G K LIETEN

S

Economic and Political Weekly January 14, 2006

of the World’s Children, took a balanced

and realistic position: In reality, children do a variety of work in widely divergent conditions. The work takes place along a continuum. At one end of the continuum, the work is beneficial, promoting or enhancing a child’s physical, mental, spiritual, moral or social development without interfering with schooling, recreation and rest. On the other end, it is palpably destructive or exploitative. There are vast areas of activity between these two poles, including work that need not impact negatively on the child’s development. …But to treat all work by children as equally unacceptable is to confuse and trivialise the issue and to make it more difficult to end abuses. This is why it is important to distinguish between beneficial and intolerable work and to recognise that much child labour falls in the grey area between these two extremes [UNICEF 1997: 24].

Child labour is thus generally defined as work that deprives children of their childhood, potential and dignity, and which is harmful to their physical and mental development. But at the same time, there is a vast grey area. This position was accepted by the ILO as well. It was agreed that it was not possible to give “a precise dictionary definition” applicable to all situations and all countries: “Whether or not particular forms of work can be called child labour depends on the child’s age, the types of work performed, the conditions under which it is performed and the objectives pursued by individual countries. The answer varies from country to country, as well as among sectors within countries” [ILO 2002a: 16].

In 1999, ILO convention 182 was accepted. It was agreed in convention 182 that not all work by children was child labour and that not all child labour could be done away with instantaneously. ILO 182 has two categories of intolerable forms of labour: the unconditional worst forms (categories 1 to 4: slave labour, prostitution, participants in armed conflicts, illicit traders, totalling 8 million) and the hazardous forms, which comprise work that exposes children to danger and jeopardises their physical and moral health.

One would have expected that the breakthrough achieved by convention 182 over 138 – the separation of fairly harmless forms of child labour from the harmful – would have led to new studies separating both and trying to estimate the real size of the problem. One then would have expected that the sheer size of the problem would have come down drastically.

Such was the intention of convention

182. Both convention 182 and recommendation 190 were based on the premise that the child labour problem was a huge one and that it would be realistic in the short term to focus on the more abject forms of child labour. Child labour in the age category 5-14 in 2000 was estimated to be 186 million; 66 per cent of this were classified as working under hazardous forms of child labour [ILO 2002b: 20].1 If despite this, the size of child labour generally and the magnitude of the worst forms of child labour are not of a substantially different order, something must be wrong either with the assumptions or with the counting, or with both.

The new UNICEF report (2005) referred to earlier states that literally tens of millions of children around the world work today long hours before they have even reached the age of 10 and that “1 in 12 children in the world was reckoned to be involved in work which put their health at risk or caused serious harm” [UNICEF 2005:3].

Could that be the case? Probably the figures appear substantially lower if we go by the qualifications elaborated in recommendation 190. Recommendation 190 recommended that any definition of hazardous work should include: (i) work which exposes children to physical, psychological or sexual abuse; (ii) work underground, underwater, at dangerous heights or in confined spaces; (iii) work with dangerous machinery, equipment and tools or carrying heavy loads; (iv) exposure to hazardous substances, agents or processes, or to temperatures, noise levels or vibrations damaging to health; and (v) work for long hours, night work, and unreasonable confinement to the premises of the employer.

Would the total number of child labourers working in such conditions be as high as 180 million? The recent UNICEF statistics made international headlines, but at no stage in the report was it made clear how the calculations of 1 in 12 children as hazardous labour had been made and that the figures were rather a rehash of old statistics.

In fact, the calculations could not have been made. The hazardous forms of child labour, many years after the adoption of

Studies

Technical meetings are being organised by the ILO and studies based on a Rapid Assessment Methodology have been conducted in 19 countries, mostly funded by USDOL, but many of these studies have been on unconditional forms of child labour. On hazardous forms, case studies are available on child domestic work (five countries), commercial agriculture (five countries), street children, garbage dump scavenging and urban informal sector (two countries each), and fishing, mining, portering and ragpicking (one country each). As many as 15 of the 24 studies have been conducted in three countries (Nepal, Tanzania, El Salvador).2

It may be difficult to extrapolate on the basis of these data for the magnitude of hazardous child labour in the world. It is not necessary to span the entire world in order to get a reliable picture, but the present case study information is obviously not enough. Hardly any studies have been conducted on sectors where most of child labour is concentrated, for example agriculture and the informal sector. The need for new investigations into the extent of child labour is clear.

The studies that have been conducted also do not unequivocally establish that hazardous child labour is involved. The study on child domestics in Bangkok for example (a study based on 115 samples under the age of 18) concludes that “their circumstances are not a priority for action in the context of worst forms of child labour” [IPEC 2002a: 16]. The conclusion of the Sri Lanka study on the other hand is unambiguous that young children were recruited for domestic work and engage in a variety of activities including those considered to be hazardous. The information gathered shows that “child domestic workers below 18 years of age3 are subject to physical and/or emotional abuse and may also be vulnerable to sexual abuse” [IPEC 2002b:96].

A similar difference can be observed between the study of sugar cane cutting in El Salvador and in Bolivia. In El Salvador, most of the children (92.7 per cent) worked near their homes for an average of five hours a day during two or three months which coincided with the school holiday season. In general, they worked together with their parents. Once the harvest is over, two-thirds of these children go to school and 22 per cent of the children work in other remunerated activities. The children were reported to be suffering during their workdays due to the hot sun, the use of sharp tools, the exposure to insects and uncomfortable positions and excessive loads [IPEC 2002c]. Conditions in the sugar cane fields of Bolivia appear to cause more concern. In the two research areas, 90 per cent and 55 per cent respectively of the “cuartas” (the children helping their parents) did not attend school. Half of the families for the duration of the cane cutting seasons migrate from the poverty-stricken high-altitude regions around Potosi and for a period ranging between five and six months stay in makeshift encampments; for these children, education is not an option. They are living under primitive conditions in highly overcrowded encampments without access to basic services, where the working day last 12 hours, not including travel time [IPEC 2002d]. While working in the cane fields in El Salvador is child labour, no doubt, it is possible that it does not constitute that of the worst kind; child labour in the cane fields in Bolivia on the other hand, obviously is. Anyway, the national law in Bolivia prohibits children under the age of 18 from working in the sugar cane harvest.

Importance of Statistics

In order to get an indicative estimate of the size of hazardous labour, many more sectoral studies are needed. What we have up to now does not seem to provide firm ground for extrapolation and it remains difficult for the ILO and its tripartite partners to develop instruments for delineating hazardous child labour in each country.

In the past, the overall figures that have been flaunted have been useful for advocacy purposes. The higher the numbers and the more miserable the circumstances, the more one could expect the world community and the national governments to rise to the occasion. But for policy purposes, a more realistic account would now be welcome.

Of all child labour in the world, 60 per cent is stated to occur in Asia, 23 per cent in sub-Sahara Africa, 8 per cent in west Asia and Latin America and 6 per cent in North Africa. In sub-Sahara Africa, 29 per cent of children are active as child labourers: the participation rate is 19 per cent in Asia, 16 per cent in Latin America, 15 per cent in North Africa and only 2 per cent in developed countries.

Economic and Political Weekly January 14, 2006 not all child work is child labour. The distinction between these words cannot be made in all languages,4 but that should not be the end of the discussion. Work done by children also includes, all activities that are undertaken as part of the normal process of socialisation, in the household, on the farm, in housework tasks after school and even by undertaking a small activity in order to earn some extra pocket money. There are activities that children in, for example, Europe are already allowed to engage in at the age of 13. “Out of school” work increasingly has come to be viewed as a healthy pastime and an embodiment of the work ethic.

These are also activities that ILO convention 138 allows for children after the age of 12 in developing countries, as long as it is only some light form of work and does not extend beyond two hours a day. Convention 138 actually goes much further in excepting certain categories of children. In respect of agricultural work (in article 5), it excludes “family and small-scale holdings producing for local consumption and not regularly employing hired workers”. The ILO was not opposed to all types of work since child work takes many forms and “in some case, such as traditional agricultural or handicraft production, it is carried out under the supervision of parents. Work of this type is often an integral part of the socialisation process” [ILO 186: 14]. UNICEF (1986: 17), in an interesting document on children in especially difficult circumstances, approvingly quoted the ILO director in his 1983 annual report on child labour:

When work by children is truly part of the socialisation process and a means of transmitting skills from parent to child, it is hardly meaningful to call it child labour. Nor can such work be divorced from poverty and underdevelopment and the absence of alternatives to child work which together generate and sustain it.

Going though the statistics provided by SIMPOC, one has a tough job finding such subtle and yet basic distinctions. The data that have been collected, and that constitute the basis for the extensive and influential work done by World Bank-associated economists, are not readily useful for an analysis of the real problem of hard core child labour.5 One even gets the impression that all forms of work are being counted as child labour and that thus the overall figure on child labour is beyond the mark.

The elaborate report on Tanzania, for example leaves crucial questions unanswered [IPEC Tanzania 2002; see Lieten 2001, 2005]. The figures cannot be directly transferred to a transparent matrix. For one reason, the age categories (5 to 9 and 10 to 14) do not correspond with the ILO conventions 138 and 182.6 Without further transparency in the reporting, it is impossible to know how many of the children have been classified as child labourers and how many actually are child labourers. In the five-nine age group, 25.5 per cent (!) are said to be economically active, but we do not know how many hours they work and whether they go to school. We know that two-thirds of all the children work less than four hours a day, but we do not know how this applies to the different age categories. Many children in the 10-11 age group may have been included as child labourers because, even though they went to school, they also did “some work” in the reference week. Economic activities, by the way, include activities like fetching water and firewood, and it seems that many young children, even if doing only “some work” for one or two hours a day, have been promoted to the status of child labourers and even to the status of working in the worst forms of work.

It is quite possible that the statistical division of SIMPOC has applied formulae in such a way that the extrapolation figures on child labour in the world (210 million) are good approximations, but for lack of transparency it remains unclear. Elsewhere, I have suggested on the basis of statistical reports about Nepal and Pakistan that the outcomes must be approached with the necessary scepticism.7 One can question whether 13-14 years old children that engage in light work at home should be considered child labourers. In many data gathering exercises, they erroneously may have been included, thereby unduly increasing the number of assumed child labourers in the world.8

The more reliable figures, in my view, are lower. I do not, with this suggestion, want to diminish the enormity and the intolerability of the problem. I simply want to make a plea for realistic perceptions and in particular for isolating those forms of child labour that are intolerable.

Fluffing up numbers and painting as dramatic a picture as possible of the circumstances in which children have to work, generalising the worst forms, so that they apply to all working children, has led to more money in collection drives as well as to greater financing by

Economic and Political Weekly January 14, 2006

institutional funds. The figures have also been handy to wake and shake up the world conscience, but they are not suitable for drawing up good policy. Since the world has already been sensitised to the child labour problem, the advocacy statistics should be replaced by realistic statistics.

Counting Child Labour in India

I will illustrate this statement by way of the situation in India and about the publications detailing the extent of child labour in this country. Since India is supposed to be responsible for one-third of the child labour incidence of the world, statistics relating to India do matter a great deal. However, the numbers regarding child labour in India differ widely. A number of NGOs as well as most western sources state that India has more than 100 million child labourers, which is to say that about half of all children between the age of six and 14 engage in child labour. According to government figures on the other hand, the number has decreased over the past 20 years, from approximately 21 million in around 1980 to 9 million in the year 2000, with a child workforce participation rate that has decreased from 11.2 per cent to 4.8 per cent. Around four out of five of these children are “working” in agriculture for at least one hour a day.

The difference between the estimates of 100 million and that of 9 million has everything to do with problem of definition. In India around 80 million children who have not been counted in the government child labour statistics do not go to school. They do not go to school because they are too young (poor children in rural areas enrol quite late, because they have finished the village school) or because they just do not go or have dropped out. Because these children cannot be found amongst the statistics of working children nor amongst the statistics of schoolgoing children, they have come to be referred to as “nowhere children”. In the high-incidence scenario, all these children have been included as child labourers.

Children who do not go to school are not necessarily child labourers. There is a large category of deprived children and since the reasons for not attending school are different from those of children who have to work, it is of critical importance that the different categories and mechanisms are separated. Otherwise one shall run the risk of getting the deprived children into school and assuming that the problem of the core category of child labourers is being solved: it is possible to solve the problem of nowhere children without even touching the core problem of child labour.

With some audacity I isolate the complexities involved in the definition of child labour and of worst forms of child labour as the essence of the problem. If we know what the concept means, where the phenomenon begins and ends, it is possible to act against it and to formulate a suitable policy response.

Conclusion

The abolition of child labour in developed countries has taken one century to bear fruition. In most developing countries a change for the better has taken place in a relatively short period of time. Governments have become aware of the importance of resolving the social question with an eye to economic development, national integration and social development in general. Policies have been put in place and there is considerable development in, for example, increasing literacy, reducing child mortality and limiting birth rates. Laws have been adopted and international treaties and conventions have been ratified. An increasing number of people seem to be sufficiently aware that labour is a bad option for childhood and large numbers of organisations and movements, national and international, are actively paving the road for new policies and new practice.

It nevertheless seems that efforts to abolish child labour are faltering and that, in combination with the commitment to reach the millennium development goals on universal education before 2015, efforts shall have to be undertaken on a warfooting.

The ILO, with the choice of convention 182 on the Worst Forms of Child Labour, has taken an important step by making the distinction between “not good but somewhat acceptable” and the “absolutely unacceptable”: “Not all work performed by children is equivalent to “child labour”…The problem is how to draw a (statistical) line between acceptable forms of work by children (which may be regarded as positive) one the one hand, and child labour that needs to be eliminated on the other [ILO 2002b: see also ILO 1998].

That job, it seems to me still remains to be done. International organisations, like the ILO and UNICEF, have brought together a mass of rich and varied data, and deserve full credit for it, but yet, at the same time, in view of their official character – working with governments as counterparts

– there is a need for independent research in this matter so as to provide more supporting evidence on where intervention is most urgently needed.

EPW

Email: g.c.m.lieten@uva.nl

Economic and Political Weekly January 14, 2006

figures include all children from 5 to 17, andthis aggregation is not useful to isolate the“child labour” category. Of the children engagedin economic activities, 2.5 million were attending school and 2.2 million were notattending school. Of these 4.7 million children,

3.8 million were working on the shamba (familyfarm), 0.9 million were working in other familyoccupations, 79 thousand were in paid employment and 60 thousand were self-employedoutside the family occupation.

7 An ILO investigation [IPEC 1996: 15] cameto the conclusion that 22.5 per cent of boys and

7.2 per cent of girls between 10 and 14 yearsof age worked, that is to say, “at least one hour(sic) every day of the week preceding theinvestigation”. In Europe, this would not qualifyas child labour. The majority worked as unpaidassistants on farms. In the cities, only 10.1 percent of the boys and 1.5 per cent of the girlsin the same age group worked (at least one houra day!). Next to the question as to whether thisactually entails “child labour”, one can questionthe quality of the material gathered: where inthe NWFP-region, 39.0 per cent of the boys inthe rural areas (10-14 years) was recorded aschild labourers, the percentage in theneighbouring province of Baluchistan was only

2.4 per cent, which could suggest that theresearch contract with local research institutes did not produce very reliable data.

8 The relevant clause in ILO convention 138, drawn up in 1973, established a minimum ageof 15 years for child labour, with exceptionsof 14 years (in countries “where the economyin educational institutes was insufficientlydeveloped”) and at 13 years for light work orat 12 years in countries where a general minimumof 14 was applicable.

Anker, Richard (2000): ‘The Economics of ChildLabour: A Framework for Measurement’, International Labour Review, Volume 139, No 3, pp 257-80.

Basu, K and P H Van (1998): ‘The Economicsof Child Labour’,American Economic Review, Volume 88, No 3, pp 412-47.

ILO (1983): Child Labour: Extract from the Reportof the Director-General to the International Labour Conference, 69th Session,International Labour Office, Geneva.

– (2002a): ‘A Future Without Child Labour’,Global Report under the Follow-up to the ILODeclaration on Fundamental Principles andRights of Work: International LabourConference 90th Session, report I(B),International Labour Office, Geneva.

IPEC (1996): ‘Summary Results of Child LabourSurvey in Pakistan’, FBS and Ministry ofLabour, ILO, Geneva, International Programmeon Elimination of Child Labour, Pakistan, mimeo.

Journal of Labour Economics, No 2, pp 29-46.

Skeldon, Ronald (2000): ‘Towards IntegratingResearch Methods on Invisible Forms of Child Labour’, A Desk Review, RWC-CL, Bangkok.

UNICEF (1986): Overview: Children in EspeciallyDifficult Circumstances, UN Children’s Fund, Economic and Social Council, Executive Board, Policy Review (E/ICEF/1986/L.6),New York.

Economic and Political Weekly January 14, 2006