Child labour in the global economy
This report proposes to examine the key effects, elements and issues surrounding child labour in relation to the global economy. Through their thorough and convincing research, the authors, Eric .V.Edmonds and Nina Pavcnik (2005) estimate the number of economically active children between the ages of 5-14 years and claim that parental poverty is its main cause. However, they are hesitant to endorse the widely held belief that international policies are the solution to this global crisis; rather they suggest more domestic measures to limit it, such as affordable education and more job stimulation. They also argue that contrary to popular perception most working children engage in domestic/market work rather than in manufacturing establishments.
Introduction
The term Child Labour is a widespread phenomenon often equated to child abuse. In Eric’s footnote on ‘What is Child Labour?’ He states the International Labour Organisation’s (ILO) definition of child labour which is”… defines an economically active child as a child labourer if she is under 12 and economically active for one or more hours per week, 12-14 and working more than 14hours per week or one or more hours per week in activities that are ‘hazardous by nature or circumstance’ and if she is 15-17 and works in ‘unconditional work forms of child labour’ (trafficked children, children in bondage or forced labour, armed conflict, prostitution, pornography, illicit activities)”(ILO 2002, cited in Edmonds and Pavcnik 2005; pp.200).
Edmonds asserts however that these forms of child labour are rare. He insists stating that children engage in activities which could either be harmful or beneficial depending on the type of activity and also the impact on the child should depend on what the child would be doing if not working (the child’s alternative foregone). Household Survey evidence by the ILO’s Statistical information and monitoring program on child labour (SIMPOC) also defines child labour as “A child is defined as economically active if he or she works for wages (cash or in-kind); works in the family farm in the production and processing of primary products; works in family enterprises that are making primary products for the market, barter or own consumption; or is unemployed and looking for these types of work (ILO 2002,cited in Edmonds and Pavcnik 2005; pp.201). This is a more practical definition taking into account that most children work at home for their families; either on the farm or domestically.
The rare hazardous forms of labour stated in the ILO’s definition are difficult to make out in household surveys which the SIMPOC are based on thus specialized surveys are conducted by the ILO and other concerned organizations. As a result of these surveys, the ILO’s SIMPOC (2002) estimates 8.4 million children engage in ‘unconditional work forms of child labour’ and of these, 68 percent are in bonded or forced labour (forms of slavery). According to Edmond and Pavncik most child labourers are employed by their parents to work on the farm or domestically which goes against the widely held view that children work mainly in manufacturing establishments and other forms of employment. They argue that children also face risks in the simplest forms of labour as they get older, for example agriculture, due to exposure to toxic chemicals, harsh weather conditions, animals/parasites. Ashagrie (1997) agrees with this point stating that “the self reported injury rate from child labour surveys of children working in agriculture is actually higher at 2 percent than the 9 percent level reported in manufacturing. (Ashagrie 1997 cited in Edmonds and Pavcnik 2005; pp.208)
Edmonds and Pavncik’s article posits an open research question as to whether the reasons children engage in these hazardous working conditions differs from the driving force of children working on their family businesses or domestic work.
Literature review.
Child labour in the global economy is estimated at 211million, ILO (2002), which accounts for 18 % of children, aged 5-14 worldwide. Of these, 60 % are in Asia and 23 % in sub Saharan Africa. Although Asia has a greater percentage of child labours Africa has a higher participation rate estimated at 30%. SIMPOC also estimates that 4 % of children work in transition economies (i.e. countries undergoing economic liberalization) and 2% in developed countries.
The United Nations children’s educational fund (UNICEF) conducted three surveys in thirty-six less developed countries in 2000 and 2001 providing information on the participation rates in both domestic and market work for 124 million children. The first survey helps buttress Edmond and Pavcnik’s point on child workers engaging mainly in domestic/ market work rather than manufacturing establishments.
Participation rates in various activities for 124million children 5-14 from 36 countries in 2000.
All children
5-14
Age
5-9 10-14
Gender
Male Female
Location
Urban Rural
Market work (MAR)
25.0
15.3 35.2
26.6 23.3
18.9 30.5
Paid
2.4
1.0 4.0
2.8 2.0
2.2 2.5
unpaid
5.8
4.4 7.3
5.6 5.9
4.0 7.3
family
20.8
12.4 29.7
22.4 19.1
14.8 26.2
Domestic work (DOM)
64.6
50.8 79.2
59.3 69.9
60.7 67.4
Any work (MAR + DOM)
68.4
53.5 84.3
64.8 72.1
64.1 71.7
20 or more hours per week
20.7
10.3 31.8
19.4 22.1
14.1 26.4
40 or more hours per week
6.4
2.7 10.3
6.1 6.7
3.6 8.8
UNICEF End of Decade Assessment microdata, (2000). See Edmonds and Pavcnik (2005, JEP).
From the table above it is evident that less than 3% of children work outside the household for pay and this is mainly the case in rural settlements rather than urban where manufacturing activities are dominant. Approximately 6% participate in unpaid work. Edmonds and Pavcnik interpret these unpaid children as those children who help their neighbours in the farm or business or could also be children receiving in-kind payments (food) as well as children who are indebted to their employees by their parents. However in contrast to this, 20.8% of children work in family businesses and of this figure, 26.2% are in rural areas and 14.8% in urban. The authors further research shows that there is an agreement with this UNICEF survey from other available data from countries like Nepal, South Africa, Vietnam, India as well as Bangladesh where child labour in its garment industry also had a figure of 1.2% unpaid children age 5-14 as cited in a 2002 child labour survey. According to the table approximately 65% of children work domestically and 68% work in both market and domestic work. There is a high participation rate pattern by older female children age 10-14 in rural areas. The table also shows that they tend to work longer hours than males. This is probably due to cultural domestic values. Agriculture takes the largest part of the employment sector in countries (e.g. Kenya 77% In 1998, Guatemala 63 %in 2000, Ethiopia 89% in 2001 etc) followed closely by domestic work and then manufacturing which is only a percentage of economically active children. Figures from the Food and Agricultural Organisation (FAO) 2004 (FOA 2004 cited in Edmonds and Pavcnik 2005)match with this indicating that most adults work in agriculture and since most children work with their parents they also work in agriculture.
There is a trade off between work and school for child labourers. Most working children attend school and the UNICEF estimates in its second survey the total hours of work in relation to different types of work as well as school attendance. The data shows that on average, children spend 26 hours on market work weekly. Children that work for their families and are unpaid spend 27hours weekly. Paid employment takes up more hours in a week and by older children. Children that take part in domestic work spend 16 hours per week .Edmonds and Pavcnik further stated that these figures should not imply that domestic work is insignificant because on average, majority of the hours spent on market work is indeed domestic work. However children that attend school spend less hours working than those who do not attend school.
The third survey below by the UNICEF reports the school attendance of children aged 5-14. From the table we observe that approximately 70% of children in that age range attend school and this attendance is mainly in favour of older male boys, in urban areas. School attendance conditional on work status accounts for 74%. There is a 14% likelihood that children who do not work do not attend school but this is reflected mainly in younger children. Less than 5 % of the 30% of children that do not attend school work in market work only and we see that domestic work is more popular in this
case as 32% of the 30% participate in it alone. Thus Edmonds states that in ignoring this figure of domestic work researchers would lose out on one of the segments of children not attending school.
A large fraction of children, 42%, well known as ‘idle’ attend neither school nor work
Work and schooling status for 124million children 5-14 from 36 countries in 2000
All children
5-14
Age
5-9 10-14
Gender
Male female
Location
Urban rural
Attend school
69.5
58.9 80.8
70.768.3
75.1 63.9
Attendance rates conditional
Any work
Not work
73.9
60.0
64.1 80.6
52.9 82.2
75.7 72.3
61.6 57.8
80.1 68.3
64.9 52.8
Conditional on nonattendance
Domestic only
Market only
32.0
4.5
30.8 34.9
2.8 8.3
27.1 36.6
6.3 2.7
31.8 32.0
4.9 4.3
Both market and domestic
Not work
22.0
41.5
13.1 42.2
53.3 14.6
20.3 23.5
46.2 37.1
12.8 26.6
50.6 37.1
UNICEF Multiple Indicator Cluster Survey End of Decade Assessment microdata.(2002)
Child labourers face consequences on schooling attainment because time spent working gives less time for extracurricular activities and might weaken their school performance. Orazem and Gunnarsson (2004) find that “third and fourth graders who attend school but never work in market or domestic work perform 28 % better on mathematics tests and 19% better on language tests than children who attend school and work”. Empirical evidence of this is given in an example by Beegle, Dehejia and Gatti (2004) wherein after five years of working and schooling they examined the status of young adults in Vietnam. They observed that a one standard deviation increase in the hours worked by children attending school is equated to a 35% decrease in educational attainment. Edmonds and Pavcnik however are of the school of thought that this negative correlation between working and grade advancement might reflect that low performing students engage in work rather than that work generates low-performing students.
Parents send their children to work not by choice but as a result of economic necessity thus Child labour is caused fundamentally by poverty and as far back as data show, there have been several policies proposed with the aim of fighting against child labour worldwide. Have these legislative acts helped solve the problem of child labour? To answer this question Edmonds and Pavncik give two examples, firstly the US Congress enacted laws prohibiting the importation of goods made by children and as a result of these sanctions, Bangladesh was said to have released over 10,000 paid child workers below the age of 14 from its garment industry in the mid 1900’s. Evidently this is not drastic as the figures are a tenth of the percentage of child workers in Bangladesh. Furthermore Moehling (1999) finds little evidence that minimum wage laws in manufacturing sectors put into practice between the years 1880 and 1910 contributed to the decline in child labour during this period. In addition, the authors state that more recent studies show that the US House of Representatives have discussed the “Child labour elimination act” which would enforce punitive measures, alter financial support, and order US oppositions from multilateral development banks to 62 developing countries affected by child labour.
Edmonds and Pavcnik argue that although these policies might help reduce child labour, they are not guaranteed to be successful in less developed nations where they are targeted at for these reasons enumerated in their article:
Lack of available resources to implement policies (i.e. bans) mainly in cases where the children engage in market work for their parents wherein the end product contributes to the family income.
Child labour policies are seen as a case of multiple equilibrium, Basu and Van (1998), whereby children that work at low wages are at one equilibrium and increased adult wages when children do not work are at another therefore defeating the purpose of the ban. Moreover according to Basu’s example (2003), if firms are fined for child labour, the cost of the fine increases the demand for cheap child labour as opposed to high earning adult workers in order to make more profit to offset the fine which in turn makes child labour necessary.
Children that are prohibited from working legitimately might move into more hazardous forms of child labour or non exporting sectors of the economy such as prostitution however there is no scientific evidence of this.
Since Poverty is the main cause of child labor it would be logical to combat child labor by
A) Improving the standard of living of families i.e. increasing adult income which in turn does the following
Diminishes the marginal utility of income which decreases the value of marginal contribution because as more income comes into the family form parents it eliminates the satisfaction derived from the income contribution of the child.
Brings about the purchase of alternatives used in place of child labour (e.g. washing machines)
Increases a child’s productive level in schooling (human capital) because the family is now able to afford necessary materials (e.g. textbooks)
B) Encouraging credit markets to give loans to poor households is another suggestive method because child labor has been said to be a result of credit market imperfections. Baland and Robinson 2000; Ranjan, 2001 agree with this stating “Several theoretical studies emphasize that if credit markets allowed households to borrow against future earnings child labor could be much reduced”
C) Providing affordable education because parents would be willing to contribute to improve their children’s long run chances yet, are however left with no choice but to send their children to work due to the perceived returns to schooling. Policies have been adopted to reduce chid labour through educational subsidies. An example of such is the Progresa program in Mexico which is predominately used in many countries worldwide. It gives parents an incentive to keep their child in school. Parents are paid if the school certifies that the child has been attending school regularly. Thus it is a demand approach aimed at reducing the cost of schooling, which varies with age of the child, and increasing family income.
Conclusively relying on laws and their enforcement as discussed previously, is a necessary but unsatisfactory solution to child labour. Overall, economic development interventions relevant to the underlying cause are more effective policy tools to reduce the incidence of child labour.
