Politik und Gesellschaft Online
International Politics and Society 3/2001

Children Labourers Without Alternatives *

Shahid Ashraf

The practice of child labour, with its numerous manifestations, is widely prevalent in different economic activities in all developing countries, and India is no exception. The industries most notorious for employing child labour include carpet weaving, fireworks and match manufacture, "bidi"-making, glass and bangle manufacture, construction work and rag-picking, not to mention child prostitution. The result is a childhood destroyed, a dismal future, a wounded psyche, and an imbalanced society and economy. The carpet industry of India-- a serious contender with countries such as Iran, China, Pakistan, Afghanistan and Nepal in the markets of North America and Europe-- has come increasingly under international pressure on the issue of child labour. This has given rise to many controversies and face-offs in the international trade arena.
The present paper, which concentrates specifically in the Indian carpet industry, examines the reasons for the presence of child labour and the effects of the international campaigns against it,. It draws much empirical information from the author's recents visits to Mirzapur and Bhadohi.


The Background

Child Labour in India

According to the 1991 Census of India (the last one to date), more than eleven million children were employed in commercial activities. This does not include children engaged in home-based work, for example children helping their parents in agriculture, basket weaving, etc. The 1991 census also shows that 90 million out of the 179 million children in the 6 to 14 years age-group do not go to school. The chances are high that many of these 90 million children are child labourers.
The causes for the prevalence of child labour in India have largely been associated with poverty. Low adult literacy levels, the absence of compulsory primary education along with a high drop-out rate are part of the child labour syndrome. This has been compounded by social and cultural factors which force the continuity of trade and skill in a particular caste or community at an early age. In fact, Indian society shows scant disapproval towards child labour.


The Indian Carpet Industry

The Persian hand-knotted pile carpets are generally woven in the south-eastern districts of Bhadohi, Mirzapur, Sonebhadra and Allahabad in the State Uttar Pradesh. Besides these, there are varieties of Nepali/Tibetan carpets as well as tufted carpets, which have different looms and methods of weaving in comparison to the Persian carpets, henceforth exclusively referred to as carpets. The quality of carpets depends on the knots per square inch, the motifs and intricacies of design as well as the number of colours used.
The value of carpet exports, which was 600 million Rupees in 1951-52, slipped to 100 million Rupees in 1984-85. The 1980s was the period of Iran's political turmoil during which their carpet exports to the Western world dropped. The vacuum was filled by increasing supplies from other countries such as India whose exports jumped to 13,500 million Rupees in 1993-94, but which have since stagnated around 15,000 Rupees. (Ministry of Commerce, Government of India, 1998-99). The main importers of Indian carpets are Germany and the USA, however the stagnation seen in carpet exports in the late nineties has been attributed in large part to the issue of child labour.
The Indian carpet industry has few large businesses and consists mostly of small family enterprises. Many of them entered the trade in the boom period of the 1980s. The exporter is the prime figure around whom the overall organisational structure of the carpet industry revolves. He is the link between the foreign importers and the local contractors, loom owners and weavers. Carpet manufacturing consists of three stages-- weaving, washing and finishing - with child labour concentrated in the first stage. Various studies have thrown up a wide range of child labour figures: from 200,000 to 1,500,000 or from 3 to 70 per cent of the industry's total labour force. But there are few systematic and objective studies. The Development Research Institute of the Tilburg University in the Netherlands estimated 229,000 weavers and 270,000 other workers of whom 23 per cent were child labourers in 1980. According to the National Council for Applied Economic Research, "the overall employment of children in the carpet industry has come down marginally from the earlier level (1992) of 8 per cent to 7.5 per cent in 1994."
A detailed report on the carpet industry for the International Labour Organisation (Ashraf and Barge, 1994) estimated that out of an approximate total of 650,000 workers, 550,000 were weavers. Of these seven per cent were definitely children. When taking into account the categories "probably child" and "absent weaver", the estimated percentage of child weavers rose to 21.7 per cent. It would therefore be safe to assume a percentage range of seven to 22 per cent for child weavers. This includes both the paid migrant child and the family member.


The Local Economy

Uttar Pradesh and Bihar (the State from which most of the migrant children arrive) are among the lowest ranked States in terms of almost all social and economic indicators. The work participation rate (total workers as percentage of total population) is low (32 per cent) compared to the national average (37 per cent). The female participation rate is only 12 per cent in Uttar Pradesh and 15 % in Bihar in comparison to the 22 per cent for all of India. Also the gross enrollment ratio in schools is lower than the pan-India level. There is a high drop-out rate after the primary classes implying that only a minority of children in the 10-12 years age-group are in school. The others are likely to be child labourers.
The rural wages for both agricultural labourers and non-agricultural labourers are lower in Uttar Pradesh and Bihar than in the rest of India. As Dreze and Gazdar (1997) point out, "ranked by their ownership of land, the bottom 40 per cent of all rural households in Uttar Pradesh owned 2.5 per cent of the total area in 1953-54 while the top 10 per cent owned 46 per cent of the area. More or less the same size distribution was observed in 1982." The situation of asset ownership has certainly not changed much in the 1990s.


Economic and Social Factors

influencing Child Labour A major cause of child labour both generally and in particular to the carpet industry - is the poverty in the eastern region of Uttar Pradesh. But there are also socio-political factors that keep pushing a child to work. Given that the literacy levels of the adults themselves are low, the parents attach little importance to the education of their children. This is then exacerbated by the fact that school attendance is not compulsory and access to schools in the countryside is often difficult. The consequence is that many children remain low-paid unskilled workers until they become adults. Even if a child completes high school, it is rare for job opportunities to come their way. Eastern Uttar Pradesh has a very thin industrial base and the work opportunities, besides those in agriculture for unskilled labourers are primarily in the carpet industry.
In the carpet industry the weavers and the loom owners are at the bottom of the hierarchy, they are paid the least and their work is the most monotonous. A study conducted in 1994 by Ashraf and Barge found that of all the loom owners (of which 96 per cent are also weavers), 77.3 per cent were illiterate and only 16.6 per cent had attended or completed primary school. 20.7 per cent of them were Muslims, 48.3 per cent were low-caste Hindus, 23.2 per cent were scheduled-caste Hindus and 6.1 per cent were from scheduled tribes (to which the caste system does not apply) . High-caste Hindus constituted only 1.7 per cent. There are social and cultural discriminations, which force the continuity of trade and skill in a particular caste or community at an early age. The low and scheduled-caste Hindus face barriers to upward mobility and their children remain in their socio-cultural straitjacket with no aspirations beyond that of the weaver, or the occupational category they have been born into. The rigid caste system tacitly encourages low and scheduled-caste parents to send their children to work at an early age.
According to Klaus Voll (1998), "conflicts between social structures ... and the requirements of modern economics and, therefore, also of social development, to which all the political parties, employers and trade union federations and also NGOs are verbally dedicated, have to be fought out primarily between the concerned actors in the changed atmosphere of rising expectations and demands of the so-called 'subaltern' sections of Indian society." Even the carpet industry is not immune to the rising expectations and a few low-caste weavers have become loom owners and exporters, but the basic social conditions of the majority remains precarious.
The carpet industry has developed a large and informal contracting and sub-contracting network. Weaving work is portioned out by the exporters through the contractors - who are usually influential local people - to the loom owners, themselves weavers, who in turn hire additional weavers. Any commitment or obligation to the workers is avoided. There is neither job security nor continuity of work in any of the enterprises involved. Wage payments are made on a piece-rate basis and virtually nobody works for daily or monthly wages. As a consequence the loom owners prefer to employ children because they cannot be unionised. Policies for protection and security of workers are absent in the carpet industry.
The initial piece-rate wages compare favourably with the rural wages prevalent in the region and apparently seem not to discriminate according to the type or the age of the weaver. Generally, the contractors take a 20 per cent commission on the piece-rate wage negotiated with the exporter. The source of exploitation is the unequal contractual relationship between the contractors and the loom owners. Once the adult weaver receives low payments he is forced to bring his own children or those of his relatives to weave along with him. Initially, the child is treated as an apprentice and no payments are made to him. After a year, 30-50 per cent of the total payment is made to a child. For migrant child weavers deductions for lodging and food are made. Full payments are made only after a period of 2-3 years of exploitative weaving.
Another exploitative practice is that wages are not paid daily or weekly. Contractors receive payments as advances equalling on average the total expected weaving charges of the carpet. The contractors subsequently loan out small amounts to the weavers during emergency or festivities, and the deductions (inclusive of interests) are made against the weaving payments. On many occasions the weavers find that they owe money to the contractors after the completion of a carpet. Thus they get "bonded" to a particular contractor and often involve their children in work to pay off the debt. Loom owners who have large looms under their control and need more weavers than the immediate family can supply prefer to hire weavers who can also bring along a child as the total payment due will be lower.
Schooling, which has been suggested by Weiner (1996) as the panacea for the elimination of child labour, gets bogged down by the absence of infrastructure, an irrelevant and outdated syllabus and the attitude of the teachers. The children from low castes are usually taunted by words such as "why do you have to come to school, you should go and do your work". A low caste parent told this author, "What can I do? I love my child as much as you love your child. I send the child to school but the teacher makes him work on his fields. Is it not better then, that he stays back and works with me?"
Besides the usual contributory factors of poverty, illiteracy, lack of employment opportunities, etc., there are certain very specific reasons for the presence of child weavers in the carpet industry and, no, it is not the nimble fingers argument. This argument has been used by the carpet industry to justify the fact that children are weavers. But weaving is a misnomer for Persian carpets is a misnomer as the process consists of tying a knot of woolen yarn to a cotton thread, cutting it with a sharp curved knife and repeating the process. The knots are compressed in a square inch, the higher the number of knots per square inch, the higher the quality of the carpet. This process needs strong fingers, particularly for high-quality carpets, and the good weavers tend to be aged between 18 and 35. A child is more liable to cut his finger while cutting the woolen thread after tying the knot in higher quality carpets, as more knots have to be compressed into a small area. For the carpet industry, the nimble fingers argument is merely another excuse for employing children, thereby also maintaining a pool of trained weavers for the future.
Carpet weaving is a continuous and monotonous activity, it requires sitting in a specified fixed position with full concentration. Generally two to five weavers work together on a single carpet, each having a horizontal area of 18 to 30 inches. The weaving has to be done together by all the weavers and the vertical level woven should not vary for any weaver. If there are variations due to a weaver being absent and his weaving area not being woven for a day, then experts will be able to point out that the carpet makes a sound like a torn cloth which brings down the overall quality of the carpet. Quality conscious exporters insist that all weavers weave together or not at all. Thus, the absence of a single weaver could stop the weaving for that particular day. A six by three foot carpet may take four to six months for its completion; whereas an adult may often be absent, a child, whether part of the household or hired, is more pliable and can be made to weave regularly over the completion period of the carpet. Also, inside the loom sheds smoking is strictly prohibited, not because of health reasons but because the woolen and cotton yarn catches fire very easily. This again restricts the preferred adult weavers to non-smokers. Therefore the loom owners for meeting quality level and delivery deadline prefer a mix of both adult and child weavers.
Female weavers are a rarity in the carpet industry. Carpet weaving is a male-dominated activity where two or more weavers sit together in close physical contact with each other. Aside from this, loom sheds are usually a part of the loom owners' house and loom owners hire the weavers and for a woman to go from her house to somebody else's house is socially not acceptable in rural India. Cultural sensitivity requires that women do not mix with men other than those in the immediate family. Encouraging women weavers would require women-only loom sheds and the formation of women cooperatives.
There are not many viable occupational alternatives in a predominantly rural economy. Agricultural labour is in demand during the sowing and harvesting season, which may not be more than 40 to 60 days a year. Actual wages are much below the official wage. During harvesting, wages are given in kind equivalent to 14-15 Rupees per day. Rock cutting and casual employment in construction sites are the alternatives to weaving. Women at times roll up tobacco in tendu leaves to make "bidi" for the local merchants. Another alternative that weavers have shifted to is making durries, or small rugs that do not require intensive training and which a single worker can complete in a week or ten days. In the city, a broad variety of skills are needed but obviously there is a limit to the number of tailors, carpenters and electricians that can get regular employment. Thus, when a rescued child is trained in these vocations, he often still turns back to carpet weaving. The migrant child workers who have come - or were forced to come - from the drought-prone poverty-stricken areas of Garwa-Palamu in Bihar end up in the carpet industry after being rescued and sent back to their villages. Either they come back to the loom owners in the Mirzapur-Bhadohi region of Uttar Pradesh or looms are set up in their villages in Bihar.


International Campaigns Against Child Labour: their Achievements …

In 1992, the International Labour Office (ILO) set up the International Programme on the Elimination of Child Labour (IPEC) with five participating countries and one donor country (Germany), to work towards the progressive elimination of child labour. Currently the IPEC has 24 participating countries and nearly a dozen donor countries. The Indo-German Export Promotion Project (IGEP) Director, Dietrich Kebschull, took the initiative with the South Asian Coalition on Child Servitude (SACCS) to set up a system for labeling carpets as being made without child labour. Along with exporters and NGOs, the IGEP developed the "Rugmark" brand for such carpets. The Rugmark Foundation was set up to act as an independent, non-commercial international certification and monitoring system. Its Rugmark label guarantees the importers that the carpet has been manufactured or exported by a company which has voluntarily committed itself to work without children. To verify the claim, Rugmark sends professional experts to do random checks at carpet looms (Martine Kruijtbosch, 1995). The Rugmark Foundation is also involved in the rehabilitation of child weavers.
Another labeling and rehabilitation exercise is STEP, in association with Switzerland and other European countries. Swami Agnivesh with Bandhua Mukti Morcha (BMM), and Kailash Satyarthi with the South Asian Coalition on Child Servitude (SACCS) are the major Indian organisations working in association with other national and international organisations towards the elimination of child labour.
The child labour elimination campaigns have had an impact. But there is also the perception that they have given rise to useless controversies, over-hyped statements and inadequate rehabilitation schemes.
One effect has been that the use of child labour in the carpet industry has been reduced. The sensitisation component of the campaigns has successfully promoted the awareness that child labour is a malpractice and is bad for the individual, the society and the nation. This awareness has trickled down to the villages. I had the opportunity to verify this during visits to different villages in the carpet weaving zone.
The loom owners say that Rugmark inspectors (altogether twelve) have been very strict with them. Rugmark inspects the loom units on the basis of information provided by the exporters. If inspectors identify a bonded child labourer they put him in their rehabilitation centre in Gopiganj near Bhadohi-Mirzapur, which has over 60 rescued child labourers. These children get free primary education and vocational training in stitching, carpentry, painting, carpet-weaving and other trades. Other NGOs like the Centre for Rural Education and Development Action (CREDA) have concentrated on providing primary education to children in different villages. Rugmark also runs four primary schools meant for the children of carpet weavers. Through the threats of exporters registered with Rugmark and the highly publicised campaigns against child labour, the fear of employing children has started filtering in the carpet industry down to the weavers' level.
The hostility that was very pronounced in my earlier visits to the villages has subsided. During the 1994-95 surveys, villagers, at times armed with sticks, would stop entry to our group of investigators. But since then, the atmosphere as well as the attitude of the loom owners have undergone a visible change for the better. Access to the weaving sheds is more open and forced absenteeism (earlier weavers would make the child leave the weaving looms and on asking would say that the weaver is absent) is less.
The sensitisation programmes and the threat of a possible ban on exports have put pressure on the employers. The All India Carpet Manufacturers Association has issued statements about their intent not to employ child labour and to make the loom owners do the same. The Carpet Export Promotion Council and the Indian Ministry of Textile and Commerce have - under the name of Kaleen - taken a drive to register children-free carpet looms.
The campaigns have also put pressure on the Indian Government to tighten legislation on Child Labour. Since 1948, it has enacted laws that have a bearing on the practice of child labour: the Factories Act of 1948, the Shops and Establishments Act of 1948 and 1961, the Plantation Labour Act of 1951, and the Bonded Labour System (Abolition) Act of 1976. These Acts largely defined the age restrictions and working hours of children. In more recent times we have the Child Labour (Prohibition and Regulation) Act of 1986, which prohibits the employment of children below the age of 14 years in specific occupations and processes. The 1986 Act also regulates working conditions in the non-prohibited occupations and processes and lays down penalties for violations of its provisions. The campaigns have put pressure on us to take responsibility as individuals in a society and on the government to search for answers.


… and their Shortcomings

The campaigns for the elimination of child labour are not uncontroversial. There have been statements on the number and percentage of child weavers which have not been properly substantiated by systematic statistical methodology. Another problem is linked to the verification of the claims made by the brands like Rugmark or Kaleen. The carpet weaving area has become so widespread that it is difficult to physically check and police each and every loom, random or otherwise. Exporters may have looms spread all over the weaving area - an are which comprises Mirzapur, Bhadohi, Allahabad, Sonebhadra and Wyndomganj in the Garwa-Palamu region of Bihar. And exporters have been known to shift looms to areas where inspection is not very regular.
However, the weakest point of the anti-child-labour campaigns lies with the alternatives for children. Unfortunately, child labourers who have been rescued and sent back to their villages do not have viable alternatives for their betterment. Follow-up investigations have revealed that most have turned back to weaving either by going back to where they worked or getting looms set up in their villages. Until the 1970s, carpet weaving was largely concentrated around the areas of Bhadohi-Mirzapur, but as a result of actions taken against child labour it has now extended to other south eastern districts of Uttar Pradesh as well as parts of Bihar. The underdeveloped transport infrastructure of these areas makes it difficult to police the looms and the weavers. Education has been put forward as the key towards the elimination of child labour. And in fact, anti-child-labour NGOs have taken efforts to promote education. The schools they have set up have concentrated on the primary level. But beyond that level it becomes difficult for the children to continue formal education because the schools are situated far from their habitat. Moreover, many parents do want to educate their child but the infrastructure and the level does not equal their need to earn even the basic minimum livelihood. As Dreze and Gazdar (1997) say, "the opportunity cost of child labour is quite low even among very poor rural households. Parents do want their children to be educated and poverty as a limiting factor is highly over rated. The willingness of parents to bear the costs (school fees, clothes, slate and books) and to coax their children into going to school may depend crucially on the 'quality' of the schooling services they obtain in return."
The focus of the international campaigns has largely been on the carpet industry in India. Indian exporters point out that the situation in other carpet weaving countries may be as bad but not much information comes out, nor is there any intense reporting in the media. The perception in the Indian carpet industry is also that importing countries are playing favourites with some exporters and trying to form cartels through morality tags. Local exporters who have profit margins of around 10-15 per cent, point out that they are under pressure by the importers to keep their prices low. Levison et al. (1996) point out that American importers usually had margins of about 30 per cent on hand-knotted rugs while retailers usually had gross margins of about 60 per cent. According to them, "if the price of carpets in India rose by more than 15 per cent the importers would stop buying them from the country. In such cases the demand for child labour is effectively international, and action to discourage it needs to encompass all the major producers so as to avoid 'beggar thy neighbours' competition." As one exporter pointed out, China, Pakistan, Afganistan and Iran would not actually be winners in any Human Rights or dignity of labour competition.
In fact, banning this product from the international market because basic social standards are violated in the production process may not be a good short-term strategy as there would be an immediate negative repercussion in the form of wage loss for the weavers and their families. The importing countries should rather put pressure - through the ILO and the WTO - on the Indian government to reorganize the worker-owner relationship in the carpet industry, to improve law enforcement with regard to labour standards and to improve the access to schools and other basic infrastructure in the countryside.


Towards an Integrated Approach

Kaleen, Rugmark and others offer the consumers in the West a choice of morality tags on the carpets they buy. But these laudable efforts fall very short of the villagers' expectation of a decent living. For many Indian peasants, carpet weaving is considered a source of additional income to the low and uncertain agricultural earnings. There are hardly any comparable employment alternatives in the regions concerned. Therefore, the carpet industry should not be pushed out of business. However, the organisational structure should be changed and the abysmal working conditions along with the exploitative system of subcontracting should be brought to an end. According to Sen (1997), "there are positive initiatives to be taken for raising the human capabilities that make life worthwhile and which can also --given the appropriate economic climate--serve as the basis of fast and participatory economic growth."
The ILO has - naturally one would say - taken the Indian trade unions as strategic partners in its endeavour to eliminate child labour. But the trade unions in India have largely ignored the carpet industry. Their sphere of work largely remains in the urban industrial sector. The international organisations can fight child labour much more successfully if they place more emphasis on cooperation with the Panchayats (the local self-government units). The Panchayats may turn out to be the most effective implementing instrument at the grassroots level given the complexities of rural India (for details see Ashraf 1998). According to the 73rd constitutional amendment, administrative and financial powers are to be devolved to the Panchayats. They are now in charge of education, anti poverty programmes, health and sanitation, safe drinking water, as well as women and child development. That puts them in a position to respond to the issue of child labour at the village level where it is really needed. The Panchayats can be used to identify and survey the child labour situation in their respective villages and prepare a data base containing such information as the number of children in different age groups, school attendance, participation in the work force, socio-economic profiles.
Mobilisation of the community via the Panchayats lends credibility as the Panchayat members are elected representatives and the mobilisation begins at the micro level, initiating the bottom-up planning. In this way, the Panchayats could monitor and publicise violations of children's rights. To find a sustainable solution to the problem of child labour, an integrated participatory approach is needed which combines awareness, legal rights, enforcement, pressure by consumers, employment and education - programmes where the villagers are not told what is good for them. The Panchayats would constitute the appropriate gateway.


Ashraf, Shahid (1998), “The Elimination of Child Labour in the Carpet Belt of North India: Possible NGO and Trade Union Strategies”, in: Klaus Voll (ed.) “Against Child Labour: Indian and International Dimensions and Strategies”, pp. 221-227, Mosaic Books, New Delhi.

Ashraf, Shahid and Barge, Sandhya (1994), “Use of Child Labour in Carpet Industry Analysis from Economic Perspective of the Industry”, Centre for Operations Research and Training, Baroda, Sponsored by ILO, Geneva.

Dre‘ze, Jean & Sen, Amartya (1997), Indian Development: Selected Regional Perspectives, Oxford University Press, New Delhi.

Government of India, 1991, Census.

Haas, Daniel, “Social Clauses in International Trade: Instruments of a Global Social Policy” in: Voll, Klaus (ed.) (1999), “Against Child Labour: Indian and International Dimensions and Strategies”, pp. 11-30, Mosaic Books, New Delhi.

Joseph, M.P. and Ashraf, Shahid (1998), “National Consultation with Trade Unions for Developing Projects on the Elimination of Child Labour under IPEC in India”, IPEC-ILO, New Delhi.

Kruitjtbosch, Martine (1995), “A Brief Resume of Concept to Reality for Visual Guarantee of Carpets Made Without Child Labour”, South Asian Coalition on Child Servitude, New Delhi.

Levison, Deborah; Anker, Richard; Ashraf, Shahid and Barge, Sandhya (1996); “Is Child Labour Really Necessary in India’s Carpet Industry?”; Labour Market Papers 15, ILO, Geneva.

Ministry of Commerce, “Annual Report 1998-99”, Government of India.

National Council of Applied Economic Research (1994), “Child Labour in Carpet Industry - A Reappraisal”, New Delhi.

Voll, Klaus (ed.) (1999), “Against Child Labour - Indian and International Dimensions and Strategies”, Mosaic Books, New Delhi.

Weiner, M. Byron (1996), “Child Labour in India: Putting Primary Education on the Politica”l

“Agenda”, Economic and Political Weekly, November 9-16.

* I am grateful to Dr. Klaus Voll for the helpful discussions and to Ms. Vinoo Hora for comments on a draft of this paper. Usual academic caveat applies.

© Friedrich Ebert Stiftung | net edition malte.michel | 7/2001