Child Labor and Household Land Holding: Theory and Empirical Evidence from Zimbabwe

The relationship between rural household productive assets and child labor in developing countries is complex. Some empirical evidence shows that child labor tends to increase as land holding increases, or there is an inverted U-shaped curve relationship between the probability of putting children to work and land holding. This paper shows that the relationship between use of children as laborers and land holding is nuanced. Child labor generally decreases as per capita land holding increases, but there can be an upward bump in the relationship between child labor and landholding near the middle of the range of land per capita. The bump can be explained theoretically by the relationship between the marginal productivity of a child worker on the farm and the marginal value placed on his/her education at different levels of wealth. This pattern is repeated in three surveys conducted in Zimbabwe, in 2001, 2007–8, and 2010–11. From the perspective of policy making, the policy maker should be alerted that the programs to promote school retention should not necessarily focus only on the poorest households in rural areas. There is a high probability that middle-wealth households put their children to work, and this probability may change by some other factors such as gender of child and agro-ecological conditions.