Wa0002.
Wa0002.
International Trade,
Economics & Environment
Research Report
Child Labour in
South Asia
Are Trade Sanctions the Answer?
#0311
Child Labour in South Asia
Are Trade Sanctions the Answer?
Child Labour in South Asia
Are Trade Sanctions the Answer?
This paper was researched and written by Dr. Jayati Srivastava, Fellow, Centre for Contemporary
Studies, Nehru Memorial Museum & Library, New Delhi. Comments on the draft were received from
Mr. Mahesh C. Verma, Adviser, Minister of Labour & Chairman, ESIC Review Committee, Govt. of
India, Mr. Sajid Kazmi, Programme Manager, Children's Resource International, Pakistan and many
others, which have been suitably incorporated.
Published by:
Edited by:
Deepmala V. Mahla and Pranav Kumar
Layout by:
Mukesh Tyagi
Cover Photo:
Courtesy – V. V. Giri National Labour Institute
Supported by:
The Ford Foundation, New York, USA
Under the Capacity Building on Linkages between Trade and Non-Trade Issues Project.
ISBN 81-87222-82-4
© CUTS, 2003
Any reproduction in full or part must indicate the title of the paper, name of the publishers as
the copyright owner, and a copy of such publication may please be sent to the publisher.
Preface .................................................................................................................................................................. i
I. Introduction ................................................................................................................................................. 8
V. Recommendations ...................................................................................................................................... 54
Endnotes ............................................................................................................................................................ 56
References .......................................................................................................................................................... 60
List of Tables and Chart
Table 4: Percentage of 10-14 age group children in the labour force in South Asia ........................................ 25
Table 5: Economically active children and child labour in total workforce in countries of South Asia ............. 26
Table 7: Child population, child labour, school education and nowhere children in South Asia 1995 ............... 27
Table 10: Economically active children (5-14 years) in Nepal 1995/96 ................................................................ 33
Table 12: Economically active children (5-17 years) in Sri Lanka 1999 .............................................................. 37
Table 13: Some labelling initiative aimed at child labour products in South Asia ............................................... 49
Demanding the inclusion of social issues in the WTO implies opening the
window for endless non-trade issues including animal welfare, gender, human
rights and the like. This would contaminate the trade agenda and at the same
time overload the WTO.
This sort of linkage has immense potential for abuse as a protectionist device
of the North. The southern countries argue that, it would help only a few rich
countries, not global welfare.
This paper includes a survey of the various economic theories on child labour.
In general terms, child labour originates in household poverty, poor adult
wages, low wages paid to child labour, lack of education etc. At the theoretical
level, scholars have come up with different models on causes of child labour.
A dominant section of these theories consider poverty and household decision
making regarding their children the fundamental variable determining the
incidences of child labour.
South Asian countries, which record one of the highest rates of child labour
practices in the world, the poverty explanation of child labour situations is
predominant. Fertility, education, labour market, adult wages, capital market
and availability of credit have been underlined as important causes.
These sanctions were exercised to alleviate the problem of child labour by the
US policy-makers and also by some countries in the EU. The import of those
goods was banned in the production process of which child labour was used,
wholly or partly. The sectors affected in South Asia were the garment industry
in Bangladesh, the carpet industry in Nepal, the football industry in Pakistan
and the carpet and bidi industries in India.
The experience as yet, asserts that the policies targeted at alleviating poverty,
illiteracy, liquidity constraints and distortions in the labour and capital markets
are more likely to retrench the virtual ubiquity of child labour.
It has been suggested repeatedly that to wean away children from work, they
should be provided free education. However, there is the feasibility aspect of
such a measure in terms of the costs of displacing the child labour. A study
was conducted by CUTS on “Eradicating Child Labour while Saving the
Child: Who Will Pay the Costs?” in 1999. As per this study, the estimated cost
of displacing child labour in India is $18.94bn.
Such argument does not have any economic rationale. According to Chatterjee
and Mehta (1998), there is no conclusive proof that the so-called poor labour
standards in the developing countries have positive correlation with export
There is no conclusive proof that the
intensities of these countries. Moreover, there is no evidence to suggest that
so-called poor labour standards in the
developing countries have positive imports from developing countries cause unemployment in developed world.2
correlation with export intensities of In the context of globalisation, two reasons have brought the issue of labour
these countries. standards to the fore, apart from the moral argument that the labour rights
should be protected. According to Chishti (2000), a growth in developing
countries’ exports of manufactured goods from 30 percent of their total
exports in 1980s to 70 percent in 1995 caused serious concerns in developed
countries, despite them being the protagonist of free trade. Moreover, the
tendency amongst MNCs and TNCs to shift labour intensive production units
to developing countries due to the availability of cheap labour, threatened the
employment opportunities in developed countries. As a result the developed
countries “called for universally agreed upon labour standards to be applied
to international trade” and also create provisions within the WTO to impose
unilateral trade restrictions on countries who fail to adhere to core labour
standards.3 Madden (1995) also points out that the renewed interest in linkages
between trade and labour standards is due to liberalisation, globalisation,
unemployment and communication. The manufacturers in the developed
world perceive cheap labour in developing countries as a potential advantage
accruing to developing country producers.4 As a result, new trade barriers are
being erected against the developing countries.
It is noteworthy that the US, which is seen to espouse the cause of child
labour as well as other core labour standards through the trade route, has
not ratified the Minimum Age Convention No. 138 of the ILO. Moreover,
amongst the eight core ILO Conventions, it has ratified only two (C182 Worst
Forms of Child Labour Convention and C105 Abolition of Forced Labour
Convention). Its commitment to the principles enshrined therein is thus
suspect in the same way as its attempt to append a social clause to the WTO.
Child Labour
The idea of using trade measures to alleviate the problem of child labour dates
back to early twentieth century when in 1917 the American Labour Federation
(AFL) demanded that the Peace Treaty to conclude World War I should
In the Singapore Ministerial in 1996, the US along with France and Norway
proposed the creation of a working group on labour standards within the WTO
to complement the work of the ILO. In face of stiff reservations and
apprehensions about protectionism expressed by developing countries, the
ILO and not the WTO was deemed an appropriate forum to deal with the issue
of labour standards. The Singapore Ministerial Declaration likewise spelt out:
“we renew our commitment to the observance of the internationally recognised
core labour standards. The International Labour Organisation (ILO) is the
competent body to set and deal with these standards, and we affirm our
support for its work in promoting them. We believe that economic growth and
development fostered by increased trade and further liberalisation contribute
to the promotion of these standards. We reject the use of labour standards
for protectionist purposes, and agree that the comparative advantage of
countries, particularly low-wage developing countries, must in no way be put
into question. In this regard, we note that the WTO and the ILO Secretariats
will continue their existing collaboration.”11
At Singapore Ministerial in 1996, due
to shift resistance and apprehension Subsequently, the matter came up again at the Seattle Ministerial in 1999
about protectionism expressed by where the US and EU were successful in putting it up for discussion in a
developing countries, it was decided working group. The EU favoured a joint ILO/WTO Standing Working Forum
that the ILO and not the WTO was
deemed an appropriate forum to deal
on the issue while Canada proposed a WTO Working Group on the relationships
with the issue of labour standards. between appropriate trade, developmental, social and environmental policy
choices in the context of adjusting to globalisation. Although these proposals
did not envisage the use of trade as an instrument of enforcing labour
standards, the US President Bill Clinton, on his way to the Ministerial did
mention that in future trade sanctions could be used to penalise violators of
labour standards. This evoked strong reactions from developing countries’
grouping and ultimately the issue of labour standards proved most divisive and
contributed (among other issues) to the collapse of the Ministerial.
Thus, within the WTO framework, the status quo is being maintained on this
issue. Meanwhile, within the ILO framework, some renewed initiatives are
being taken to alleviate the problem of child labour. These new ILO initiatives,
among other things (campaigns on ratification, awareness drives, etc.) include
New ILO initiatives, among other
the establishment of the Working Party on the Social Dimension of Globalisation
things (campaigns on ratification,
awareness drives, etc.) include the to address the issue of trade liberalisation, employment and investment with
establishment of the Working Party on a special focus on poverty reduction and a proposal to form a global commission
the Social Dimension of Globalisation of eminent persons to examine the social aspects of globalisation. In 1999, the
to address the issue of trade ILO Member Countries agreed to prohibit and eliminate the worst form of
liberalisation, employment and child labour defined to include all forms of slavery, child prostitution and
investment with a special focus on
pornography, use of children in drug trafficking and work which is harmful
poverty reduction and a proposal to
form a global commission of eminent to health, safety and morals of children. It was also recognised that child
persons to examine the social aspects labour is a product of poverty and the solution to the problem lay in economic
of globalisation. growth and development.
The Convention stipulates that signatory states must initiate adequate national
policies for the effective abolition of child labour and to progressively raise the
minimum age for employment or work to a level consistent with the fullest
physical and mental development of children. Further, according to the Worst
Form of Child Labour Convention, 1999 (No. 182), unconditional and worst
forms of child labour are defined to include slavery, trafficking, debt bondage
and other forms of forced labour, forced recruitment of children for use in
armed conflict, prostitution and pornography, and illicit activities.13
In the second model, children have no or very limited bargaining power and
parents take decision regarding children’s work keeping in mind their own
selfish and economic interest. This model argues that bargaining takes place
between parents and employer and the wage of the child is dependant on the
bargaining strength of the household determined by household income and
access to credit. Gupta (1998) on the basis of his analysis of some villages in
West Bengal, India has propounded the argument of negligible bargaining
power of children in a household.16
According to the model propounded by
Becker (1960), number of children/
Some of these household behaviour theories consider fertility as an important
fertility is determined by cost-benefit
analysis regarding children in a explanatory variable in the incidences of child labour. According to the model
household, considered as a single propounded by Becker (1960), number of children/fertility is determined by
decision-making unit. cost-benefit analysis regarding children in a household, considered as a single
decision-making unit. He argues that “an increase in income increases both
quantity and quality of children but the quantity elasticity of demand for
children is smaller than the quality elasticity of demand for children.” According
to Easterlin (1969), in societies where children enter labour force at a younger
age, the economic expectations from children’s prospective income would
determine the fertility decision of the household and produce more children
with a view to augment household income.
Some scholars view the need for more children as an indispensable input in
peasant economies. The work of Mamdani (1972) on peasant’s households in
Studies have documented that the cost India highlighted the significance of the children’s work for the poor peasant
of bearing and rearing farm children families. He argues that high fertility is essential in peasant economy as it
is lower than the non-farm children requires a large number of working children to complement the work of
and that opportunity cost of rural adults. UNICEF views the lack of investment in basic services and labour-
children is smaller compared to those saving technologies in an underdeveloped economy as a crucial factor which
in urban areas – a factor explaining
makes a large number of children desirable, even essential, as a source of
high fertility in rural areas.
help in fields and homes.18
Furthermore, a large family size may also explain high incidences of child
labour. A study on developing countries by Lloyd (1994) argues that larger
household size reduces children’s participation and progress in schools as well
as parent’s spending on education of their children. It is likely that the larger
families increase the possibility of children working. It is also suggested that
children with more siblings are likely to work longer hours when they are
older. The impact of high fertility on child labour are felt more by the first
a large family size may also explain born children and the gender dimension makes it likely that boys are engaged
high incidences of child labour. It is in market work while girls are engaged in domestic or farm work.21
likely that the larger families
increase the possibility of children According to Dessy (2000), low parental skills coupled with high fertility
working. It is also suggested that
results in an increased reliance on child labour and the creation of a poverty
children with more siblings are likely
to work longer hours when they are trap which reinforces the trend towards child labour through generations.
older. “The greater the number of children, the less income is earned by the parent
[due to time devoted to child care activities] and the more the incentive to rely
on child labour rather than send children to school.”22
The study by Sharif (1994) points out that children from landless and marginal
farms are more likely to engage in wage labour whereas those with larger
farm lands are engaged in agricultural activities. As the farm size increases,
child labour also increases. However the situation reverses where larger farm
households cease to operate their own land and rent it out.” Christian and
Kanbur (1995) also infer that both the size of the farm and the mode of
Eswaran’s (1996) model links child labour, poverty and fertility. He argues
that a poverty trap is created when child mortality is high and child labour
is socially acceptable. Eswaran proposes that “when children are needed to
provide old-age security, allowing parents to put their children to work [it]
may induce parents to substitute away from small, more educated to large,
less educated families.”24 Similarly, according to Cain, Mazumdar and Alam
A poverty trap is created when child
mortality is high and child labour is
(1980), for poor households, child labour is part of the household self-insurance
socially acceptable. strategy to minimise risk arising out of loss of employment, failed crops,
disability, illness, etc.25
The above discussion highlights the fact that parents’ decision to send children
to work is predominantly a product of household poverty, apart from other
factors such as fertility, size of the household and size of land holding. In
poverty situations, children are considered as an asset to supplement household
incomes and ensure subsistence of families. In addition, the level of economic
development of a country is considered to be an important factor in explaining
the prevalence of child labour as discussed below.
According to Fallon and Tzannatos (1998), “both the incidences of child labour
and children as a proportion of the total labour force decline with per capita
Bhalotra (1999), using data for 2,400 households in rural Pakistan, points out
that acute poverty makes the opportunity cost of sending a child to school
very high who may then join the labour force. This evidence is stronger for
boys than girls.33 Ray’s (2001) comparative study on Bangladesh, Nepal and
Pakistan shows a “U-shaped relationship between inequality and child labour34
It must also be pointed out that similar poverty levels may produce different
outcomes for child labour and schooling. e.g, two states in India – Kerala and
Uttar Pradesh (UP) – despite showing similar economic growth, have shown
diverse pattern vis-à-vis child labour as well as educational attainments of
It must also be pointed out that
similar poverty levels may produce
children. Compared to 98 percent school attendance of 12-14 years rural girls
different outcomes for child labour in Kerala, only 32 percent of rural girls attend school in UP. Here, social and
and schooling. e.g, two states in India cultural factors as well as the literacy levels of the parents have been cited
– Kerala and Uttar Pradesh (UP) – as factors influencing incidences of child labour.36
despite showing similar economic
growth, have shown diverse pattern Studies have also documented a negative correlation between child labour
vis-à-vis child labour as well as
educational attainments of children.
rates and adult literacy levels. Ray (2000) points out that all countries in South
Asia including India show a positive correlation between rising adult education
and children’s schooling, although such correlation is weakest in Bangladesh.37
The model developed by Grootaert (1998) based on his study of Côte d’ Ivoire
has identified certain factors which affect a household’s decision to choose
between child labour and schooling. For example, children from parents with
no or low education are more likely to work and this effect is stronger for
younger children and in rural areas. Besides, father’s education and the
mother’s employment contribute to increasing a child’s probability of going to
school and not working, particularly in urban areas. In addition, the age and
the gender of the child play an important role. For example, in urban areas
girls are less likely to attend school exclusively, combine work and school
relative to working only, and are more likely to be involved in home care
activities. In rural areas, a female head of household would further increase
the odds that a child will work. Apart from these factors, “the poverty status
of the household matters the most in the decision between the work-only
option and the work-school combinations”.38
Overall, poverty, lack of education and child labour reinforce each other
because “child labour excludes or restricts access to education and jeopardises
the chances of upward social mobility, it perpetuates poverty since lack of
education impacts on earnings throughout life [as] it depreciates the human
capital needed for economic and social development”.39
Overall, poverty, lack of education
and child labour reinforce each other
because “child labour excludes or Based on the above discussion, it can be said that the linkages between child
restricts access to education and labour, education and schooling, and poverty are significant and that child
jeopardises the chances of upward work largely comes at the expense of schooling. Thus, any positive policy
social mobility. intervention to alleviate the issue of child labour should look at these inter-
linkages in totality.
Certain studies such as those by Bequele & Boyden (1988) and Jomo (1992)
also point to a wage differential between adult and child worker, thus
perpetuating child labour. It has also been pointed out that monopolistic
demand conditions arising out of concentrated ownership of land, credit and
In a competitive market where wages product monopoly, share cropping arrangements, restrictions of the mobility
are flexible, child labour can easily
substitute adult labour in the labour
of labour and lack of alternative job opportunities also keep the wages of child
market and this may encourage child labour very low. Since child labour is prone to lower mobility and lack of
labour. alternative jobs, the wage becomes a function of the elasticity of supply.
The studies by Levy (1985) on Egypt and Rosenzweig on India (1981) points
out that children’s labour supply is elastic with respect to their wages or the
own-wage elasticity of child labour is positive and higher for younger children.
However, the cross-wage elasticity with the labour supply of the mother was
found to be negative which implies that an increase in mother’s wage would
reduce the participation of female children in the workforce. In case of men’s
wages, the opposite hold true as it has a cross-wage elasticity of 1 in case of
boys and zero in case of girls.40
Ilahi, Orazem and Sedlacek’s (2000) Brazilian case study on the relationship
between child labour and future adult earnings and poverty status, point out
that entry in the workforce before age 13 results in a reduction in adult wages
of 13-17 percent and an increase in the probability of being in the lowest two
income quintiles of 7-8 percent. According to them, policies which delay age
of entry into the labour market may have significant impact on adult incidence
of poverty. Besides, although child labour reduces the productivity of schooling,
even if the child works while in school, the net effect of an additional year
of schooling on adult wages would be positive. Therefore, policies which delay
school drop-out rates even in cases of working children would be effective at
mitigating adult poverty. The findings support the theoretical argument of
Baland and Robinson (2000) that restricting child labour would be welfare
It has been argued that prevalence of
child labour reduces household
improving.41
income as it keeps adult wages
depressed and contributes to high It has been argued that prevalence of child labour reduces household income
rates of adult unemployment. Hence a as it keeps adult wages depressed and contributes to high rates of adult
ban on child labour would raise unemployment. Hence a ban on child labour would raise adult wages and
adult wages and households’ income households’ income and enable them to send their children to schools. However,
and enable them to send their
children to schools.
according to Khan et al (forthcoming), given the excessive supply of labour,
a large-scale withdrawal of child labour would not have any impact on the
adult wages. Thus, “from the point of view of child welfare, it may be easier
to approach the issue with a set of policies that eradicate work hazards,
encourage schooling along with work on raising household income of the
poor.”42
The model developed by Basu and Van (1998) theoretically analyses the
phenomenon of child labour in the household decision-making context and
argues that child labour is a facet of low adult wages in the labour market
and consequent poverty. They postulate a negative correlation between child
work and parent’s income and argue that a child’s “non-work is a luxury good
In sum, these theories’ main argument for the prevalence of child labour rests
on the centrality of the labour market failures, arising out of wage differential
between child and adult labour and low adult wages.
Baland and Robinson (2000) argue that there is a trade-off between child
labour (current income) and accumulation of human capital through education
(future income). Accordingly, child labour is inefficient when it impairs the
Putting children to work may future earning ability of a child. Putting children to work may augment
augment current income but may
current income but may undermine their human capital development and
undermine their human capital
development and parents’ failure to parents’ failure to “internalise the trade-off between child labour and earning
“internalise the trade-off between ability” would result in the child labour to persist.46
child labour and earning ability”
would result in the child labour to Further, based on the inter-generational flow of wealth theory, Baland and
persist. Robinson argue that child labour can be Pareto-inefficient even when parents
are fully altruistic towards their children and may persist “either when parents
leave their children no bequests or when capital markets are imperfect.”
According to them, child labour is inefficient when it is used by parents to
transfer income from children to parents (negative bequest) or used as a
substitute for borrowing due to capital market imperfections (transferring of
income from future to the present). “Child labour is inefficient when the
family is so poor that parents do not leave bequests to their children. When
bequests are positive, parents completely internalise the adverse impact of
child labour on the future income of their children since, by reducing bequests
accordingly, they can compensate themselves for the current income they lose
The model developed by Basu et al (1998) points out that child labour is “a
function of returns on schooling (skilled or educated wages), the child wage,
the cost of education and the degree of difficulty in obtaining funds for
education.” The model argues that decision whether a child goes to work or
school is determined by asset distribution across households as well as capital
market imperfections (when borrowing is costlier than lending due to difficulty
in monitoring loans).49
Ranjan, (1999), also has postulated a two-period model in which child labour
arises as the result of non-existence of market for loans against future
earning of children. Ranjan (2001) further developed an overlapping generations
general equilibrium model where inefficient child labour arises due to credit
constraints. It derives a positive relationship between inequality in the
distribution of income and the incidence of child labour and argues that as
long as returns on education are high and the credit markets50 allow sufficient
flexibility to allow parents to borrow for their children’s education, parents
would prefer to send their children to school thereby reducing or eliminating
child labour altogether.51 The finding that poverty compels child work is
consistent with households being chronically poor or else with their being
transiently poor and credit-constrained.
The new global estimates on child labour put the number of economically
active children ages 5 to 17 years at 352 million which is 23 percent of their
total population. The child labour constitutes nearly 70 percent (245 million)
of the economically active children. On average, more boys tend to work as
child labour than girls although between age groups 5-14, there is no substantial
gender difference (Table 3).
Gender and Economically Child Child Labour Children in CHW CHW as Work Ratio
Age Group Active Labour as percent Hazardous as percent percent
Children (‘000s) of EAC Work of EAC of child
(EAC) (‘000s) (CHW) (‘000s) labour
However, the data about girl child labour is often underestimated as it does
not take into account the activities in the household enterprises and unpaid
economic activity around the household. Participation rates of children are
higher in rural areas than in urban centres and most of the working children
are found in the developing regions of the world where one out of every four
Among the developing regions, South child in the age group 5-14 is a working child.53
Asia region54 accounts for the largest
concentration of child labour and of
History of Child Labour in South Asia: An Overview
children not attending school. A
conservative estimate puts the number Among the developing regions, South Asia region54 accounts for the largest
of such children (5-14 years) between concentration of child labour and of children not attending school. The region
20-30 million in five countries of
South Asia.
thus has become a focus of the international campaign to eliminate child
labour. These children work either in a family business, outside home or
employed as wage earner. A conservative estimate puts the number of such
children (5-14 years) between 20-30 million in five countries of South Asia.55
The South Asian Coalition on Child Servitude (SACCS) however estimates
16000
14802
14000
12000
Child Labour (in '000)
10000
8000
6000 5368
4000
2835
2000 1200
48
0
Bangladesh India Nepal Pakistan Sri Lanka
Since the 1980s, the percentage of children (10-14 years) in the labour force
has declined in all countries of the region, although the rate of decline varies
between the countries of South Asia (Table 4).
Based on the ILO data on labour force participation for the years 1950-1995
and projections for the year 2010, Castle, Chaudhri and Nyland (2000) have
given an overview of the child labour situation in South Asia over a period
of time. According to them, although the latest ILO world estimates on child
labour indicate that the problem has intensified, “it is largely a statistical
illusion.” The increase, they argue is due to better survey methods with a
larger focus on the informal sector as well as girl and unpaid child workers.
The section below discusses the overall child labour situation in South Asia.
A look at the index of total child labour (10-14 years of age) in South Asia
with 1970 as base year shows that in the years 1950 and 1960, Sri Lanka had
a higher child labour index compared with all other countries of the region.
Amongst the countries of South Asia, In the year 1995, the same figure had declined to become the lowest amongst
Sri Lanka in 1995 recorded the best
the countries of South Asia. This year also saw a marginal decline in the index
figures in terms of high proportion of
children in school education as well of child labour for India whereas the other 3 countries (Bangladesh, Nepal and
as a low percentage in child labour Pakistan) recorded an increase in the child labour index. Projecting this trend
compared to other countries of the into the future indicates that by the year 2010, Sri Lanka would reach zero
region. on the child labour index. All other countries would also show a varying
degree of decline, except Nepal (Table 6).
Amongst the countries of South Asia, Sri Lanka in 1995 recorded the best
figures in terms of high proportion of children in school education as well as
a low percentage in child labour compared to other countries of the region.
The percentage of child labour in India 4.6 percent, 10.1 percent in Bangladesh
while it was 5.3 percent in the case of Pakistan and 13 percent for Nepal
(Table 7).
There is extensive trafficking in both women and children, within the country
and overseas (primarily to India, Pakistan), often for prostitution and at times
for forced labour. Some children are also trafficked to the Middle East to be
used as camel jockeys.61 UNICEF has estimated that there are about 10,000
child prostitutes in the country. Other estimates have been put the figure as
high as 29,000.
According to the census figures of 1981, the total number of working children
in India was 13.6 million which came down to 11.28 million according to the
The NSS 55th round (1999-2000) has
1991 census (Table 9). The NSS 55th round (1999-2000) has estimated the
estimated the number of working
children at 10.4 million while the number of working children at 10.4 million while the 43rd Round NSS conducted
43rd Round NSS conducted in 1987- in 1987-1988 had estimated 17.5 million working children. As per these
1988 had estimated 17.5 million estimates, there is a progressive decline in the number of working children.64
working children. As per these In 1981, children constituted 6 percent of the total labour forces but in 1991,
estimates, there is a progressive they constituted 3.59 percent of the total labour force. Likewise, in 1981 the
decline in the number of working
number of working children was 13.6 million compared to 11.28 million in
children.
1991. Amongst the states, Andhra Pradesh has the highest number of working
children at 1.66 million. Besides, Madhya Pradesh, Maharashtra and Uttar
Pradesh have more than one million working children. Majority of children
are engaged in rural areas, in agricultural and allied activities including
cultivation, livestock, forestry and fisheries.65
Forced/bonded child labour is also reported from various parts of the country
although the number is not very significant. Industries that involve forced or
indentured child labour are considered to be brassware, hand-knotted wool
carpets, explosive fireworks, footwear, hand-blown glass bangles, hand-made
locks, hand-dipped matches, hand-broken quarried stones, hand-spun silk
thread and hand-loomed silk cloth, hand-made bricks, and bidis. A number
of these industries also expose children to hazardous working conditions.68
Human rights organisations estimate that the carpet industry alone accounts
for nearly 300,000 working children working, many of them under conditions
The older and urban children tend to be more involved with the non-
agricultural work. The non-agricultural households show lower work
participation rate amongst children. Economic participation rate of children
also decreases with a household occupational shift from agricultural to non-
agricultural activities. Besides, work participation rate of children tends to
decrease as the size of the family increases. No direct correlation exists
between work participation rate of children and the size of household
landholdings although economic participation rate increases with increase in
the size of the household landholding.76
The child labour estimates in the carpet industry in Nepal vary between
official figures of 0.8 percent to 50 percent by the NGOs. A survey conducted
by Government of Nepal on 23 carpet units estimated that 9 percent of the
Most of these child labourers are migrants from the rural areas, 47 percent
of whom are brought by brokers, the rest by relatives. The amount paid to
the family for taking children to work in the carpet industry ranges between
The amount paid to the family for US$ 40-100. This is the only income that the family receives for the child’s
taking children to work in the carpet work as very few children were known to send money back to their families
industry ranges between US$ 40-100.
The children are paid a flat rate of
later. The children are paid a flat rate of Nepali Rupees 350-750 per month
Nepali Rupees 350-750 per month (US$ (US$ 7.95-17). According to CWIN, 7-8 percent of the child labourers work to
7.95-17). repay debt incurred by their parents, which often is a pittance: Nepali Rupees
100-15,000 (US$ 2.50-US$375).77
Another large export earner in Nepal is the tea industry. Here again, it has
been estimated that children form 10 percent workforce in this industry, of
which 20 percent are the sole earner of the family.78
Like in other countries of South Asia, agriculture is the largest child labour
employer; followed by the informal sector including domestic work, street
vending, and family businesses; and hazardous work, such as the leather,
surgical instruments, and brick kiln industries. According to the president of
the Punjab Labourers Front, nearly 100,000 children between the ages of 5
and 12 years are working in more than 4,500 brick kilns in Punjab. Besides,
children as young as three also work in certain sectors. According to the
Human Rights Commission of Pakistan 1999 survey, approximately 4,000
children (mostly aged 3-8 years) are employed in auto workshops in the
Mardan district of the NWFP.
Pakistan is also a source and conduit for trafficking in women and children
for sexual exploitation, prostitution and for use as bonded labour. In remote
areas of rural Sindh, bonded agricultural labour and debt slavery have a long
history. According to human rights groups, about 4,500 bonded labourers were
maintained by landlords in rural Sindh. A 1998 study done by the trade
In remote areas of rural Sindh, federation noted that nearly 200,000 families work in debt slavery in the brick
bonded agricultural labour and debt kiln industry. The carpet industry is also known to use bonded child labour
slavery have a long history. According
to human rights groups, about 4,500
based on the peshgi system (where worker is advanced money and raw
bonded labourers were maintained by material for the carpet to be made by him) and is often exploited due to lack
landlords in rural Sindh. A 1998 of education and ability to calculate. 82 In carpet weaving sector in particular,
study done by the trade federation children from illegal immigrant families from Bangladesh and Burma are
noted that nearly 200,000 families involved in debt bondage. Bonded labours usually though not exclusively
work in debt slavery in the brick belong to the unskilled, low-caste and non-muslim section of the population.
kiln industry.
Certain studies indicate that children from all the sections of society are
involved in debt bondage.83 Besides, phenomenon of child soldiers is not
unknown. Although government does not recruit children for this purpose,
some groups such as the Tehrik-e-Nifaze-e-Shariat-Muhammadi have recruited
teenagers to fight alongside the Taliban in Afghanistan.84
The sporting goods industry in Pakistan, again in the Sialkot and adjoining
area, is another export industry that uses child labour in its production
process (football stitching, etc.) employing between 20-25 percent of children
(12-15 years) in its labour force. Here again majority of them work in family-
based cottage set ups.85
One of the most destructive forms of child labour in Sri Lanka prevails in the
fishing industry. Children are recruited in the fishing ‘vaadiyas’. Vaadiyas are
very remote and children are kept in conditions of virtual slavery. (ILO-IPEC,
Country Paper: Sri Lanka, September 1999)
The LTTE uses high school level children for work as cooks, messengers, and
clerks and occasionally, assist help build fortifications. LTTE is also known to
recruit children as young as age 10 for 2 to 4 years in special schools and
impart mixture of LTTE ideology and formal education and use children as
young as 13 years of age in battle. Beginning in May 1999, a programme of
compulsory physical training and mock military drills was initiated for all
sections of the population including school children in the LTTE controlled
area to provide ready army of people, particularly children. 87 An assessment
of the LTTE fighters killed in combat reveals that 40% of its fighting force are
both males and females between 9 and 18 years of age. Another study done
When a child’s family is very poor, there is no creative growth, no play for
the child. There’s no time that the child can spend doing something that is
not essential for the family’s survival.
Poverty
Poverty, both at country and household level constitutes the most significant
determinant of child labour in South Asia. In Bangladesh, for example, about
55 million people in Bangladesh live below the poverty line. In this situation
of abject poverty, child labour contributes about 20-25 percent of the family
income, most of which is spent on food, thus making child income critical to
their survival. Similarly, in India, poverty constitutes the major factor in
driving children to work. The backward and drought-prone areas record high
incidences of child labour. The population of people living below poverty line
in India is nearly one third of the total population. Studies have found that
nearly half of the families with working children are below the poverty line.
As estimated number of about 77 million children under 14 years belong to
families below the poverty line. This provides a ready pool of abundant child
labour supply that outpaces demand. The result is a downward trend in wages
which keeps the child and his family in perpetual poverty thereby further re-
enforcing child labour.
In Pakistan and Sri Lanka too, household poverty and need to supplement
household income is one of the important factor for sending their children to
work. But, in variance of the general trend in other countries of South Asia,
children’s assistance in the household enterprise is also cited as a major factor.
The survey conducted in Pakistan likewise, noted that 54 percent of the
parents let their children work primarily to assist in household enterprises to
supplement household incomes. Approximately 27 percent parents cite need
to supplement household income as another reason and 14 percent says that
their children work because no one else is available to undertake the household
chores. About 0.85 percent parents/guardian cited payment of outstanding
debt and 3.05 percent cited ‘other’ reason. The Pakistan survey also did not
ascertain correlation between household income levels and working children
as nearly 39 percent of children are from households, whose income is
between Rs. 2501-4000, followed by 31 percent in the income range of Rs.
1501-2500. From household with income of Rs. 4001 and above, 21 percent
children work and 9 percent come from income group with less than Rs.
1500.91
Sri Lanka too shows similar pattern as majority, i.e. 58.3 percent of parents
cited assistance in the household/family enterprise as the reason for letting
their child work while 28 percent parents let their children work for economic
The pattern of landholdings in India gains in order to supplement household incomes. Poverty factor influences
due to a limited land reform
programme has perpetuated debt
nearly 28 percent of children who work for money.92
burden of rural population and
plight of rural labourers and Mode of Production and Pattern of Landholdings
marginal farmers forcing their
children to work.
The predominantly agricultural economies along with the feudal landholding
pattern also perpetuate the phenomenon of child labour in the region. For
instance, the pattern of landholdings in India due to a limited land reform
programme except in the state of West Bengal (one of the lowest child labour
rates) has perpetuated debt burden of rural population and plight of rural
labourers and marginal farmers forcing their children to work. The consequent
migration from rural to urban areas to escape rural poverty in search of jobs
is fairly widespread in India which also contributes to child labour.
In Sri Lanka, the number of children who are sent to work to supplement
household income is 243,893 out of which 103,664 do not attend school which
means that due to lack of resources/poverty, parents cannot send their children
to school. According to the survey, 20 percent of children have stopped
schooling to engage in economic activities and 12 percent have dropped out
due to financial difficulties and 11 percent due to uncertainty of higher
The phenomenon of child labour in education.94
India is closely linked to illiteracy,
in addition to poverty. According to Similarly, the phenomenon of child labour in India is closely linked to illiteracy,
ORGB 1983 survey, over 64 percent of in addition to poverty. According to ORGB 1983 survey, over 64 percent of
the child workers were illiterate. the child workers were illiterate. According to Myron Weiner, prevalence of
child labour is attributable “largely to the failure of the educational system.”
Since primary education is not compulsory in India, children drop-out of
school and work as labourers. The lack of adequate infrastructure and link
between life opportunities and education in addition to economic circumstances
act as another contributory factor to the causes of child labour in India.
Other Factors
Poverty continues to be the main reason for the ample supply of cheap child
Enforcement of the child labour laws labour and provides the most compelling economic rationale for child labour.
remains a problem due to lack of However, on the demand side, children are also employed because (1) children
resources and corrupt practices in all
are docile, (2) they are less likely to be absent – this in important in the
countries of South Asia. Since
majority of children are located in informal sector where labourers are employed on a daily basis and thus
informal sector and household require full strength of workers every day, (3) they do not form unions and
enterprises, monitoring and their employment reduces the possibility of hartal or strike, (4) management
enforcement becomes a problem due to of children is easy, and (5) the owners feel sorry for poor children and give
resource constraints. them work. In some sectors, children are believed to be uniquely suited for
the work such as carpet and gem industries. It is said that nimble fingers can
weave 41 greater number of knots in carpets and polish tiny gems more
efficiently.97
Impact of Trade-Sanctions
The Nineties saw the use of trade sanction to alleviate the problem of child
labour by the US policy-makers and also by some countries in the EU. The
sectors that were affected in countries of South Asia were apparel/ garment
industry in Bangladesh, carpet industry in Nepal, football industry in Pakistan,
and carpet and bidi industry in India. The immediate cause of these trade
The Nineties saw the use of trade
measures was some legislative initiatives taken within the US Congress,
sanction to alleviate the problem of particularly the 1993 the Harkin Bill – the Child Labour Deterrence Act,
child labour by the US policy-makers introduced by Senator Tom Harkin which proposed to ban the import of those
and also by some countries in the EU. goods into the US that used child labour in its production process, either
The sectors that were affected in wholly or partly.101
countries of South Asia were apparel/
garment industry in Bangladesh,
carpet industry in Nepal, football Bangladesh
industry in Pakistan, and carpet and The impact of the bill was most pronounced on the garment industry in
bidi industry in India. Bangladesh. In addition to this bill (which was not passed but caused severe
indirect impact), the American television newsmagazine – NBC Dateline carried
a feature on child labour in Bangladesh garment industry which manufactured
The overall impact on economy was also significant due to the loss of an
estimated half billion US dollars in 1994-1995. Critics of the trade sanction
approach point to the fact that the “loss of foreign exchange earnings of this
magnitude actually denies Bangladesh the opportunity to finance the industrial
and economic growth which would create the conditions under which child
labour would disappear.”103
However, by this time many children had already lost their jobs and were
neither given compensation or schooling. A study carried out in the fall of
1995 found that the number of child workers in BGMEA factories came down
to approximately 11,000. However, despite the fact that by September 1996,
Pakistan
In the case of Pakistan, the United In the case of Pakistan, the United States revoked trade benefits under the
States revoked trade benefits under GSP in 1996 for certain goods, such as leather sporting goods, surgical
the GSP in 1996 for certain goods, instruments, and hand-loomed carpets, due to the tardy progress on various
such as leather sporting goods, worker rights issues including child labour. Simultaneously, similar to the
surgical instruments, and hand-loomed situation in Bangladesh, the revelation about the rampant occurrence of child
carpets, due to the tardy progress on
labour in the football-making industry at Sialkot led to demand from western
various worker rights issues including
child labour. consumer/pressure groups, trade unions and NGOs for a boycott of products
from Sialkot unless child labour was banned.
In response, in 1996 the MNCs such as Nike and Reebok agreed with their
contractors in Sialkot to ban child labourers in football stitching. As a result,
the production of football shifted from houses to well-monitored centres. The
consequences particularly for women were difficult. According to Sajid Kazmi
of the Sustainable Development Policy Institute, Islamabad, “it is very easy for
women to stitch footballs in household, now many of them have lost work
because they cannot leave their homes; and if children cannot work in the
football industry, they will shift to more hazardous work.” An illustrative
example of such processes is that of Nabeed Arain, who used to stitch footballs
at home with his mother. When the ban was introduced, the 13-year-old went
to a small school funded by Reebok. In 18 months, he dropped out after his
teacher dismissed him as a slow learner. For the past six months he has been
grinding scissors in a dirty workshop in Sialkot for 12 hours a day as he learns
the complex process of making surgical instruments.” Although he gets more
money than in making football (Rs. 1,000, £ 11) a month, he is candid in
admitting that there is no choice that he has.105
Many of the children shifted to hazardous industries such as brick kilns and
car workshops and the production of surgical instruments. The consequences
for companies has been the rise in cost who felt that they are not being
According to some exporters of rewarded for their effort to eliminate child labour and little effort has been
football, implementing child labour made to improve actual working conditions. According to some exporters of
reforms has increased their football, implementing child labour reforms has increased their production
production costs, making their costs, making their products less competitive in the world market.106
products less competitive in the world
market.
As a remedial measure, a number of donor agencies and concerned MNCs
came forward for help. In 1997, Reebok pledged US$ one million as aid to
educational programmes in Pakistan in areas where child labour are
widespread. The UK Department for International Development (DfID)
provided funding for the National Rural Support Programme aimed at a credit
and savings scheme to augment families’ income to reduce the need for
children to work. The targeted villages were the ones where more than 30
percent stitcher households were located. Around 101 million Rupees (US$ 1.9
million) has been disbursed amongst 70,000 people and it is estimated that it
has benefited 40,000 children and their families until December 1999, including
former child stichers, their siblings and children involved in other work and
The project to eliminate child labour from the football industry based in
Sialkot monitors the production of footballs at established stitching centres,
and has been instrumental in setting up 185 rehabilitation centres to educate
former child labourers and their younger siblings. The project also identifies
unemployed adults, especially women, from the families of former child stitchers
to take up stitching work and replace lost income.109 Since the ban, 6,000
children have enrolled in new schools set up by the ILO to prepare them for
mainstream education. Up to 2001, 1,660 children have joined the mainstream
education; another 1,760 were dropouts who simply shifted to other jobs,
sometimes hazardous.
ILO has also funded inspectors to monitor small home-based football stitching
centres as it has been permitted to establish home-based workshops with 3
or more stitchers registered in order for women to work in their community
and safeguard family incomes. By the year 2000, 358 home-based centres
were set up and 146 larger centres for female stitchers. By the end of the
year, the number of home-based centres had increased to 360, larger centres
to 186, and combined centres to 64. The ILO, which monitors more than 90
percent of export production, reported that it had found no evidence of
children working in any of the registered stitching centres and that it found
In manufacture of surgical no unregistered centres.110
instruments and also carpet industry,
both export earning industries, the Simultaneously, some brand name leaders changed practices to reduce the
ILO along with SCCI has also exposure to child labour. While Nike reduced number of their suppliers,
initiated projects aimed at promoting Reebok opted for direct monitoring using local NGOs and in developing child
educational opportunities to ban
employment of child labour.
labour education programs.
India
In India, a select brand of bidi was subjected to a ‘detention’ order (preventing
marketing of the product in America) in 1999 when the US banned the import
of a particular brand of Indian bidi, alleging use of bonded child labour in the
industry. The move was prompted by a CBS News story, which ran a feature
on ‘60 Minutes’ regarding the use of bonded child labour by Mangalore-based
Ganesh Bidi Work.114
Since the US ban applied only to a select company, the immediate impact was
not significant in terms of loss of market. If the ban on import had extended
to the entire industry, India may have lost substantial revenue.115 The
consequent impact on the household industries employing child labour would
have been disastrous.
Impact of Labelling
Apart from outright ban, the trade intervention approach is also being pursued
through the instrument of labelling of child labour products. The campaign on
labelling was launched by humanitarian and religious organisations of consumer
Apart from outright ban, the trade
countries through public awareness and media campaign against child labour
intervention approach is also being
pursued through the instrument of in hand-knotted carpet industry in India, Pakistan and Nepal. The resulting
labelling of child labour products. decline in the demand for South Asian carpets prompted the governments,
The campaign on labelling was manufacturers and exporters to accede to the child labour-free labelling
launched by humanitarian and schemes being advocated. Subsequently, a number of social labelling initiatives
religious organisations of consumer in South Asia such as Rugmark, Kaleen, Smiling Carpet, etc. were introduced.
countries through public awareness
Other products from South Asia were also brought under the ambit of
and media campaign against child
labour in hand-knotted carpet labelling schemes such as apparel and textiles, etc. Some of the labelling
industry in India, Pakistan and initiatives and their details are given in Table 13.
Nepal.
The most successful amongst these was the Rugmark labelling programme
which was initiated as a result of consumer initiative in Europe, particularly
in Germany. In December 1995, the label became a legally-binding international
trademark in Germany and in the US in 1996 – the largest markets for carpet
exports from South Asia.116
*
Initiated by Government of India through its Carpet Export Promotion Council in 1995 in response to Rugmark.
Source: Janet Hilowitz, Labelling Child Labour Products: A Preliminary Study, Geneva: ILO/ IPEC, 2000.
Another ILO/ IPEC study (2000) has not reached a final conclusion about the
decline in the overall incidences of child labour in the carpet industry as a
result of labelling initiatives. However, it noted a decline in the incidence of
hired child labour during the last decade, particularly since 1995. For example,
in comparison with the 45 percent hired child labour figure quoted by the 1993
NCAER report,118 this study put the proportion at 11 percent in the core
carpet-belt. A disturbing finding of the study was that in the extension areas
of Allahabad and Kosambi and in the new areas of Bihar, the proportion of
hired child labour remained high at 36 percent and 40 percent respectively.
Overall, the impact of labelling was limited and largely restricted to individual
registered looms and even there, awareness amongst the weavers about these
initiatives was non-existent. Also, the sub-contracting system of production
method in the carpet industry makes monitoring a gigantic and difficult task.
The study has concluded that the legislative measures such as the effective
enforcement of the Child Labour (Prohibition and Regulation) Act, 1986 has
played the most important role in contributing to the decline in child labour,
apart from media campaigns, NGOs’ interventions, increased enforcement
and vigilance by government agencies and labelling initiatives.119
An empirical study conducted by the OECD does not establish any link
between labour standards and production costs but it has been noted that
labour standards are more closely adhered to in sectors exposed to international
competition compared to the sectors not exposed to international markets.
The analysis of the relationship between the differences in labour standards
and trade and foreign investment in OECD and some non-OECD counties has
concluded that it is not possible to provide empirical evidence on the linkage
between non-observance of core labour standards and unfair trade advantage.
Also, it is difficult to establish a correlation to show that countries with lower
labour standards show better export performance than countries with high
labour standards.122
According to this study, ‘core labour standards’ do not play a significant role
in shaping trade performance. “The view which argues that low-standards
countries will enjoy gains in export market shares to the detriment of high-
standards countries appears to lack solid empirical support. ... These findings
also imply that any fear on the part of developing countries that better core
According to the OECD study, ‘core
standards would negatively affect either their economic performance or their
labour standards’ do not play a competitive position on world markets has no economic rationale.”123 The
significant role in shaping trade converse logic of the same argument would indicate that leveraging trade
performance. “The view which argues intervention to address compliance with labour standards (such as eradication
that low-standards countries will of child labour) is unlikely to bear many fruits.
enjoy gains in export market shares
to the detriment of high-standards
countries appears to lack solid
Further, the idea that trade under the WTO regime can be an effective
empirical support. instrument to secure labour rights including an end to the problem of child
labour assumes that universalisation of labour standards across countries and
regions is desirable, is untenable both theoretically and empirically. According
to Robert M Stern (2000) “because of the diversity of labour standards in
countries with differing national characteristics, policies, and institutions, the
case for devising WTO rules and disciplines to improve core labour standards
in low-income countries cannot be convincingly made.”124 According to
Srinivasan (1998), differences in labour standards among various countries is
• Since incidences of child labour declines with rise in per capita income of
countries, facilitating free trade and open markets is the best way to
ensure economic development and the consequent elimination of child
labour and also raising other labour standards.
• At the international level, ILO not the WTO, with its long and credible
record in the field of child labour and other labour standards should form
the basis of multi-lateral understanding and solutions to the problem. The
ILO’s approach based on consensus, prescription and persuasion on labour
standards is sustainable in the long term rather than the punitive measures
as contemplated by the trade-based approach within the WTO forum.
Recently ILO has also launched a highly successful effort to obtain universal
commitment to the core labour standards. Since 1995, he number of
countries that have ratified all the core ILO conventions has more than
doubled.
• The trade unions along with the civil society needs to demonstrate so as
to why governments should prioritise children and public spending on
children, rather than cutback. This can be done through practice, through
good programmes, good work with children and education, health,
livelihoods, etc. The civil society has also to do an analysis and be able to
raise issues related to the economic models in their societies and its
negative social impacts on children. Strong advocacy is required on budget
allocations to improve the well being of children.
• There are various ways to arbitrate and the first and foremost is prevention.
The second one is withdrawal, but experience has shown that it is not
enough to just withdraw children from work, but to provide a package of
assistance. It had to be ensured that children are not just taken away, but
are given alternatives. The third area of intervention is protection but it
is a transitional measure. It remains the goal that children below the age
of 15 should not be working. However, often children simply cannot be
immediately removed, there can be programmes to protect children and
improve their working conditions.
1996 2000
1. The Freezing Effect - Lack of Coherence in the New 1. Implementation of the WTO Agreements: Coping with
World Trade Order the Problems
2. Competition Policy In A Globalising And Liberalising 2. Trade and Environment: Seattle and Beyond
World Economy 3. Seattle and the Smaller Countries
3. Curbing Inflation and Rising Prices - The Need for 4. Dispute Settlement under the GATT/WTO: The
Price Monitoring Experience of Developing Nations
n Foreign Orders - please pay by a Bank Draft (if possible drawn on any Indian Bank)
Name: _____________________________________________
Address: ____________________________________________________
____________________________________________________
____________________________________________________
1. ___________________________________________________
2. ___________________________________________________
3. ___________________________________________________
4. ___________________________________________________
5. ___________________________________________________
6. ___________________________________________________
Goals
Enable and empower Create a questioning Promote equity between and
representatives of the civil society, society through empowerment among the developed and
from developing countries in of civil society representatives developing countries through
particular, to articulate and thus ensuring transparency well-argued research and
advocate on the relevant issues at and accountability in the advocacy on the emerging and
the appropriate fora. system. relevant issues.
ISBN 81-87222-82-4