A child works at a brick kiln. Phan Hien for the ILO/Flickr. Creative Commons.

The first modern international convention specifically for the protection of children appears to have been the International Labour Organisation’s (ILO) Minimum Age (Industry) Convention, no. 5, of 1919. It applied internationally what had been a primarily European standard against child labour, a legal minimum age below which children were forbidden to be employed as industrial workers. It was widely believed by many educated Europeans that childhood everywhere should be privileged, devoted to play and school and separated from adult cares and responsibilities. Serious work was thought to harm children by robbing them of their essential innocence, an idea that lingers today.

Over time, the idea that children should be protected from work was increasingly applied to non-Western societies that had not yet industrialised. Many of these societies thought children should be prepared for adulthood rather than buffered against it, and considered children’s family and social ties to include obligations as well as privileges. Hewing to its global north ethnocentrism, the ILO gradually expanded the scope of prohibitions on child work until, in 1973, it passed the Minimum Age Convention (no. 138). This broadly sought “the effective abolition of child labour” and obligated ILO member countries to legislate age limits for which “no one under that age shall be admitted to employment or work in any occupation”.

Whereas the original 1919 goal was to keep kids out of undesirable workplaces like mines and factories, the expanded 1973 purpose was to discourage work more broadly. Economists and policy makers at the time believed, and many continue to believe today, that children are set to work by their parents on the basis of rational economic calculus. This core idea, which is based more on microeconomic theory than data, assumes that families below a certain level of wealth necessarily value additional income more highly than education. In order to fully safeguard children’s access to school, so the logic goes, the legal prohibition of child labour is required.

This thinking, along with the idea that a legal minimum age for work would keep children in school, represents 140 years of conventional wisdom now fossilised into doctrine. The assumptions of work-education conflict and of a need for minimum age policies to keep children in school were never tested by proper research. Nor was the effect of minimum-age policy on children evaluated. The rationale underlying the whole project was nothing more than group-think taken on faith.

Social scientists, child advocates and others first began to probe the realities of children’s work in the 1980s. Their findings, based largely on interviews with children and parents, did not easily jibe with either official pronouncements on child labour or justifications for universalised minimum age standards. We have learned that decisions regarding the place of work in children's lives respond to much more than economics. The initiative, or at least final decision, often rests with the children themselves. Furthermore, both families and children overwhelmingly value education and typically arrange for it in deciding children's work responsibilities.

 By the mid-1990s a clear gap had opened between the international politics of child labour as pursued by the ILO and its allied organisations on one side, and, on the other, empirical understanding by researchers of how children’s work, and interventions in it, affect the well-being and development of children. Based on the evidence, there is no reason to consider work inappropriate for even quite young children, as long as it is adapted to their age, ability and developmental needs. In most of the world, children are expected to help with family maintenance and livelihood tasks as a normal part of growing up. Their work is usually treated as an educational and developmental activity that prepares them for adulthood. Children reared without work responsibilities would, in many cultures, be considered deprived and their parents negligent. There is no evidence that child-rearing systems separating children from work are in any way superior to those in which children assume work responsibilities.

There is furthermore little conflict between work and education; most school-age children who work also attend school. As such, it is invalid to justify universal minimum age policies as a means of defending children’s access to education. When data are properly disaggregated to control for economic status and other key social and economic variables, there is little evidence that work reduces school attendance. In many cases work supports schooling, for example by providing money to pay fees, and is itself educational. The value of schooling is widely known and appreciated, and the demand for good schools is far greater than the supply. Except in cases of exceptional poverty, evasion of school in favour of full-time work is increasingly rare. The evidence suggests that most school evasion has more to do with school dysfunction than with competition from work. For similar reasons, banning work seldom promotes school attendance. Improving schools and making them more accessible and attractive, however, does raise attendance and achievement.

There is some indication that the most effective policies for protecting working children aim at keeping children out of seriously hazardous work. The ILO’s Worst Forms of Child Labour Convention (No. 182) takes this approach. Adopted in 1999, this convention specifies which kinds of work and working conditions are too dangerous, demeaning or exploitative for children. Many researchers and child advocates now suggest that the ILO retract the Minimum-Age Convention (no. 138) and replace it with this new convention as the international standard. While prohibiting children from work strictly on the basis of age is a bad idea, it is necessary to ensure the work they perform is safe and appropriate to their age. Policies and projects to do that must rest on empirical evidence rather than group-think.

This essay summarises findings and policy arguments more fully presented in Michael Bourdillon, William Myers, and Ben White (2009) ‘Re-assessing minimum-age standards for children’s work’, in International Journal of Sociology and Social Policy; and Michael Bourdillon, Deborah Levison, William Myers and Ben White (2010) Rights and Wrongs of Children’s Work from Rutgers University Press. These issues are complex and demand more lengthy explanations than are possible in the current format. The reader is referred to these publications for more complete information and discussion.