15 July 2003
Many international organisations have large programmes intended to reduce the prevalence of child labour. But are their intuitively plausible policies effective?
In societies where parents face a choice between sending a child to work or to school, the common presumption is that it is the constraints of poverty that drive children into work. If this is so, the appropriate policy would be to relax this constraint by, for example, offering cash transfers to very poor households. But if schools are costly, unavailable, or of poor quality, then, even when the household is not very poor, it may be rational to send the child to work. This is especially so for work such as farm work, which provides valuable experience if the child grows up to inherit the farm. This is the case where incentives dominate: the rewards to work exceed the rewards of schooling. In this case, policies that make schooling more attractive are likely to reduce child labour.
A third interesting possibility relates to parent altruism. Parents typically decide whether the child works, but their interests may not always coincide with the child’s interests. While economists have tended to assume altruism, historians and anthropologists studying child labour have questioned parents’ motives. If there were in fact only limited altruism then a case could be made for legislative action involving a ban on child labour or compulsory schooling laws. Overall, it is clear that research is needed to establish whether policy should primarily address poverty, inadequate schooling or the freedom that parents have to decide what their children do. To accomplish this, Dr Bhalotra uses economic theory to generate hypotheses which are tested using large representative household survey data that paint a fairly reliable ‘average’ picture for developing countries such as Pakistan and Ghana.
Is Child Work Necessary?
Bhalotra’s research investigates the hypothesis that household-level poverty compels child work. In this case, the child works to a target income defined as the shortfall between subsistence needs and adult income. A testable implication is that a drop in the child wage results in an increase in child hours of work and vice versa. Since economic theory would predict the converse (that people respond to a lower wage offer by working less), this is a good test of the role of ‘survival constraints’.
Attempting to curb the prevalence of child labour is an area in which policy has run ahead of research
To capture this, a theoretical model was constructed and applied to children in wage work drawn from 2,400 households in Pakistan. The results suggest that boys work when necessary for household survival but they are more ambiguous for girls, indicating that they work even when their income contribution is not critical. This may be due to parental favour for boys or to the perception that girls’ schooling is associated with lower returns.
The Wealth Paradox
A simple tabulation of data from rural areas of both Ghana and Pakistan shows, contrary to previous research, that children from land-rich households are more likely to work and less likely to attend school than are children from land-poor households. Bhalotra calls this the ‘wealth paradox’ as it seems to fly in the face of the popular presumption that child labour is less likely in wealthier households. The paradox is resolved by appeal to imperfections in the markets for land and labour.
Households that own more land do tend to generate more income which, by itself, does depress the extent to which they use the labour of their own children. However, where it is difficult to sell land and to hire labour, this effect may be overwhelmed by the fact that an extra hour is more productive on a larger plot of land, giving larger land-owners a greater incentive to employ child labour. The force of this incentive effect is investigated using the survey data from the very different environments of Ghana and Pakistan. The main result is that the ‘wealth paradox’ persists for girls, but for boys it vanishes once account is taken of variation in income and demographics. In other words, when all other observable influences on child labour are taken into account, in households with relatively large plots of land, girls are more likely to work rather than attend school.
Parent labour and child labour
It is easy to see that when parents work more, household income increases and children are less likely to work. This is called ‘the income effect’. But does the number of hours worked by the mother or father have any influence on child labour other than this income effect? Bhalotra found that indeed boys and girls were sheltered by the income effect. However, if the mother worked, daughters were also more likely to work, whatever the household income. In other words, the labour supply of girls is complementary to that of mothers. This is an important, if puzzling, result as it drives a wedge between two common policy objectives. Policy would like to encourage women in developing countries to work and gain economic independence, but this would appear to conflict with the objective of getting more girls into school. The precise mechanism involved merits further research.
Children from land-rich households are more likely to work and less likely to attend school than children from land-poor households
In a further development of the research on child labour, Bhalotra devised a test of parent altruism that can be applied to large-scale data to produce an ‘average tendency’. The basic idea is that if parents are altruistic, then expenditures on children increase as expenditures on adults increase. Intuitively, if we looked across households in a village and saw that in some households adults bought more clothes or tobacco for themselves, then, assuming altruism, we would expect children in these households to be more likely than others to attend school rather than work. Results for Pakistan decisively reject parent selfishness. However, at the same time they indicate that households spend less on children when at least one adult smokes systematically.
Overall, the findings of the different strands of this research are consistent with the view that boys work under poverty compulsions whereas the engagement of girls in work is probably best explained in terms of the rewards from work being perceived as greater than the rewards from school attendance. Therefore, cash transfers to households with working boys are likely to be effective, whereas to get girls out of work and into school would appear to require investments in improving access and quality of schools.
In Pakistan there is a remarkable gender gap in schooling, with only 31% of 10-14-year-old girls enrolled as compared with 73% of boys. Girls in this age group are twice as likely to be in waged work and a bit more likely to be working on the household farm, as compared with boys. In Ghana too, girls are relatively disadvantaged in schooling, although the gap is smaller. Clearly, therefore, policies targeted at closing the gender gap will substantially lower the overall incidence of child labour.