![Sample our Global Development journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5929/TOC-Subject-Sample-2021-Global-Development.jpg"})
The main purpose of this paper is to study market (outside the home) and non-market (domestic) child labour in rural India and see how this is influenced by a mother's participation in the labour force. The paper also investigates whether this participation has a different impact on sons as compared with daughters. The empirical analysis is based on household survey data from rural households in northern and eastern India. The results show that a mother's labour is not a substitute for, but a complement to, market and non-market child labour, while a mother's education, along with the father's education, reduces the likelihood of child labour. Gender-based analysis lends support to existing literature regarding the gender bias in domestic child labour. Additionally, a mother's participation in the labour force is found to increase the likelihood of daughters working outside the home as well. Thus, an increase in the opportunity for mothers to work in the labour-intensive agricultural sector makes child labour more likely. The results of this paper have important policy implications.
Notes
1 Analysis was also carried out using a continuous variable measuring the hours of child labour as the dependent variable and the hours of mother's labour as the explanatory variable. The results were very similar and are available on request.
2 Results using this as an explanatory variable produce very similar results to the other measures and are not presented or discussed here.
3 These and later figures in this section are rounded to the nearest whole number.
4 These correlations could be due to the endogenous nature of some of these variables, namely father's education and child labour and poverty and child labour. Given the limited availability of data, both of these variables are difficult to instrument for. Father's education enters in only one specification of the model. Given that the results hold up with and without this variable, we can assume that the endogeneity is not affecting the overall results. In order to deal with the possible endogeneity of the POOR variable, we ran IV estimations instrumenting for POOR using a variable that measures the proportion of poor in the village. This is a valid instrument given that it does not have an impact on the outcome. These results are not presented here but are available on request. The basic results remain robust to instrumenting for POOR.
5 First stage F-statistics confirm the validity of the instruments used.
6 These are not presented here.