Working Paper No. 05-38
Literature on migration and network effects suggests that the rate of migration is positively related to the extent or degree of personal and community level networks potential migrants have at the destination. However in this particular paper it is shown that when the decision making unit is the extended family and there is a minimum wage induced Harris-Todaro type job rationing in the urban sector having greater numbers of previous migrants at the urban end does not necessarily lead to more migration from the hinterland. This counter intuitive result is generated as a consequence of juxtaposing an extended family framework with urban equilibrium unemployment in the model. A larger stock of previous migrants at the urban end has a positive effect on new migration which comes specifically through a greater flow of remittance income from the migrants to their rural counterparts - the two households comprising the extended family. The increased remittance income provides a positive stimulus to migration as it relaxes the migration cost constraints facing the extended family. On the other hand limited jobs in the urban sector and the resultant job rationing implies that a greater number of previous migrants also crowds out job opportunities for the new ones thus simultaneously reducing incentives to migrate. The direction of the net effect however depends on the economic characteristics of the extended family and the initial employment conditions in the urban sector.