Risk aversion, rural-urban wage differentiation and migration

In this paper I first present evidence to show that there is a labor shortage problem caused by insufficient rural-urban migration in Chinese urban/sub-urban areas. Moreover, people in poor rural areas migrate less than people in rich rural areas to cities. These phenomena cannot be explained by existing theories. A theory is provided to explain those puzzling phenomena. In my theory, migration is regarded as an instrument of the income portfolio if a household: facing high risks of food price fluctuation, a geographically extended cooperative household which has land and has out-migrants regards city jobs as high-risk high income opportunities, and regards agricultural production on its own land as low-risk low-income opportunities. In order to insure themselves, risk averse rural households would keep more labor input in agriculture (I.e. reduce the number of rural-urban migrants) compared with risk neutral households. Therefore, the "excess" labor input in agriculture and the resulting lower marginal labor productivity in agricultural production is virtually the payment for the insurance of rural households. If poorer households are more risk averse than rich households, then concerning their insurance, a poor and the rich is widened when there are opportunities for rural laborers to migrate to cities, and migration provides better chances to earn higher incomes