25-26 Information on household chores is not available for those children who are in employment, leading to an underestimation of the number of children in household chores For the year 25-26, information on sector and status of employment is collected only for children in employment exclusively National Sample Survey Office (NSSO) 24 NSSO 29-1 Does not capture children who combine work and school LFS 21 LFS 211 From the age of 1 Multiple Indicator MICS 211 Up to the age of 14 Cluster Survey (MICS) 2 Africa MICS 2 Demographic and Health Survey (DHS) 21 Up to the age of 14 Cameroon MICS 2 MICS 211 Up to the age of 14 Second Integrated Household Survey (IHS2) 24 Third Integrated Household Survey (IHS3) 21-11 MICS 2 DHS 21-11 Up to the age of 14 MICS 2 MICS 21 Up to the age of 14 MICS 2 MICS 21 Up to the age of 14 National Household Survey (NHS) 25-6, CLS 211-212 Information on weekly working hours from CLS 211 excludes individuals working in subsistence agriculture LFS 25 LFS 28 Latin America Bolivia Encuesta Continua de Hogares (ECH) 23-24 ECH 29 Pesquisa Nacional por PNAD 211 Amostra de Domicílios (PNAD) 21 CLS 21 CLS 211 Encuesta Nacional de Empleo, Desempleo y Subempleo (ENEMDU) 21 ENEMDU 211 Encuesta de Hogares des Propositos Multiples (EHPM) 21 Encuesta Nacional de Condiciones de Vida (ENCOVI) 2 EHPM 211 ENCOVI 211 Encuesta Nacional de ENOE-MTI 211 Ocupación y Empleo con modulo the trabajo infantil (ENOE-MTI) 27

3 UCW WORKING PAPER SERIES, MARCH 215 5 The datasets were selected to ensure data comparability across time within countries However, survey instruments, reference years and reference periods differ somewhat across countries This means that caution should be exercised in drawing cross-country comparisons 6 Moreover, while in general the descriptive statistics are available for the group of children up to the age of 17 years, for some countries, ie,,, Cameroon,,, and, the information is available only for children up to the age of 14 years Another exception is, where the information on employment is only available starting from the age of 1 Other survey-specific data limitations relating to Bangladesh, and are detailed in Table 1 7 The indicators used in the Report are based on the definitions provided in Panel 1 The Report relies primarily on the concept of children s employment as a measure of their involvement in work (the concept of household chores is dealt with in Panel 2) Children s employment is a broader concept than child labour, covering all market production and certain types of non-market production (principally the production of goods and services for own use) It includes forms of work in both the formal and informal economy; inside and outside family settings; work for pay or profit (in cash or in kind, part-time or full-time), or as a domestic worker outside the child s own household for an employer (with or without pay) Panel 1 Definitions Employment: a child is considered to be in employment if he/she has worked during the week prior to the survey for at least one hour for pay (or without pay), profit, in kind, or family business A person is also considered to be in employment if was not working but had a job to go back to School attendance: a child is considered to be attending school if he/she is currently attending school at the time of the survey Activity status: the identification of children in employment and children in school allows for the disaggregation of the child population into four non-overlapping activity groups - children in employment exclusively, children attending school exclusively, children combining school and employment and children doing neither Weekly working hours: refers to the average weekly working hours worked in the main employment Status in employment: the indicator of status in employment distinguishes between the three main categories (a) wage (or paid) employee, (b) self-employed, (c) unpaid family workers, and (d) other information not elsewhere classified Sector of employment: the indicator sector of employment divides employment into different groups of economic activity The groupings used in the present Report are (a) agriculture, (b) manufacturing, (c) services and (d) the category other, which includes mining, constructions, electricity gas and water, and activities not classifiable by economic activity Household chores: Children in other productive activities includes children who perform unpaid household services, that is, the production of domestic and personal services by a household member for consumption within their own household, commonly called household chores

4 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 3 EVOLUTION OF CHILDREN S TIME USE PATTERNS 8 The ILO global estimates for the 12-year period beginning in 2 indicate that this period was one of significant progress against child labour There were almost 78 million fewer child labourers aged 5-17 years at the end of the period than at the beginning In relative terms, the age of children in child labour fell from 16 in 2 to less than 11 in 212 9 Trends in children s employment in our 19 countries are largely consistent with this global picture Children s involvement in employment for the entire 7-17 years age range has fallen in all 19 countries In many cases the decline has been dramatic In, for instance, children s employment fell from 7 to, and in from 73 to 41 In, the share of children in employment fell more than three-fold, from 24 to seven In absolute terms, the decline in children s employment from 11 to less than seven in populous is especially noteworthy 1 Most children in employment in the 7-11 years and 12-14 years age ranges are also in child labour in accordance with international legal standards 2 and trends for these age ranges are therefore of particular interest We see from Figure 1(b) and Figure 1(c) that there was a general decline in children s employment for both the 7-11 and 12-14 year-olds in the 19 countries Indeed, only one of the countries Bangladesh experienced an appreciable rise in children s employment, and this rise was limited to the 7-11 years age group But despite this progress, the prevalence of employment in these age groups remains far from negligible This is especially the case for the Sub-Saharan Africa countries included in our sample Over one-fourth of 7-11 year-olds, for example, remain in employment in, Cameroon,,, and 11 Children s employment also fell among 15-17 year-olds in all countries in our sample except Bangladesh 3 This result is more difficult to interpret, however, as child labour among 15-17 year-olds is limited to those in hazardous employment rather than those in employment generally The general drop in employment could represent a (desirable) drop in hazardous employment but it could also reflect (undesirable) difficulties experienced by young people in finding decent jobs in the labour market 12 The general decline in children s employment was accompanied by a general rise in school attendance in our 19 sample countries across all age ranges (Appendix Figure A1) We now turn to changes in the interplay between employment and schooling in the 19 countries Figure 1 Recent years have seen a fall in children s employment across all 19 countries 2 ILO Convention No 138 (Minimum Age) (C138) calls on Member States to set a general minimum age for admission to work or employment of at least 15 years of age (Art 23) (14 years of age in less developed countries), and a higher minimum age of not less than 18 years for employment or work which by its nature or the circumstances in which it is carried out is likely to jeopardise the health, safety or morals of young persons, ie, hazardous work (Art 31) However, the Convention states that national laws or regulations may permit the employment or work of persons from 13 years of age (12 years in less developed countries) on light work which is (a) not likely to be harmful to their health or development; and (b) not such as to prejudice their attendance at school, their participation in vocational orientation or training programmes approved by the competent authority or their capacity to benefit from the instruction received (Art 7) Owing to the flexibility clauses in C138 left to the discretion of the competent national authority, the specific minimum working ages vary somewhat across the 17 sample countries 3 Data were not available for this age group in countries where the MICS survey was used as the data source

Bolivia Camerron Bangladesh 153 242 23 184 225 168 315 299 278 276 39 263 31 374 359 5 3 373 536 498 475 623 63 583 578 546 Bolivia Camerron Bangladesh 79 57 49 92 65 52 145 19 167 171 131 137 118 19 36 258 28 248 219 212 345 9 375 417 377 435 473 461 465 535 522 524 54 542 724 672 79 83 Bolivia Camerron Bangladesh 11 9 44 47 17 44 35 15 16 48 38 85 66 67 45 65 77 161 167 121 12 186 15 19 273 265 274 34 276 277 343 38 377 444 614 595 716 682 5 UCW WORKING PAPER SERIES, MARCH 215 Percentage of children in employment, base year and most recent year, by age group and country (a) (a) 7-11 years age group 9 8 7 6 5 3 2 1 year year 1 Latin America and Caribbean Sub Saharan Africa Asia (b) 12-14 years age group 9 8 7 6 5 3 2 1 year year 1 Latin America and Caribbean Sub Saharan Africa Asia (c) 15-17 years age group 9 year 8 7 year 1 6 5 3 2 1 Latin America and Caribbean Sub Saharan Africa Asia Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1)

6 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Figure 2 The overall decline in children s employment frequently masks very different patterns for children combining employment and school and children working only Percentage of children combining employment and schooling and of children only in employment, base year and most recent year, by age range and country (a) (a) 7-17 years (b) 6 5 Only employment 3 2 Employment and schooling 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia El Salvador Sierra Leone Cameroon (b) 7-11 years 6 5 Only employment 3 Employment and schooling 2 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia El Salvador Sierra Leone Cameroon

7 UCW WORKING PAPER SERIES, MARCH 215 Figure 2 Cont'd (c) 12-14 years 6 5 Only employment 3 2 Employment and schooling 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Sierra Leone Cameroon (d) 15-17 years 6 5 3 Only employment Employment and schooling 2 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Sierra Leone Cameroon Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country; (b) Figures for,,, and Cameroon refer to the 7-14 years age range Source: UCW calculations based on national household surveys (see Table 1)

Bolivia Camerron Bangladesh 8 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 13 The population of children in employment can be divided into two broad groups, those combining employment and school and those only in employment The overall decline in children s employment presented above masks very different patterns for these two groups We can see from Figure 2that for the 7-17 years age group as a whole there has been a consistent and often substantial decline in the age of children only in employment across the 19 countries Children s employment that completely precludes schooling, in other words, appears to be becoming rarer over time 14 The trend for the group of children in the 7-17 years age group combining employment and schooling, however, is less consistent While the majority of the 19 countries have also seen a decline in the share of children who are in this group, there are a number of noteworthy exceptions to this trend In Bolivia, and, the overall decline in children s employment masks an actual rise in the share of children combining school and employment In populous Bangladesh and, the overall decline in children s involvement in employment is accounted for entirely by children only in employment; the share of all children who combine employment and schooling in the two countries remained virtually unchanged between the base and most recent years 15 A similar picture emerges when we look at the narrower 7-11 and 12-14 years age ranges The overall rise in employment among 7-11 year-olds in Bangladesh and the marginal overall rise in Bolivia were driven entirely by rises in the age of children combining employment and schooling; the age of children only in employment actually fell in both countries The overall decline in employment among 7-11 year-olds and 12-14 year-olds in, and among 12-14 year-olds in, masked a rise in the age of children combining employment and schooling Figure 3 Children who are excluded from schooling are diminishing as a share of total children in employment Children only in employment as a age of total children in employment, 7-11 years age group, base year and most recent year, by country (a) 8 7 6 5 3 2 1 Base year Most recent Latin America and Caribbean Sub Saharan Africa Asia Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1) 16 The net effect of these changes is that the remaining population of children in employment is increasingly made up of children who combine employment and schooling In the context of the overall decline in children s employment, in

Cameroon 9 UCW WORKING PAPER SERIES, MARCH 215 other words, there has also been a change in the interplay between employment and schooling a greater share of children in employment also go to school and a smaller share is excluded from schooling This point is illustrated in Figure 3, which reports children only in employment as a share of total children in employment for the 7-11 years age group The figure indicates that the share children in employment who are only in employment has fallen in all of the sample countries except Panel 2 Household chores and schooling Employment is not the only category of work involving children As reported in Figure A, an even larger proportion of children are engaged in a regular basis in performing household chores in their own homes in most of the Report countries where data on chores is available This is especially the case for female children, who are more likely than their male peers to be assigned responsibility for household chores across all of the countries Household chores include activities such as cleaning, cooking, washing clothes and dishes, child care and shopping They fall outside the technical definition of employment and therefore are not considered in the estimates in the main text of this report Figure A A large share of children in most countries also shoulder responsibility for household chores within their own homes Percentage of children performing household chores in their own homes, 7-17 years age range, most recent year, by sex and country 1 9 8 7 6 5 3 2 1 Male Female Latin America and the Asia Sub Saharan Africa Caribbean Source: UCW calculations based on national household surveys (see Table 1) The performance of chores frequently overlaps with school attendance in the sample countries In other words, a large share of students must undertake chores each day alongside their studies As reported in Figure B, the age of students also performing chores exceeds 5 in all but three of the 13 countries where data are available The age is especially high in the Sub Saharan African countries In and Cameroon, for example, over 9 of students also perform household chores on a regular basis As also reported in Figure B, however, the share of students also performing household chores appears to be declining in most of the Report countries This share has fallen in eight of the 13 countries, while in only three is there a clear trend in the opposite direction

Cameroon 12 5 97 68 62 95 11 165 325 376 369 353 595 Cameroon 172 423 484 534 636 732 756 722 757 926 917 857 846 1 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Panel 2Cont d Figure B The share of students who also perform household chores is substantial but appears to be declining over time Percentage of children attending school who also perform household chores in their own homes, 7-17 years age range, base and most recent year, by country 1 9 8 7 6 5 3 2 1 Base year Most recent Latin America and the Caribbean Asia Sub Saharan Africa Source: UCW calculations based on national household surveys (see Table 1) A smaller but nonetheless substantial share of students must perform both household chores and employment, ie, triple duty, with obvious consequences for the time and energy they have for their studies and for leisure Again, this appears to be an especially important issue in Sub Saharan Africa As reported in Figure C, at least one-third of all students are also involved in both household chores and employment in, Cameroon,, and The share of triple-duty students, however, appears to be declining in most countries, as also reported in Figure C Figure C A smaller but nonetheless substantial share of students must perform both household chores and employment Percentage of children attending school who are also in employment and performing household chores, Share of children carrying out triple duties (work, school and household chores) over total children in school, by year, 7-17 years age range, base and most recent year, by country 8 7 6 5 3 2 1 Base year Most recent Latin America and the Caribbean Asia Sub Saharan Africa Source: UCW calculations based on national household surveys (see Table 1)

Bolivia Camerron Bangladesh 46 68 1 126 129 113 12 134 13 hours 161 179 167 21 212 211 21 232 251 239 239 271 263 258 25 25 287 289 278 274 285 311 38 299 316 346 35 35 377 11 UCW WORKING PAPER SERIES, MARCH 215 4 TRENDS IN THE TIME INTENSITY OF CHILDREN S EMPLOYMENT 17 The decline in the incidence of children s employment has been accompanied by a reduction in the time intensity of employment in most countries In other words, not only is a smaller share of children working in the 19 countries, but those still working are spending less time each week on average actually doing so As reported in Figure 4, in many countries the fall in working hours has been dramatic in, for example, against a backdrop of a 1 age point fall in the proportion of children in employment for the 7-17 years age group, there was a 21 hours fall in the weekly working hours of the children remaining in employment There are on the other hand only a few exceptions where the time intensity of children s employment has increased Bangladesh (among 12-14 and 15-17 year-olds), (7-11 year-olds and 12-14 year-olds), (15-17 year-olds) and (7-11, 12-14 and 15-17 year-olds) (Appendix Figure A2 ) Figure 4 The decline in the incidence of children s employment has been accompanied by a reduction in the time intensity of employment in most countries Average weekly working hours in employment, base year and most recent year, 7-17 years age range, by country (a) 35 Base year Most recent 3 25 2 15 1 5 Latin America and Caribbean Sub Saharan Africa Asia Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1) 18 It is again interesting to look at differences between children only in employment and children combining employment and schooling underlying these general trends These differences are reported in Figure 5 A first point worth noting in this regard is that children only in employment log far more working hours than their counterparts attending school in all locations in both the base and most recent years Nonetheless, children combining employment and schooling log enough hours in most countries to significantly constrain the time and energy that they have for their studies (the impact of work on school performance is discussed in Section 6 of this report) 19 As also reported in Figure 5, most countries have seen a fall in the average weekly working hours of both children only in employment and children

12 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 combining employment and schooling There are, however, some notable exceptions to this general pattern In the countries where there has been a rise in the time intensity of children s employment, ie, in Bangladesh, and for the overall 7-17 years age group, this rise has been driven by sharp increases in the working hours of the group of children only in employment The working hours of children combining employment and schooling have either decreased (Bangladesh) or increased only marginally ( and )

hours hours 13 UCW WORKING PAPER SERIES, MARCH 215 Figure 5 Changes in the time intensity of children s differ between children combining employment and schooling and children only in employment Average weekly working hours in employment, base year and most recent year, children only in employment and children combining employment and schooling, by age range and country (a)(b) (a) 7-17 years age group 5 45 35 3 25 2 15 1 5 45 35 3 25 2 15 1 5 Only employment Employment and schooling Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Cameroon Sierra Leone (b) 7-11 years age group Only employment Employment and schooling 5 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Cameroon Sierra Leone

hours hours 14 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Figure 5 Cont d (c) 12-14 years age group Only employment Employment and schooling 5 45 35 3 25 2 15 1 5 45 35 3 25 2 15 1 5 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Cameroon Sierra Leone (d) 15-17 years age group Only employment Employment and schooling 5 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Cameroon Sierra Leone Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country (b) In, the calculation of working hours does not include those in subsistence agriculture Source: UCW calculations based on national household surveys (see Table 1)

15 UCW WORKING PAPER SERIES, MARCH 215 5 TRENDS IN THE STATUS OF CHILDREN S EMPLOYMENT 2 We have looked thus far at the prevalence and time intensity of children s employment We now turn to the composition of children s employment, how it is changing over time and differences in this regard between children only in employment and children combining employment and schooling 21 Figure 6 reports the status in employment of children only in employment and children combining employment and schooling for the most recent year The figure suggests important differences in terms of the work arrangements of the two groups: those combining employment and schooling are much more likely to be working for their own families while those only in employment are relatively more likely to be in paid work, and, to a lesser extent, in self-employment These patterns are especially pronounced in the countries of the LAC and Asian regions as well as in and 22 The differences in this regard between those only in employment and those combining employment and schooling appear to be growing over time As reported in Figure 7, in 13 of the 17 countries, the group only in employment has seen a reduction in the relative importance of unpaid family work and a corresponding increase in the relative importance of work outside the family and in particular in the relative importance of paid work It should be recalled that these changes in work arrangements occurred against the backdrop of a consistent cross-country decline in the prevalence of children only in employment As the share children only in employment has declined, in other words, this group has become increasingly concentrated in work outside the family 23 At the same time, the changes in the work arrangement of children combining employment and schooling have been smaller and less consistent across the 19 countries The relative importance of unpaid family work increased in nine countries but the pattern operated in the opposite direction in the seven other countries 24 Information on children s sector of employment, available for fewer of the Report countries, also points to differences between children only in employment and children combining employment and schooling (Appendix Figure A4) Those combining employment and schooling are more likely to be found in agriculture while those only in employment are more likely to be found in non-agricultural sectors ( and are exceptions) There are no clear patterns in terms of how the sectoral composition of employment is changing for over time for the two groups of children in employment (Appendix Figure A5)

16 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Figure 6 There are important differences between children only in employment and children combining employment and schooling in terms of work setting and work modalities Children s status in employment (% distribution), children in employment only and children combining employment and school, most recent year, by country (a) 1 Unpaid family worker Paid worker Self-employed Other status 9 8 7 6 5 3 2 1 ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO ES EO Bangladesh Bolivia Sierra Leone Cameroon Asia Latin America and Caribbean Sub Saharan Africa Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country The category "Unpaid family work" for estimates based on MICS surveys includes also "unpaid work carried out for non-family members" Source: UCW calculations based on national household surveys (see Table 1)

17 UCW WORKING PAPER SERIES, MARCH 215 Figure 7 The differences in work arrangements between children only in employment and children combining employment and schooling appear to be growing over time (a) Children s status in employment (% distribution), children in employment only, base year and most recent year, by country (a) 1 9 8 7 6 5 3 2 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bangladesh Bolivia Sierra Leone Cameroon Asia Latin America and Caribbean Sub Saharan Africa (b) Children s status in employment (% distribution), children combining school and employment, base year and most recent year, by country (a) 1 9 8 7 6 5 3 2 1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Y Y1 Bolivia Sierra Leone Cameroon Asia Latin America and Caribbean Sub Saharan Africa Notes: Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country The category "Unpaid family work" for estimates based on MICS surveys includes also "unpaid work carried out for non-family members" Source: UCW calculations based on national household surveys (see Table 1) Other status Self-employed Paid worker Unpaid family worker Other status Self-employed Paid worker Unpaid family worker

Bolivia Camerron Bangladesh 311 394 474 545 532 65 67 648 753 745 742 8 844 818 85 845 85 888 784 822 856 798 838 875 968 958 931 955 923 941 95 861 93 863 889 976 18 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 6 RELATION BETWEEN GRADE-FOR-AGE AND CHILDREN S WORK 25 Children are clearly disadvantaged in terms of being able to attend school As reported in Figure 8, the school attendance rate of working children is lower than of their non-working peers in all 19 countries Differences in this regard are often very pronounced in populous Bangladesh and, for example, non-working children are more than twice as likely to be in school as those who must also shoulder the burden of work Figure 8 School attendance is consistently much lower for children in employment than for their non-working peers Percentage of children attending school, 7-17 years age group, most recent year, by employment status and country (a) 1 9 8 7 Children in employment 6 Children not in employment 5 3 2 Latin America and Caribbean Sub Saharan Africa Asia Notes: Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1) 26 Nonetheless, as Figure 8 also shows, there are more working children in school than out of it in almost all of the 19 countries (more than half of working children are in school in all but, Bangladesh and ) This raises another important question relating to the educational impact of work: the extent to which work interferes with the ability of children to learn effectively once in the classroom It stands to reason that the exigencies of work limit the time and energy children have for their studies, in turn negatively impacting upon their academic performance But in the absence of test scores or some other direct measure of achievement, information that is unfortunately not available from the datasets we are using for this Report, it is difficult to draw concrete conclusions regarding the link between school performance and children s employment 27 Average grade-for-age, reported in Figure 9, offers one indirect measure of school performance The figure indicates the children combining employment and schooling lag substantially behind their non-working peers in terms of grade progression at the age of 13 years in all but one of our sample countries The lag in grade progression is likely in large part due to repetition arising from poorer

Bolivia Cameroon Bangladesh 24 grade 32 34 34 41 46 41 47 44 41 4 41 42 47 45 5 53 53 53 51 52 58 6 57 62 63 66 67 71 76 Bolivia Cameroon index 19 UCW WORKING PAPER SERIES, MARCH 215 performance, although other reasons of course cannot be ruled out 4 What is more, because children in employment are more likely to drop out at early ages, and because drop outs are presumably those with higher accumulated delay, the gap in Figure 9 is likely to underestimate the true gap in grade-for-age, ie, the gap that would be observed in the absence of selective drop out These results point to the difficulty that working children face in keeping up in the classroom with children that are not burdened with work responsibilities, and constitute another indication of the educational cost associated with children s employment Figure 9 Children in employment also lag behind in terms of grade for age (a) Average grade at age of 13 years of children combining employment and school expressed as a age of average grade for age at 13 years of children who are in school exclusively, (a)(b) 7-14 years age group, most recent year, by employment status and country (c) 18 16 14 12 16 1 8 6 94 91 93 94 89 95 77 93 71 87 8 95 87 94 4 2 Latin America and Caribbean Sub Saharan Africa Asia (b) Average grade at age of 13 years, most recent year, by employment status and country (b) 8 Schooling exclusively Combining employment and schooling 7 6 5 4 3 2 1 Latin America and Caribbean Sub Saharan Africa Asia Notes: (a) Distance from the value of one indicates the size of the difference in grade between children combining employment and school and children attending school only Values greater than one indicate that children combining employment and school have higher average grade at age of 13 years while values of less than one indicate that children combining employment and school have lower average grade at the age of 13 years (b) The upper and lower bars show the standard deviation from the ratio (c) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1) 4 The lag in grade progression might also, for example, be due to higher incidence of late entry among children who are identified as workers, or to higher absenteeism among child labourers in turn leading to grade repetition

Bolivia Camerron Bangladesh Bolivia Camerron Bangladesh Bolivia Camerron Bangladesh 2 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 STATISTICAL APPENDIX Figure A1 The general decline in children s employment has been accompanied by a general rise in school attendance Percentage of children in attending school, base year and most recent year, by age range and country (a) (a) 7-17 years age group 1 9 year year 1 8 7 6 5 (b) 7-11 years age group 1 9 8 LAC SSA Asia year year 1 7 6 5 LAC SSA Asia (c) 12-14 years age group 1 9 year year 1 8 7 6 5 LAC SSA Asia

Bolivia Camerron Bangladesh 21 UCW WORKING PAPER SERIES, MARCH 215 Figure A1 The general decline in children s employment has been accompanied by a general rise in school attendance Percentage of children in attending school, base year and most recent year, by age range and country (a) (d) 15-17 years age group 1 9 year year 1 8 7 6 5 LAC SSA Asia Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1)

Bolivia Camerron Bangladesh hours Bolivia Camerron Bangladesh hours Bolivia Camerron Bangladesh hours 22 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Figure A2 The decline in the incidence of children s employment has been accompanied by a reduction in the time intensity of employment in most countries Average weekly working hours in employment, base year and most recent year, by age range and country (a) (a) 7-11 years age group 45 35 3 25 2 15 1 5 year year 1 Latin America and Caribbean Sub Saharan Africa Asia (b) 12-14 years age group 45 35 3 25 2 15 1 5 year year 1 Latin America and Caribbean Sub Saharan Africa Asia (c) 15-17 years age group 45 35 3 25 2 15 1 5 year year 1 Latin America and Caribbean Sub Saharan Africa Asia Notes: Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country Source: UCW calculations based on national household surveys (see Table 1)

23 UCW WORKING PAPER SERIES, MARCH 215 Table A1 Changes in the age of children combining employment and schooling and of children only in employment, base year and most recent year, by country (a) sex, residence and age range Country Bolivia Cameroon Bangladesh Year Total Sex Residence Age range Male Female Urban Rural 7-11 12-14 15-17 E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) Y 252 54 198 271 49 222 234 59 175 126 3 96 443 91 353 161 1 151 36 55 251 374 142 232 Y1 243 38 25 265 34 232 22 42 178 135 21 113 428 66 362 167 6 161 258 29 229 359 13 256 Y 149 29 12 192 38 155 15 21 84 18 22 86 33 62 268 44 2 43 145 16 13 315 86 23 Y1 99 2 8 13 27 13 68 12 56 76 17 6 21 32 169 17 17 79 3 76 242 64 178 Y 165 6 15 225 88 137 11 3 71 - - - - - - 85 1 75 19 61 13 299 161 138 Y1 151 37 113 194 54 1 13 19 84 124 27 97 222 64 158 66 5 62 167 3 137 278 11 177 Y 235 94 141 288 114 174 181 73 18 133 48 85 392 164 228 121 8 113 28 11 17 5 242 163 Y1 66 3 36 87 42 45 44 18 26 37 18 19 115 5 65 17 1 16 57 19 38 153 88 66 Y 136 64 72 188 9 98 83 38 46 91 33 58 186 98 88 11 29 171 58 113 276 168 18 Y1 119 47 72 162 66 96 73 27 46 8 32 48 17 67 13 35 3 31 131 27 15 23 133 97 Y 28 153 127 362 193 169 194 11 84 212 99 112 317 182 135 12 3 9 345 162 184 536 396 1 Y1 23 95 18 274 124 151 13 66 64 161 65 96 236 118 118 67 1 58 248 87 161 3 259 144 Y 145 61 85 192 85 17 97 36 61 1 6 182 78 15 48 2 47 137 34 13 39 182 127 Y1 122 48 74 164 69 95 79 27 52 85 31 54 15 61 89 38 1 37 118 25 93 263 147 116 Y 37 179 191 384 171 213 357 185 172 183 75 16 383 186 197 273 146 127 535 235 3 - - - Y1 319 58 26 313 57 257 324 6 264 179 27 153 331 61 27 265 59 26 9 57 352 - - - Y 654 113 541 676 13 573 631 123 57 531 44 488 715 147 567 614 1 513 724 135 588 - - - Y1 62 93 527 636 78 558 64 18 496 478 31 448 733 143 59 595 84 512 672 112 559 - - - Y 415 89 3 441 85 372 389 94 39 92 32 65 456 97 375 274 31 251 522 79 464 623 245 7 Y1 35 6 247 317 59 26 293 61 234 122 29 96 335 65 272 186 2 167 375 48 329 498 168 335 Y 369 26 163 449 2 29 289 171 118 277 112 167 424 263 161 343 182 161 417 251 169 - - - Y1 174 92 83 29 116 92 141 67 73 12 62 22 125 96 15 72 78 219 128 91 - - - Y 7 398 343 752 386 368 727 411 318 681 218 463 764 473 294 716 392 326 79 412 38 - - - Y1 399 86 313 8 9 318 391 83 39 291 51 241 445 11 344 38 79 31 435 99 336 - - - Y 725 21 515 734 171 563 716 25 466 576 11 467 792 255 537 682 29 474 83 212 591 - - - Y1 41 59 351 417 49 368 3 7 333 258 21 237 479 77 3 377 47 33 473 82 391 - - - Y 434 67 368 449 68 382 418 66 353 23 59 144 471 68 4 34 19 285 524 48 476 63 28 423 Y1 416 58 359 427 64 365 6 53 353 182 52 132 453 59 394 276 11 266 54 44 461 583 171 411 Y 52 145 358 511 141 371 493 15 343 145 39 16 692 22 491 444 134 31 542 17 437 578 214 365 Y1 395 12 294 7 13 35 383 12 282 195 44 152 52 133 369 277 57 221 461 76 384 546 213 333 Y 211 142 69 29 193 98 119 83 36 178 128 5 22 146 75 45 33 12 377 19 188 373 333 41 Y1 175 16 69 267 169 98 71 35 36 151 11 5 181 17 74 65 18 47 19 92 98 475 4 71 Y 113 - - 118 - - 88 - - 71 - - 115 - - 11 - - 92 - - 31 - - Y1 66 - - 81 - - 49 - - 42 - - 74 - - 9 - - 49 - - 184 - - Y 113 85 27 132 99 33 92 71 21 73 56 16 141 16 35 15 5 1 65 37 29 225 188 38 Y1 9 62 28 14 74 31 74 5 24 62 46 16 19 73 36 16 6 11 52 24 28 168 13 38 Y 292 49 243 31 42 26 282 56 226 13 17 85 333 56 277 19 18 172 465 11 364 - - - Y1 13 21 19 125 19 17 135 23 112 72 1 62 15 24 125 77 4 73 212 46 166 - - - Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country; (b) E denotes overall involvement in employment regardless of schooling status (c) EO denotes employment only; and (d) ES denotes employment and schooling Source: UCW calculations based on national household surveys (see Table 1)

24 EVOLUTION OF THE RELATIONSHIP BETWEEN CHILD LABOUR AND EDUCATION SINCE 2 Table A2 Changes in average weekly working hours in employment, children only in employment and children combining employment and schooling, base year and most recent year, by country(a) sex, residence and age range Country Bolivia Cameroon Bangladesh Year Total Sex Residence Age range Male Female Urban Rural 7-11 12-14 15-17 E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) E (b) EO (c) ES (d) Y 287 46 239 274 457 233 32 462 247 329 514 271 269 433 226 226 416 213 268 441 23 357 475 285 Y1 232 462 19 224 455 19 242 467 189 28 552 229 26 411 169 163 381 155 229 464 2 29 468 218 Y 289 394 263 291 41 262 285 365 265 316 7 293 25 375 221 17 248 167 238 361 223 338 5 313 Y1 278 376 254 283 388 255 267 347 251 34 391 28 232 341 21 148 236 147 24 335 198 312 378 288 Y 251 413 157 259 41 161 231 421 148 - - - - - - 142 287 122 244 432 156 318 421 197 Y1 21 335 16 214 35 164 177 291 152 21 362 17 189 36 143 116 197 11 176 311 148 254 352 2 Y 311 411 244 317 426 245 31 387 244 33 442 265 31 397 232 26 363 195 33 377 256 377 431 297 Y1 274 369 195 289 384 22 244 333 182 311 396 229 254 353 178 12 253 114 227 329 175 318 379 237 Y 38 395 231 33 389 224 318 8 244 316 437 246 33 379 219 223 37 192 284 377 236 346 412 244 Y1 299 387 241 291 371 235 318 429 253 315 418 248 288 368 236 219 315 29 259 362 232 3 395 265 Y 377 45 29 38 459 289 372 434 293 3 513 38 368 432 282 262 326 241 374 46 3 432 465 3 Y1 316 437 21 32 454 29 38 3 21 331 472 236 38 422 193 164 276 145 271 421 19 391 453 279 Y 271 413 169 282 425 169 249 386 168 285 438 182 265 3 163 113 357 15 28 358 158 339 425 216 Y1 263 397 171 27 5 168 249 377 177 268 415 179 261 39 168 125 323 117 21 355 16 316 5 22 Y 179 211 149 185 216 16 174 27 137 154 198 117 18 212 15 167 189 142 189 235 154 - - - Y1 126 15 121 129 147 125 123 153 116 11 146 13 127 15 121 114 123 111 139 197 13 - - - Y 212 268 199 219 284 28 23 255 19 161 223 155 229 275 217 194 264 18 237 274 228 - - - Y1 129 211 114 1 244 125 117 185 11 82 152 77 15 22 134 116 185 15 152 251 132 - - - Y 134 21 115 1 223 121 127 198 17 23 342 141 132 25 114 12 149 96 129 195 119 173 235 137 Y1 113 147 15 119 165 19 17 13 11 161 313 114 111 135 15 91 99 89 11 135 17 136 165 121 Y 161 2 112 192 242 133 113 138 75 118 176 8 177 25 132 144 181 12 186 224 13 - - - Y1 211 297 117 239 326 132 17 247 99 248 438 112 21 267 119 185 266 111 244 328 127 - - - Y 239 255 233 251 265 246 228 249 216 195 231 187 254 26 251 216 22 214 277 318 262 - - - Y1 68 86 63 66 75 63 7 97 63 65 81 61 69 87 64 62 75 59 77 12 7 - - - Y 239 255 233 251 265 246 228 249 216 195 231 187 254 26 251 216 22 214 277 318 262 - - - Y1 1 166 88 95 142 88 15 184 88 123 316 14 95 151 84 92 168 81 11 164 98 - - - Y 13 259 17 139 292 113 12 225 11 221 452 146 124 237 15 99 211 92 125 247 113 172 274 125 Y1 167 384 117 166 366 117 169 4 117 331 578 146 156 35 116 14 281 97 143 343 113 237 6 144 Y 258 278 25 259 281 25 257 275 25 277 419 221 256 263 253 237 242 236 261 288 255 289 323 269 Y1 278 316 263 28 329 263 274 32 264 31 398 282 271 32 26 251 275 245 272 35 265 36 3 284 Y 285 335 183 37 368 188 223 248 167 332 385 195 275 323 181 254 27 28 241 314 167 357 368 269 Y1 346 469 153 359 473 158 292 445 1 399 511 157 334 457 153 213 421 135 28 412 157 461 52 183 Y - - - - - - - - - - - - - - - - - - - - - - - - Y1 - - - - - - - - - - - - - - - - - - - - - - - - Y 326 379 158 31 362 157 351 6 16 41 48 16 295 341 158 19 257 157 253 334 15 353 39 165 Y1 319 393 154 32 366 15 344 434 159 426 512 18 276 3 145 17 233 135 222 311 146 354 411 162 Y 21 295 193 217 323 2 22 273 185 197 398 156 211 288 196 174 226 169 235 316 213 - - - Y1 12 292 87 11 286 79 131 298 96 127 336 9 119 286 87 89 241 79 138 3 92 - - - Notes: (a) Specific reference years differ across countries; caution should therefore be exercised in drawing cross-country comparisons See Table 1 for details on reference years for each country; (b) E denotes overall involvement in employment regardless of schooling status (c) EO denotes employment only; and (d) ES denotes employment and schooling Source: UCW calculations based on national household surveys (see Table 1)