Applied Managerial Economics -GRBU513
DU, Palumbo·Donahue School of Business, Spring 2014
Lu Zhang
Child Labor

INTRODUCTION
Child labor has always been a world trend topic. No matter what punishment our governments prepare for the law-breakers, what policy our governments conduct to help to reduce the rate of child labor, how many times people say forcing children to work is cast aside by them, how many campaigns people launch to show child labor is unacceptable, it just keep happening in any corner of most developing country. Why such thing cannot stop? Some people may think: because the factory or labor agencies are all run by some bad, irresponsible people. But let’s think this through, you know some areas are just not that well-developed to allow all the eligible children to go to school, government should do something, but it will take a really long time to achieve their goals. And I think there are more considerable factors.
Gender bias
Previous works on son-preference have studied its effects on different access to health and education (Alderman and King, 1998, Behrman et al., 1986, Davis and Zhang, 1995 and Orazem and King, 2007). Empirical evidence also suggests that both incidence and the intensity of child labor are higher for female children than male children. For example, Edmonds and Pavcnik (2005) using the UNICEF MICS (Multiple Indicator Cluster Survey) data find that the incidence of child labor among female children (72.1%) is much higher compared to male children (64.8%). In Fig 1, we can also find that female children are more likely to work long hours than male children. Some companies also say that they would like to hire girls more because girls are easier to control and better-behaved.

Fig 1. Participation rates in various activities
Another reason is that the effects of gender bias depend on both its form as well as economic conditions of parents. In the case of son-preference, when parents are relatively better-off and can give bequests, both male and female children receive an equal amount of schooling. However, female children work more than male children. In the case of an earnings function bias, male children not only receive more schooling than female children, but also work less.
When parents are so poor that they cannot give bequests, child labor is inefficiently high and schooling is inefficiently low for children who do not receive bequests. Son-preference interacting with poverty can lead to less investment in the human capital of female children relative to male children. Social norms such as dowry and marriage expenses can lead to inefficiently low level of human capital investment and high level of child labor. Son-preference interacting with dowry and marriage expenses can aggravate gender differential in the human capital investment (Kumar, 2013).
About the work
Both boys and girls are reported to have lower participation in agriculture and more in domestic duties using the short module. The marginal result of the allocation across three categories: agriculture and other sectors, domestic work, and the category “no work” by presenting the regression analysis result.
Table 1. Regression analysis of main activity

Change to another angle to see the allocation across the three categories. Let’s see the predicted distribution in Table 2. From Table 2 we can see the actual number as well.
Table 2. Predicted distribution across employment

Both girls and boys are more likely to be classified as working in domestic work than identified as not working when given a short module. This effect is also larger for girls than boys. The proxy module is not associated with significant changes in sector classification.
The multivariate analysis confirms that the largest difference between the short and the detailed modules is in the allocation of children across the two categories “domestic work” and “no work”. Although the detailed module captures higher