Child labour and schooling in the context of a subsistence rural economy: can they be compatible?

International Journal of Educational Development 23 (2003) 167–185 www.elsevier.com/locate/ijedudev

Child labour and schooling in the context of a subsistence rural economy: can they be compatible? A. Admassie PO Box 8437, Addis Ababa, Ethiopia

Abstract International concern over the problem of child labour often focuses on children working in the manufacturing and export sectors. Yet, most children in the developing countries provide unpaid agricultural and domestic labour, often at the expense of their education. A holistic approach that considers schooling in conjunction with children’s participation in child care, farm work, and other household responsibilities is, thus necessary to formulate policies that will encourage greater educational attainment. This paper has investigated the implications of children’s participation in household activities on schooling using a household survey from rural Ethiopia. The result suggests that combining work with school attendance is common among school children, although some activities may hamper school attendance more than others. Hence, a gradual policy towards child labour may be necessary in the context of subsistence economies, where initial interventions should aim at making the combination of work and school attendance possible rather than eliminating immediately child labour. To achieve this, introduction of a flexible school system that recognizes the peak demand seasons for family and agricultural labour may be necessary.  2003 Elsevier Science Ltd. All rights reserved. Keywords: Child labour; Education; Gender; Rural economy; Ethiopia

1. Introduction Child labour is a complex and growing problem in developing countries of Africa, Asia and Latin America. Although the Convention on the ‘Rights of the Child’ clearly underscores the need to protect children from economic exploitation and from performing any work that is likely to be hazardous, child employment still continues to prevail in many

developing countries.1 Recent, ILO estimates show that some 250 million (140 million boys and 110 million girls) children between the ages of 5–14 are obliged to work with 61% (152 million) being in Asia, 32% (80 million) in Africa, and 7% (18 million) being in Latin America (Kebebew, 1998). Besides, more than half of the school-aged children worldwide receive no primary education suggesting that they might be in the labour market. 1

E-mail address: [email protected] (A. Admassie).

Estimates show that about 80,000 new children are recruited globally into child labour every day (Grimsrud and Stokke, 1997). So, if one were able to stop the recruitment of child labourers one would rescue about 80,000 children every day.

0738-0593/03/$ - see front matter  2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0738-0593(02)00012-3

168

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

The above estimates of child labour have to be interpreted with caution since they exclude the whole categories of children working in the invisible domestic services, subsistence agriculture, and the informal sector in the form of unpaid work. If those invisible child workers were included, then the estimates of child labour would have been much higher than those estimates. In addition, such figures could also be deficient because of the likelihood of high under-reporting, as employment of children is generally prohibited in many countries by law. Moreover, the magnitude of the problem cannot be captured only with numbers alone since the conditions of child labour may vary between countries. Although Asia has the highest number of child labourers in absolute terms, Africa is on top of the list in terms of the child labour force participation rates. Of the relevant age group who participate in work activities, more than 40% of all children between the age of 5 and 14 are participating in economic activities in Africa, while the corresponding figures for Asia and Latin America are 21 and 17%, respectively. That means, the incidence of child labour in Africa is more than twice the level in Asia. Within the African continent, Eastern Africa ranks top in the rate of children’s participation in the labour force with 33%, while West Africa and Middle Africa account for 24 and 22% of the child labour, respectively (Kebebew, 1998). Specific country estimates also show that 27, 51 and 49% of all children between the age of 10 and 14 years participate in work activities in Benin, Burkina Faso, and Burundi, respectively (see Fig. 1). In Kenya, Ethiopia, Niger, Uganda, and Tanzania the participation rates are between 40 and 46%. In the case of Mali, 54% of the children are estimated to be working (ILO, 1997). Child labour has emerged as one of the most important issues on the global agenda and has become a subject of heated debates.2 However, 2 For instance, the recent international conferences on child labour in Amsterdam in February 1997 and in Oslo in October 1997 as well as the different regional consultation meetings in Africa, Asia and Latin America are clear manifestations of the growing conviction that this global problem needs a global solution. Child labour was also one the main agenda items of the ILO conference in June 1999.

Fig. 1. Participation rates of children, 10–14 years in economic activities (%). Source: Kebebew, 1998.

even today there are no simple and easy short-term solutions to the problem. Although, there may be several reasons why the issue of child labour should be a concern, at least three deserve serious considerations (Anker, 2000). Child labour is considered as a violation of children’s human rights since young children are forced to undertake abusive, harmful, and exploitative work activities, which may often be beyond their physical capabilities. Child labourers are deprived of their freedom of childhood, their right to recreate and play, and their natural development. Childhood should in principle, be a period of learning, recreation, physical, mental, and social development. The second concern about child labour arises from its effect on the educational development of the child. Child labourers are often deprived of any form of education. Hence, child labour is a dis-investment in human capital formation and has detrimental effects on the subsequent private and social returns from it. Debates about child labour also arise from its micro and macro economic impacts. It is often argued that eliminating child labour may entail untold sufferings to poor families3 since poverty is 3 For instance, Patrinos and Psacharopoulos (1997) argue that prohibiting children from all kinds of work without first raising family real income from other sources might have tragic

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

considered as the main, though not the only, reason for child labour in developing countries. The survival of poor families may depend heavily on the cash or in kind contribution of their children. With regard to the macroeconomic and labour market effects, there is a general belief that children displace adult labour and as a result depress adult wage rates thereby contributing to adult unemployment since child labour is often significantly undervalued in the labour market.4 Thus, understanding child labour is important not only because young children are made to undertake work obligations that may be beyond their physical capability but also because of its long-term effect on human capital formation and the children’s future welfare, and its micro and macroeconomic effect. The perpetuation of child labour is neither in the best interest of the children and their families; nor is it in the best interest of the social and economic wealth of nations (UNICEF, 1997). Most of the literature on schooling focuses mainly on children’s educational attainment and on school enrolment figures ignoring information on alternative household activities to education. Domestic and farm work, which could be serious impediments to school attendance have often been neglected in studies on schooling. The problem of child labour and its implication for children’s school attendance in rural economies, is also limited. Very little is known about the magnitude, nature, and distribution of child labour within the context of subsistence rural economies. Information on whether child work and school attendance are necessarily inherently incompatible or whether the two activities could sometimes be compatible is lacking. A holistic approach that considers schooling in conjunction with child work is essential to formulate policies that will encourage greater educational attainment for children in subsistence economies. In the case of Ethiopia, it is one of the

consequences for the poor creating even more destitute children, since these children contribute up to one-third of the family income. 4 Children’s earnings are consistently lower than those of adults, even where the two groups are engaged in the same tasks. Children receive no fringe benefits, insurance or social security payments (Bequelle and Boyden, 1988).

169

least researched areas that have been given little attention either in the media or in the academic circles despite its significance. In the light of the above discussion, the main aim of this paper is to examine and reflect on the link between child work and school attendance within the framework of a subsistence economy using a recent household survey from rural Ethiopia. The importance of such a study is warranted since it enables policy makers to identify areas of possible intervention to combat child labour and promote school attendance. The frequently proposed measures of trade sanctions will not be effective to bring child labour to an end in the context of a subsistence economic framework since the proportion of output exported is rather minimal. As Basu (1999) and Grote et al. (1998) stressed policy interventions to curb the problem of child labour should be context specific and need to be based on the types of child labour arrangements and the institutional setup. Identifying the extent of child labour and its impact on educational attainment could also serve as an effective instrument for informing the public and creating awareness. The study is also expected to stimulate debate on the subject and contribute to the empirical literature on the subject.

2. Reasons for expecting a high incidence of child labour in Ethiopia While no one actually knows how many children are at work at present, there is little doubt that child labour could be exploitative and harmful in Ethiopia due to the country’s unique economic, social, and developmental circumstances. The factors that accentuate child labour including rapid population growth, economic crisis, insufficient public investment in essential economic and social services such as in education and health, frequent wars, the breakdown of family structures and an increase in female headed households, and the HIV/AIDS epidemic are more prevalent in Ethiopia than anywhere else. Ethiopians have probably one of the lowest rates of life expectancy and standard of living in the world. The problem will continue to persist as economic conditions deteriorate and expen-

170

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

diture in social development, especially in the area of health and education decline. A large proportion of the Ethiopian children find themselves as victims of famine, disease, poverty, civil war, family displacement, and social instability all of which are contributing factors for a high incidence of child labour. Hundreds of under aged children are engaged in work activities at home, on the farm, or on the street because they are deprived of any form of education that is necessary for their proper development. The AIDS epidemic is potentially a big threat since under aged orphaned children may be increasingly pushed into work activities to support themselves. Recent estimates show that the number of HIV/AIDS orphans in Ethiopia has reached the one million mark, which is the highest in the world save for India and South Africa (Walta, 2001). These orphans face a host of social problems, and many of them are being forced to live on the street. The number of street children is also growing everyday (Addis Tribune, 1999). The ILO estimates also show that Ethiopia has one of the highest child labour participation rates in the world (Fig. 2).5 Ethiopia is the second most populous country in sub-Saharan Africa with an estimated population of over 65 million. More than 50% of the total population is reported to be below the age of 14 years (CSA, 1994). The fertility rate is among the highest in Africa and the world with about seven children per women. Due to the rapid increase in population, a growing proportion of the population, including children, are entering into the labour force at a faster rate and at an early age. Agriculture is the mainstay of the economy accounting for more than 45% of the GDP, 90% of the foreign exchange earnings, and employing more than 85% of the population. The sector is also characterized by subsistence production and low technological

5

The participation rate of children between the age of 10 and 14 in economic activities in the world in 1995 was around 13% while the corresponding figure for Ethiopia was more than 42%. And although there is a declining trend in the incidence of child labour globally by the year 2010 still more than 40% of the children between the age of 10 and 14 years could be participating in economic activities.

Fig. 2. Trends in the participation rates of children in selected countries, 10–14 years. Source: ILO, Economically active population, 1997.

development both of which lead to high incidences of child labour. The literature on child labour often states that poverty is the root cause of child labour (see for instance, UNICEF, 1986; Basu and Pham Van, 1998; Blunch and Verner, 2000; Cockburn, 2000; Ray, 2000). Ethiopia is one of the least developed and poorest countries on earth with a per capita income of less than US$ 110. More than two-thirds of the Ethiopian population live on income below that required to meet the minimum basic needs (UNICEF, 1992). A recent survey in Ethiopia showed that more than three-quarters of the working children work for more than 9 hours a day and give all the money they earn to their families (Walta, 1999) underscoring children’s significant contribution to family income. Thus, the rapid population growth coupled with the predisposing poverty put Ethiopia as a suitable candidate for a high incidence of child labour. Schooling is probably the most effective instrument for combating child labour. Ethiopia has one of the lowest school enrolment rates in the world primarily due to the poor and underdeveloped education system and due to lack of a compulsory education system.6 For instance, the gross enrolment 6 Generally public schools do not charge official school (tuition) fees, but unofficial expenditures on supplies, regis-

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

rate in primary schools was only 35% in 1996/1997, which means that more than 60% of the primary school-aged children herd livestock and completely miss any form of education, despite the universal declaration that children should have the right to free primary education. When these children grow older and finish herding they face difficulty in finding work because they have few other skills other than herding and they fall back in an environment where there is already high unemployment. Moreover, about a third of those students enrolled in grade one drop out and some 20% repeat the grade (MOE, 1994).7 The drop out rates for junior secondary and secondary schools stand at 10 and 17 %, respectively.8 Evidences from other developing countries show that there is a lot of gender disparity in school enrolment. For instance, Murard (1998) stated that 25 % fewer girls than boys are attending schools in sub-Saharan Africa. Moreover, while school enrolment is much lower for girls, school dropout rates are much higher for girls than boys (Bequelle and Boyden, 1988). In the case of Ethiopia, while the primary gross enrolment rates for boys was over 40% the corresponding rate for girls was less than 30% in 1996/1997 (see Table 1). The enrolment of girls also declines as the level of education increases. The school dropout rates are also much higher for girls in Ethiopia. Parents in Ethiopia would prefer their boys to stay in schools than the girls and feel that school attendance is more essential for boys than for girls (ILO, 1995). Parents might think that a child could learn a skill (for example agriculture) through work and supplement family income. As the majority of the

tration fees, and payments for school uniform are frequent and could at times be exorbitant. The availability of private schools, which normally charge tuition fees in rural areas, is limited. 7 For instance, in 1993/1994 28 % of the girls and 24 % of the boys dropped out before completing the first grade of school (Colclough et al., 2000). More recently it was reported that about 44,000 primary school students dropped out of school in just one zone (province) alone during the 2000/2001 academic year (Walta, 2001). 8 The present education system in Ethiopia is organized into primary level, consisting of grades 1–6, junior secondary referring to grades 7 and 8 and secondary levels consisting of grades 9–12.

171

Ethiopian population is illiterate, parents may be unable to visualize the long-term benefits of education and invest in their children’s schooling. The inability of the education system to accept every child and the indirect costs associated with schooling might also encourage households to send their children to work and try their luck in the informal sector. When a child is not attending school it is not only the welfare of the child that is endangered, but also the welfare of the nation as a whole. Full time working children could not acquire the necessary human capital, which allows them to have a fair chance in the labour market when they grow up. This translates at the national level into a labour force, which is ill equipped to compete in the global economy where development will be increasingly based on skills and competition (Anker and Melkus, 1996). The level of technological development and the structure of the economy could also be important factors contributing to the increased supply of child labour. Studies show that the higher the share of agriculture in the GDP, the greater will be the incidence of child labour. If the production system is backward and labour intensive it requires greater participation of children. Since Ethiopia is an agrarian country dominated by subsistence agriculture, there are adequate reasons to expect a high incidence of child labour. Actually, current estimates show that more than 57% of the total child employment in Ethiopia is found in rural areas (ILO, 1995). Child labour is more prevalent in the rural areas also because the capacity to enforce minimum age requirements for schooling and work is lacking. The level of enforcement is very weak since the law covers only undertakings where a formal employment relationship exists, leaving out the agricultural and the informal sectors. On the basis of the prevailing social and economic circumstances, it is natural and logical to expect a high incidence of child labour in Ethiopia. There are, therefore, adequate reasons to be concerned about the problem of child labour in Ethiopia. Although the complete abolition of child labour will take a long time, there is a need to ensure that a start is made since as stated earlier child labour jeopardizes children’s physical and intellectual development and hinders their possibilities of

172

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Table 1 Gross enrolment ratios by education level and region (1996/1997) Regional states

Tigray Afar Amhara Oromiya Somali Benshangul Gumuz SNNP Gambella Harari Dire Dawa Addis Ababa Ethiopia

Grades 1–8

Grades 9–12

Male

Female

Total

Male

Female

Total

49 10 30 41 16 67

41 7 26 20 7 29

45 8 28 31 12 49

7 2 7 8 1 6

4 1 5 5 0 3

5 2 6 6 1 5

60 82 74 54 80 43

28 49 57 47 81 26

44 66 66 51 80 35

10 16 40 22 47 10

5 4 32 18 38 7

8 10 36 20 42 8

Source: Ministry of Education, Education statistics: Annual Abstracts, May 1998.

becoming productive adults in the future. Therefore, the issue of child labour deserves to be a matter of priority concern for Ethiopia.

3. Child labour and schooling in the rural economy of Ethiopia: empirical evidence Children are usually employed in activities that are technically simple. Hence, child labour is overwhelmingly a rural phenomenon, since as many as 70% of the child labourers are involved in agricultural production worldwide (ILO, 1999).9 Subsistence oriented rural communities often feel that it is more important to involve children in economic activities and equip them with the basic life skills for their future survival. This practice has pushed several rural children into exploitative and hazardous work conditions, thereby limiting their ability and right to grow in a healthy and conducive environment.

9

The significant labour contribution of children even at an early age in agriculturally based societies has always been recognized. For instance, an ILO study in Indonesia indicated that 75–85% of the agricultural labour force consists of children of which more than 20% could be below the age of 10 years.

Children in rural Ethiopia begin participating in work activities as shepherds looking after the animals, collecting firewood and fodder, fetching water, and then taking on more strenuous tasks such as farming and cooking as they grow up. Agricultural work includes activities like ploughing, weeding, harvesting, transporting, threshing, and post harvest and marketing activities. Climatic exposure or working under extreme weather conditions, work that is too long and too heavy for young children, work that exposes children to snake bites and cuts from some sharpened tools, working barefooted for long hours, and lifting heavy loads are some of the health hazards children may face in rural areas. Work in the rural informal sector like shop keeping, traditional weaving, making of agricultural tools, trading are also part of children’s activities. Some children are also hired out as labourers to other households particularly if the family does not have adequate farmland. Domestic service, which is one of the world’s oldest occupations and one in which children have traditionally played a part, is also one of the most common forms of child labour in rural Ethiopia. Domestic chores include activities like fetching of water and firewood, minding of small children, preparation of food, cleaning, feeding and milking

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

domestic animals, family business work as well as the sale of grain and livestock in local markets or combinations of these activities. Some children carry significant domestic workloads in their own homes while many more work in other households in exchange for room, board, care, and sometimes remuneration. Some children also work as shop helpers or hotel workers. Although children participating in domestic work in their own households may not be subjected to the harsh treatments and abuse many are still deprived of formal education. Those domestic child workers who are employed outside their homes typically live and routinely work for long hours without any access to education. Girls are particularly affected by domestic work because it is considered to be an essential component of the girl’s upbringing exercise. Many believe that girls should be prepared for their future roles as mothers and wives.10 Some argue that some of domestic activities could be considered as normal family obligations and thus do not necessarily give rise to exploitation and abuse. Whether work has a detrimental effect on children’s development depends on the working conditions (hours of work, exposure to physical hazards, etc.), age of the child, and other risks and abuses (psychological and social adjustment risks, prostitution, child slavery, etc). But, it is clear that sending out children to work is not primarily designed to serve the child’s interest, but is the outcome of a transaction in which the traded commodity is the child labour. So, domestic service whether in ones own home or outside the household, can cause serious psychological and social adjustment problems. To examine the implications of work on school attendance within the context of a rural economic framework, information from the rural household surveys that were collected by the Department of Economics, Addis Ababa University in collaboration with the Centre for the Study of African Economies (CSAE-Oxford) was utilized. Detailed 10

For instance, more than 35% of all the girls between the age of 10 and 14 in rural Brazil work as domestic workers compared to a national average of only 20%. In Ecuador the proportion of working girls in the rural areas could go as high as 44% (ILO, 1999).

173

socio economic data were collected in three rounds between 1994 and 1995 from a total of 15 peasant associations (Kebeles) spread over four major regions (see Fig. 3 for location of survey sites). A total of 1477 households, which were selected randomly and proportional to size from each survey area, were interviewed using structured questionnaires. The four regions constitute almost 90% of the total population and occupy more than threequarters of the country. The 15 survey sites have attempted to capture the different agro-ecological zones (Dercon and Krishnan, 1996).11 Some of the survey sites belong to the poorest regions of the country where frequent famine is observed while some are better off. While cereal production dominates the northern and central regions chat,12 coffee, and enset13 crops prevail most in the South. Some salient features of the survey sites are presented in Appendix A, Table A1. 3.1. Main types of work activities performed by rural children In order to identify the implications of work on the normal development of the child it is essential to have a stock of the type of activities performed by rural children. Respondents were asked to identify the primary, secondary, and tertiary occupations for every member of the household including children. Table 2 presents the primary responsibilities of rural children expressed as a proportion of the surveyed children of the relevant age group. Rural children except the inactive category, which consists of the very young and the disabled ones, participate in domestic and farm work activities usually contingent on the gender of the child. While most children have been participating in 11 The three major traditional agro-ecological zones in Ethiopia are the Woyena Dega, which refers to the mid altitude where the temperature is temperate, the Kola, which is the equivalent to an arid and semi arid zone, and the Dega, which is the high altitude zone where the temperature is rather cool. 12 Chat is a kind of stimulant crop whose leaves are mostly consumed in the horn of Africa and in the Arabian Peninsula and widely grown in Eastern Ethiopia. 13 Enset is a root crop that is mostly used as food in most parts of southern Ethiopia.

174

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Fig. 3.

Map of Ethiopia and location of the study areas.

Table 2 Primary responsibilities for rural children by age group and sex (%) Type of work Pre-school childrena

Student Domestic and farm work Too young to work Others Total Number of children

Aged 7–10b

Aged 11–14b

Total

Male

Female

Male

Female

Male

Female

Male

Female

1.8 10.0

1.2 9.7

18 54.8

11.8 59.5

30.4 62.3

18.1 75.5

15.6 40.8

9.6 45.6

88.1

89.1

26.6

28.1

5.3

4.2

42.8

44

0.2 100.1 570

0.0 100 586

0.7 100.1 579

0.7 100.1 595

2.1 100.1 438

2.1 99.9 425

0.9 100.1 1587

0.8 100 1606

Source: Ethiopian Rural Household Study, 1994. a The pre-school children group consists of mainly the inactive children. Primary schooling begins at the age of 7 years in Ethiopia, although some children below 7 years may be enrolled in traditional or religious schools, where they learn the basic literacy and numeracy. b The others category includes those children looking for job, the disabled, and those not in the labour force as well as those engaged in other informal activities.

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

a multiple of work activities some children have been attending schools during the time of the survey. Some were also able to combine work with school attendance. As expected, domestic work and farm work are the two primary work responsibilities for most of the rural children after controlling for the inactive group. According to the information presented the number of children who have been working14 either on the farm or in the household was about three times the number of children going to school. School attendance has been the main task for not more than 10% of the surveyed children. Even preschool children, who are normally too young to work, participated in domestic and farm work activities. The participation rate of children in domestic and farm activities usually increases with age. For instance, while the participation rates for children between the age of 7 and 10 years were 55 and 60% for boys and girls, respectively, the rates increased to 62 and 76%, respectively for the next higher age category. An important issue when dealing with child labour is the need to recognize gender disparities in work and school participations. Experiences from other countries show that girls’ work participation rates are much higher than boys’ participation rates. In the process of preparing the girls to become good wives in the future, female children are required to learn cooking and other skills from their mothers or from their older sisters through increased participation in household activities. This situation leads to greater school non-attendance and dropouts for girls. More girls also stay out of schools than boys due to early marriage practices. In the case of rural Ethiopia, female children were participating more in household activities such as looking after younger siblings, preparing and cooking food, cleaning the house, and fetching water and firewood. Boys’ participation was higher in farm work such as ploughing and harvesting of

14

This category of working children includes all those children whose primary activity was indicated to be farm work, domestic work such as helping in the family, work in the informal sector, etc. Only children whose main responsibility was schooling, those who were labelled as too young to work, and those who were disabled and unable to work have been excluded from the working children category.

175

crops, and looking after the livestock. While girls’ school participation rate was lower than boys by about 10%, their participation in the labour force was greater than boys by the same margin. Only a little more than 10% of the girls in the 7–10 age group and about 20% of the girls in the 11–14 age category reported that their primary responsibility was school attendance. The corresponding figures for boys were 20 and 30% for the 7–10 and 11– 14 age categories, respectively. This evidence indicates that there is a clear gender bias in work and school attendance in rural Ethiopia. Moreover, the bias increases, as children get older more girls than boys work while more boys go to school. 3.2. Average work starting age The issue of child labour should be a concern to policy makers not only because children do participate in economic activities but also because of the early age at which they start to participate in work-related activities. In general, the poorer the household is, the lower would be the age at which children start to participate in work activities. Fig. 4 illustrates the age at which rural children start to participate in household work activities as a percentage of rural children who begin work related activities at the indicated age. Children in rural Ethiopia start assuming household responsibilities even before celebrating their fifth birthday. Some 15% of the boys and about 20% of the girls started participating in work activities before the age of five. Early age work participation is more prevalent among female children, the average work starting age being about 6 years (6.37 and 6.27 years for male and female children, respectively). By the age of 7 years, which is normally the school starting age, about 80% of the rural children have started to participate in work activities. Almost everyone starts to participate in household and farm work by the age of 10 years. The very early age at which children start to participate in labour activities is an important aspect of child labour since the younger children start to work the more vulnerable they become to workplace hazards, exploitation and ill treatment. The physical and mental development of very young children could be seriously affected through long

176

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Fig. 4. Age at which rural children start participating in household work activities (%).

hours of work, climatic exposure, work that is too heavy for young bodies, and accidents. These children are also as usual either staying out of school or even if they are going to school they will not be motivated to study well. So, children’s participation in work related activities at an early age could bring undesirable consequences on the normal development of the child. 3.3. Work participation and schooling The relationship between school attendance and child labour is perceived to be negative since education and child labour are believed to be mutually incompatible. According to this argument, if a child is not going to school then he/she is most likely engaged in some kind of work activity, which may have long-term impact on the child’s intellectual development. Thus, uninterrupted

school attendance reduces the time the child has for work at home or in the labour market. Whether school attendance and participation in work could be compatible or not largely depends on the type of work activity and the duration of work, which we will examine later in this section. The highest grade completed by rural children including participation in traditional and religious schools and in literacy programs is reported in Table 2, which gives the percentage of children who have completed the stated educational level. Although children in Ethiopia are expected to start attending school at the age of seven more than 70% of the boys and more than 85% of the girls had not started school attendance (formal or informal) by that age during the time of the survey. Hence, as many as 80% of the rural children in Ethiopia will remain illiterate and may not be able to take advantage of human capital formation in the future. The gender bias in school attendance is also clearly reflected by the higher proportion of girls who are out of school for all age categories. More male children are enrolled in formal schools as well as in traditional or religious schools than female children (Table 3). In order to have a much broader picture of the implications of work on education it might also be important to examine past school attendance, a proxy for school dropout rates. School dropout rate is very high in rural Ethiopia as observed by the proportion of children who have attended school in the past and who were not in schools at the time of the survey.15 About 4% of the children between the age of 7 and 10 years and some 10% of those between the age of 11 and 14 years have discontinued their education most of whom are working (see Table 4). Even those who were enrolled in schools at the time of the survey were likely to dropout because of their delayed enrolment age. More than 50% of the children aged between 7 and 10 years who were participating in domestic and farm work and more than 60% of children between the ages of 11 and 14 years engaged in similar 15 A recent report indicated that more than 10% of the female students drop out of school per year due to responsibilities at home, early marriage, fear of abduction, and several other reasons (Walta, 2001).

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

177

Table 3 Highest level of school attained by rural children by age and sex (%) Highest level Pre-school children of schooling attained Male None (no 93.1 schooling) Some nursery 0.2 school Primary 4.0 education Junior 0.0 secondary school Adult literacy 0.4 program certificate Traditional or 2.4 religious school attendance Other literacy 0.0 programs Total number 549 of children

Aged 7–10

Aged 11–14

Total

Female

Male

Female

Male

Female

Male

Female

96.9

71.9

85.1

42.2

66.5

72.9

82.6

0.2

2.4

1.8

1.6

1.2

1.4

1.1

1.8

16.9

13.9

35.2

24.6

17.5

12.4

0.0

0.0

0.0

2.7

1.7

0.8

0.5

0.0

0.9

0.4

1.6

2.5

0.9

0.8

1.1

7.5

3.3

8.8

2.7

6.1

2.4

0.0

0.5

0.4

1.6

0.7

0.6

0.3

1579

1518

545

587

570

443

403

Source: The Ethiopian Rural Household Survey, 1994.

Table 4 Rural children’s past and present school attendance status (%) Activity

Aged 7–10 Attending

Student at 14.1 present Domestic/farm0.9 work Too young 0.3 to work 0.01 Othersb Sub Total 15.3 Total count 1175

Aged 11–14 Nevera

Attending

0.0

0.0

23.7

3.4

52.8

0.3 0.3 3.9

Attended

Total Nevera

Attending

Attended

0.0

0.0

12.1

0.0

0.0

0.8

7.8

60.2

0.5

3.5

39.2

26.7

0.0

0.2

4.5

0.2

0.6

42.4

1.3 80.8

0.0 24.5 864

1.0 9.0

1.7 66.4

0.03 12.9 3193

0.5 4.5

1.0 82.6

Attended

Source: Ethiopian Rural Household Survey, 1994. a Never refers to children who have never been to schools. b Others include children looking for work, disabled children, children participating in informal activities, etc.

Nevera

178

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

activities have never been enrolled in schools. The percentage of girls who have never been enrolled in schools is also much higher than the percentage of boys that never visited schools. It may not be sufficient to know that a large proportion of the rural children do not go to school. From a policy perspective it might also be useful to identify the main reasons why children are not attending school since children might be staying away from schools also for non-work related reasons. Child employment interferes with schooling by absorbing too much time of the child. Work also requires a lot of energy so that children cannot have the necessary energy for school attendance or for an effective study. Even for those children who might be able to combine farm work or domestic work with schooling, long hours of work will leave them exhausted. Faced with fatigue, these children will have very little mental stimulation the results of which will be neglecting of their studies. The primary reasons for not attending schools by age and gender are presented in Table 5. Work related reasons are among the most important factors hindering school enrolment in rural Ethiopia. Many children fail to enroll in schools simply because their families cannot afford to do without their labour either for domestic or farm work. On the average work related reasons have been identified as the primary reasons for non-school attendance in about 23% of the cases and become increasingly important as the child’s age increases. For instance, more than half of the surveyed children between the age of 11 and 14 years identified work related reasons as the primary causes for school non-attendance. These results suggest that work participation may be in direct conflict with school attendance. Reasons linked to the direct costs of schooling as proxied by the distance to school and expenses on school supplies also contribute significantly to the low school enrolment. Some 15% of the surveyed children indicated that either schooling is too expensive or far away from their homes prohibiting them from attending school. But compared to work related reasons direct costs of schooling have relatively lower effect on school attendance. Whether a child should be going to school or should be engaged in work obligations is primarily

a family decision. Therefore, parental attitude towards investment on human capital is an important determinant of school enrolment. The decision to send a child to school is partly influenced by the labour requirement of the family, social norms like marriage practices, and on the expected financial return from schooling. If families felt that the present or expected return from schooling (relative to the perceived cost of so doing) is rather low they would prefer to send the child to work instead of sending them to school. The education offered may sometimes be too academic having very little relation with their daily needs and thus even when they send their children to school they might withdraw them after a few years. In particular parents attitude towards girls education is a critical factor determining the level of girls school enrolment. Family decisions also explain part of the high school dropout rates. Parental attitude towards the education of their children in general and their opinion about the relative importance of schooling for boys and girls is reported in Table 6. Most of the respondents believe in the importance of schooling for both male and female children (about 70% believe that education is equally important for both boys and girls). About 99% believe that boys should be sent to school while 93% felt that schooling is an important investment for girls as well. In other words, while less than 1% of the respondents indicated that it is not useful to send boys to school some 7% of the respondents believed that schooling is not appropriate for girls. However, there are significant response variations on the relative importance of schooling for boys and girls. More than 20% of the respondents think that boys’ education should be given greater priority than girls’ education. Parents’ perception about the benefits of schooling is linked to expectations of the child’s role when the child grows up. Households perceive that the return to investment in male children’s education is greater than female children’s in Ethiopia because the son remains at home after marriage whereas the daughter is expected to move away so that the investment in her education is lost to her family. Boys in most cases inherit the family assets and continue to support their parents in old age

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

179

Table 5 Primary reasons for not attending school by age group and sex (%) Reason

Too young to attend school Required for farm work Required for domestic work Child takes care of the sick and/or elderly Required to work for wages Total work related reasons School too expensive School too far /no places Total reasons related to direct costs of schooling Other reasons Total number of valid cases

Pre-school children

Aged 7–10

Aged 11–14

Both sex

Male

Female

Male

Female

Male

Female

89.5

91.4

46.1

45.0

10.8

9.3

57.7

2.9

1.9

22.3

7.8

42.2

12.2

11.6

0.6

1.5

7.8

18.9

9.0

37.6

10.8

0.0

0.0

1.2

0.2

0.0

1.1

0.4

0.0

0.0

0.5

0.2

0.4

0.7

0.2

3.5

3.4

31.8

27.1

51.6

51.6

23.0

3.2

2.1

10.9

12.4

22.4

19.7

9.8

1.8

1.7

4.3

6.7

5.8

8.2

4.3

5.0

3.8

15.2

19.1

28.2

27.9

14.1

3.0 507

1.4 526

6.9 421

8.7 460

9.4 223

11.2 279

6.0 2406

Source: Ethiopian Rural Household Survey, 1994.

whereas female children become a part of their husbands’ families after marriage. As a USAID (1994) report underscored, investing in girls education might be perceived as not being beneficial to her own family, because any benefits of her education will go to her future husband’s family, rather than her own. Examples from Togo also showed that parents prefer to send girls into domestic services because the girl’s income helps to support the schooling of her brothers (Black, 1997). Although it has been indicated earlier that domestic work and farm work are the primary responsibilities for rural children, many children participate in more than one activity. Many children combine school attendance with work. Without looking into the implication of work on the scholastic achievements of children it can be argued that some work

activities may be more compatible with schooling than others. If any work activity is incompatible with schooling then intervention is surely needed. Hence, the type of household work activities performed by school going (in-school) and non-school going children have been identified and analysed. Table 7 presents the different types of work activities and the proportion of school going children who are at the same time participating in these work activities. Combining school attendance with activities like herding and the collection of firewood and/or water is very common in rural Ethiopia. The empirical evidence shows that around 50% of the schoolgoing boys and some 35% of the in-school girls have been participating in firewood and/or water fetching exercise. Children who attend school are

180

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Table 6 Parents’ attitude towards the (relative) importance of education (%) Attitude of parents about education

Male

(a) Opinion about the usefulness of sending boys to school Useful to send boys to school 98.6 Not useful to send boys to 0.8 school Do not know 0.6 Total respondents 1438 (b) Opinion on the usefulness of sending girls to school Useful to send girls to school 93.3 Not useful to send girls to 5.8 school Don’t know 0.8 Total respondents 1437 (c) Opinion on the relative importance of education for children Education is more important 6.7 for girls than boys Education is more important 19.3 for boys than for girls. Education is equally important 73.8 for both boys and girls Others 0.2 Total respondents 1438

Female

Both sexes

98.8 0.5

98.7 0.6

0.8 1066

0.7 2504

92.7 6.7

93.0 6.2

0.7 1064

0.8 2501

7.5

7.0

23.1

20.9

69.2

71.7

0.2 1061

0.2 2499

Source: The Ethiopian Rural Household Survey, 1994.

able to fetch water or fire wood either on their way to and from school or before and after school hours. Similarly, about 30% of the in-school girls and 20% of the school-going boys have been participating in livestock herding. Collection of firewood and water absorbs a significant portion of rural children’s time and becomes a particularly important responsibility for children during the summer vacation in which schools will be closed. Public schools operate in two or three shifts in most places allowing school-going children to participate in work activities. The data show that school-going children also make significant contribution to farm and domestic work. On the other hand, the participation of school-going children in child minding activities has been relatively low suggesting that taking care of younger children is one of the most incompatible household activities with school attendance. When these young children are forced to take care for even younger siblings, they lose educational opportunity at school. In addition the younger child may be prevented from

developing sufficient verbal and conceptual skills to benefit from formal education. On the basis of the presented evidence one can generally conclude that farm work, herding, and fetching of water and/or firewood seem to be relatively more adapted to schooling than other forms of household activities like child minding. Although some work activities may be more compatible with school attendance than others, this does not necessarily mean that children’s work participation will not have negative implications on their scholastic achievement. If school-going children are made to work long hours then they will be exhausted from the work and will have less time to concentrate on their studies. So, identifying the primary household responsibilities of children might not indicate the full scale of the problem. In fact, the issue may not be whether school-going children also participate in work activities or not, but rather whether work participation has negative implications on children’s school performance and health. Although children’s participation in work

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

181

Table 7 Primary work responsibilities for school-going children by age and sex (%)a Primary work responsibility

Fetching firewood/water Farm work Family business work Domestic work such as cooking and cleaning Cattle herding Minding small children None i.e. no work responsibilityb

Aged 7–10

Aged 11–14

Total

Male

Female

Male

Female

Male

Female

50.0

36.4

55.6

36.0

47.4

34.7

14.7 2.9

18.2 0.0

5.6 0.0

16.0 0.0

12.8 2.6

17.3 1.3

7.1

9.1

16.7

24.0

11.5

14.7

23.5 0.0

36.4 0.0

11.1 0.0

24.0 0.0

20.5 1.3

29.3 1.3

1.9

0.0

11.1

0.0

3.8

1.3

Source: Ethiopian Rural Household Survey, 1994. a School attendance is defined in terms of whether the child was attending school during the last four months prior to the survey. b None indicates the percentage of school going children that have not been engaged in any work responsibility.

activities for up to 7 h a week may not be terribly harmful, long hours of child employment will have an adverse effect not only on the child’s physical development but also on his/her mental and psychological development.16 Long hours of work participation have detrimental effects on the private and social return from education (Psacharopoulos, 1997). If children who are attending school have to work excessively long hours after schools and over the weekends then there is no leisure or playing time available for them. Long working hours are often responsible for fatigue that can cause accidents and impair intellectual development. Thus, the number of hours children are made to work each day could be an additional indicator of the scale of the prob-

16

According to international standards, an individual who is engaged in an economic activity for at least 1 h per week is considered as employed (Assaad et al., 2001).

lem. Table 8 presents some information on the duration of work activities for school-going children in the survey areas. Table 8 shows that school going (in-school) children are often required to work from 2–4 h every school day and up to 13 h over weekends and on holidays. The daily working hours for these children are definitely in excess of any reasonable threshold. In particular cattle herding occupies more than one-third of their whole available time. Child shepherds who are contracted out to other families are subjugated to long hours of work and are denied any form of education (Hachallu, 1998). Girls work longer hours than boys on the average since they are mostly engaged in domestic services, which demand longer hours of work. So, the risk of poor intellectual development as a result of the excessive working hours for children in rural Ethiopia is very high. Poor education or poor training could seriously hinder the welfare of these children and the society at large since opportunities for future employment increasingly depends on literacy and possession of other skills and technical progress.

182

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Table 8 Duration of work related activities for school going children (%) Duration of work Aged 7–10 Male Average number 0.86 (1.47) of weeks taken out of school to help with farm work Average number 3.28 (2.04) of hours worked after/before school per school day Average number 10.42 (7.14) of hours worked per weekend

Aged–14 Female

Male

Total Female

Male

Female

0.36 (0.93)

0.66 (1.15)

0.55 (1.06)

0.74 (1.29)

0.46 (0.99)

3.40 (1.95)

3.05 (1.83)

3.72 (1.81)

3.13 (1.93)

3.57 (1.89)

10.83 (6.21)

9.07 (6.70)

12.60 (8.20)

9.45 (6.88)

11.81 (7.29)

Figures in parentheses are standard deviations.Source: The Ethiopian Rural Household Survey, 1994.

Examining the type of work activities of schoolgoing children alone may not provide a complete picture about the compatibility of child work and schooling. For a more complete and better understanding of the effect of child work on school attendance we need to examine the type of activities performed by non-school going children as well. If there is a significant difference between the types of activities performed by school-going and non-school going children then that may be an indication of the incompatibility of the work with schooling. The different types of work activities performed by non-school going (out-of-school) children are reported in Table 9. For all categories of non-school gong children fetching of firewood and/or water, herding, and taking care of younger children have been the main work responsibilities. Compared to school-going children, out of school children have relatively greater participation in childcare activities, but lower participation rate in activities such as fetching of firewood and water, farm work, and domestic work. Minding of younger children was the main responsibility for about 15% of the nonschool going male children and for about 10% of the out-of-school female children, while the corresponding figure for the in-school children was less

than 2%. There was no significant difference between the participation rates of school-going and non-school going children in family based business activities.

4. Conclusions and policy implications There is ample evidence suggesting that the problem of child labour poses a major challenge in rural Ethiopia. Supply and demand related factors as well as the social attitude prevailing in the society of looking at child labour as normal phenomenon even if it is detrimental to the normal development of the child are the main contributing factors in Ethiopia. Neither the vulnerability of children to work related damages nor the undesirable consequence of work on children’s education and health have been given due attention and recognition by policy makers in Ethiopia. The empirical data presented in this study indicate that children in rural Ethiopia are expected to perform a multiple of tasks, which may often be beyond their physical capabilities even at the expense of their education. Nearly one out of every two children participate in domestic and farm work activities, while only one out of every fifth child goes to

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

183

Table 9 Work participation rates for non-school going (out-of-school) children (%) Primary responsibility

Aged 7–10

Male Fetching firewood/water Farm work Family business work Domestic work such as cooking and cleaning Cattle herding Minding small children None

Aged 11–14

Total

Female

Male

29.5

31.4

23.5

25.2

25.0

30.2

7.6 3.0

11.4 2.1

9.2 3.1

9.2 3.4

8.5 3.1

11.6 2.1

15.2

12.1

9.2

11.8

12.2

11.6

22.7 12.9

27.9 6.4

29.6 17.3

28.6 13.4

28.4 14.2

27.3 9.3

6.1

8.4

9.1

8.6

Female

Male

8.5

Female

8.0

Source: The Ethiopian Rural Household Survey, 1994.

school, the figure for girls being slightly less than that for the boys. The results also indicate that school-going children are expected to undertake domestic and farm work activities on top of their school attendance. Some household and farm work activities may be relatively more compatible with school attendance than others, even though, they may retard the scholastic achievement children. The immediate elimination of child labour is neither feasible nor desirable in view of the significant children’s labour contribution to family labour and household production system within the framework of a subsistence rural economy. But, a start has to be made to reduce the undesirable consequences of children’s work participation on schooling. The frequently suggested simplistic solutions like banning child labour or implementing international trade sanctions will not have appreciable impact either on promoting school attendance or reducing the incidence of child labour in subsistence oriented economies. The problem of child labour can only be properly addressed through a combination of short- and long-term approaches and no single intervention may be sufficient in itself. In the short term, education is probably the most important and powerful tool to combat the problem of child labour. In the context of the Ethiopian

social and economic setup, the provision of relevant, universal, and compulsory primary education of good quality should be the centrepiece of any short- and long-term strategy to combat child labour. The result of this study suggests that a gradual policy towards child labour may be necessary. Initially interventions should aim at making the combination of work and school attendance possible rather than eliminating immediately child labour. Combining work with school attendance is a fact and that non-harmful child labour is common among school children in rural Ethiopia and often may be necessary for family survival. This situation demands that school calendars should be synchronized with peak demands seasons for family labour, such as agricultural seasons. In other words, the school system needs to be flexible enough to recognize and adapt according to the constraints of rural communities in the short term. Flexibility of school hours, hours of work, and vacations periods, which coincide with harvest time and peak agricultural season may potentially be effective measures to facilitate the work school combination. In the long run, the problem of child labour can only be minimized through a comprehensive and

184

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

integrated approach, encompassing legal, economic and social measures including advocacy, public awareness raising, and community mobilization and empowerment measures since it cuts across economic, social, political, cultural and legal aspects. Measures to reduce poverty and bring sustainable development, appropriate population policy to slow down the rate of population growth, reliable statistical data, and appropriate legislation that does not exempt from coverage work activities like agriculture, family undertakings, small workshops, and domestic services are all important.

Appendix A

References Addis Tribune, 1999. Child Labour is Increasing in Ethiopia, October 1, Addis Ababa, Ethiopia. Assaad, R., Levison, D., Zibani, N., 2001. The Effect of Child Work on School Enrolment in Egypt, Mimeograph. Anker, R., H. Melkus, 1996. Economic Incentives for Children and Families to Eliminate or Reduce Child Labour, Mimeograph, ILO, Geneva. Anker, R., 2000. The economics of child labour: a framework for measurement. International Labour Review 139, 257– 280. Basu, K., Pham Van, H., 1998. The economics of child labour. The American Economic Review 88, 450–477. Basu, K., 1999. Child labour: cause, consequence and cure, with remarks on international labour standards. Journal of Economic Literature 37, 1083–1119.

Table A1 Some salient features of the survey areas Survey site

Location

Number of households interviewed

Number of children/ Main crops household

Harresaw

Tigray

84

2.5

Geblen Adele Keke

Tigray Oromiya

66 97

2.8 3.6

S. Godeti T. Kecheme

Oromiya Oromiya

98 102

3.6 4.0

Korodegaga

Oromiya

109

3.9

Yetmen Shumsheha

Amhara Amhara

61 149

2.4 2.3

Dinki Debre Birhan Adado

Amhara Amhara SNNPS

87 184 132

2.1 3.0 3.3

Aze debo

SNNPS

75

4.3

Imdibir

SNNPS

67

3.4

Domaa Gara Goda

SNNPS SNNPS

74 96

3.6 4.1

Total

1477

Barely, wheat, lentils Barely, maize, Teffa Millet, maize, chat, coffee Teff Wheat, barely, Teff, potato Maize, Teff, barely, beans Teff, wheat, beans Teff, lentils, peas, sorghum Millet, teff Teff, barley beans Coffee, Enset, maize, Enset, coffee, maize, Teff Enset, chat, maize, coffee Enset, maize Enset, maize, barely, teff

3.3

Source: Ethiopian Rural Household Survey and Bevan and Bereket, 1996. a Teff is a very small domestic cereal and staple crop mainly in the central highlands of Ethiopia.

Agro-ecology

Woyena dega Dega Woyena dega Woyena dega Woyena dega Kola Woyena dega Kola Kola Dega, W.dega Woyena dega Woyena dega Woyena dega Kola Woyena dega

A. Admassie / International Journal of Educational Development 23 (2003) 167–185

Bequelle, A., Boyden, J. (Eds.), 1988. Combating Child Labour, ILO, Geneva. Bevan, P., Kebede, B., 1996. Measuring Wealth and Poverty in Rural Ethiopia: A Data Based Discussion. In: Kebede, B., Taddesse, M. (Eds.), The Ethiopian Economy: Poverty and Poverty Alleviation. The Ethiopian Economic Association, Addis Ababa, Ethiopia. Black, M., 1997. Child Domestic Workers: A Handbook for Research and Action, Anti-slavery International. Blunch, N.-H., Verner, D., 2000. Revisiting the Link between Poverty and Child Labour: The Ghanaian Experience, Policy Research Working Paper No. 2488, The World Bank, Washington DC. Central Statistical Authority, 1994. Population and Housing Census, Addis Ababa, Ethiopia. Cockburn, J., 2000. Child labour versus education: poverty constraints or income opportunities? Paper presented at the Conference on Opportunities in Africa: Micro Evidences on Firms and Households. CSAE Oxford, April Colclough, C., Pauline, R., Mercy, T., 2000. Gender inequalities in primary schooling: the role of poverty and adverse cultural practice. International Journal of Educational Development 20, 5–27. Department of Economics, Addis Ababa University, 1994. Ethiopian Rural Household Survey. Unpublished data. Dercon, S., Krishnan, P., 1996. A consumption based measure of poverty for rural Ethiopia in 1989 and 1994. In: Kebede, B., Taddesse, M. (Eds.), The Ethiopian Economy: Poverty and Poverty Alleviation, Addis Ababa, Ethiopia. Grimsrud, B., Stokke, L.J., 1997. Child Labour in Africa: Poverty or Institutional Failures?: The Case of Egypt and Zimbabwe, Fafo Report No. 233. Grote, U., Basu, A.K., Weinhold, D., 1998. Child labour and the international policy debate: the education—child labour trade—off and the consequences of trade sanctions. Working Paper Series 1. Centre for Development Research (ZEF), Bonn. Hachallu, B., 1998. A survey report on the problem of children in shashemene town. Paper presented on the Workshop on the Problem of Child Labour, 15–16 January 1998. Shashemene, Ethiopia.

185

International Labour Office, 1995. Report of the National Workshop on the Problem of Child Labour in Ethiopia, ILO, Nazareth. International Labour Office, 1997. Child Labour in Commercial Agriculture in Africa, ILO, Geneva. International Labour Office, 1999. Targeting the Intolerable: A New International Convention to Eliminate the Worst Forms of Child Labour, Information Kit, ILO, Geneva. Kebebew, A., 1998. Statistics on Child Labour: A brief Report. Bulletin of Labour Statistics 3, ILO, Geneva. Ministry of Education-MOE, 1994. Education Sector Strategy, Addis Ababa, Ethiopia. Ministry of Education. 1998. Education Statistics: Annual Abstracts. Education management Information Systems. Addis Ababa, Ethiopia. Murard, F., 1998. Schooling for Girls Lags behind. The Courier (Number 167, January–February), pp. 59–60. Patrinos, H., Psacharopoulos, G., 1997. Family size, schooling and child labour in Peru: an empirical analysis. Journal of Population Economics 10 (4), 387–405. Psacharopoulos, G., 1997. Child labour versus educational attainment: some evidences from latin America. Journal of Population Economics 10 (4), 377–386. Ray, R., 2000. Analysis of child labour in Peru and Pakistan: a comparative study. Journal of Population Economics 13, 3–19. UNICEF, 1986. Exploitation of Working Children and Street Children. UNICEF Executive Board, 1986 Session Report E/ICEF/1986/CRP.3, New York. UNICEF, 1992. A Situation Analysis of Children and Women in Ethiopia, Addis Ababa, unpublished material. UNICEF 1997. The State of the World’s Children 1997. Oxford University Press, New York. USAID, 1994. The Demand for Primary Schooling in Rural Ethiopia, A Research Study, Mimeograph, Addis Ababa. Walta Information Center, 1999. Children Deployed to Physical Labour: Manifest Negative Behaviour, 11 October, Addis Ababa, Ethiopia. Walta Information Center, 2001. English News, 21 August, 2001, Addis Ababa, Ethiopia.