The “new economics of labor migration” has added explanatory power to the neo-classical model by focusing on a household’s decision to send migrants in a context where migration serves to mitigate the impact of insurance and market imperfections on “emitting” households. The realization of this idea can be traced back to Stark and Bloom’s 1985 paper.
The old Harris-Todaro model of migration is the traditional, and more individualized, model of migration. Contemporary research on the determinants of migration, however, has focused on the importance of economic and noneconomic factors in the decision to migrate in a “dualistic” economy, that is, one in which there rigid wages in the urban sector and flexible wages in the rural sector.
This “cumulative causation” of migration provides a framework for understanding migration by looking at individuals and households.
In the individual model model of migration, wage flexibility in rural areas effects decisions to migrate from rural to urban areas, where rigid wages are often higher. Urban utility may be higher due to unionization, proximity to policymakers, and thus more likely to have minimum wages, unemployment benefits, day cares, and pension schemes. And employers will often pay higher wages to “buy the threat” from loss of productivity caused by competing low-wage employment. According to the Harris-Todaro Model, an equilibrium obtains when the expected urban wage is equal to the marginal product of a rural worker.
Household models of migration, on the other hand, consider several noneconomic factors that influence the decision to migrate. Migration networks, for example, serve as a means of conveying information from those with migration experience to potential migrants, and network members assist new migrants, and therefore networks serve to influence the expected income gains from, and the uncertainty associated with, migration. There is also a distinction to be made between community and family networks. Family networks are considered “strong ties” and community networks “weak ties” for various reasons.
The household model therefore adds sophisticated variables like information, insurance, and social capital to the model. This is more difficult for the neo-classical model to absorb, however. It doesn’t mean it cannot be done. Presumably everything can be reduced to mathematical formalisms, right? Perhaps, but models do have obvious limitations. Even the household model has its own. It opens the door to various other migration factors, such gender and totalitarian biases in the household itself, and the dynamics of labor- and wealth-endowments of households.