For expository purposes, let’s stick with the extreme W. Arthur Lewis assumption of MPL=w=0 in the Chinese countryside by 1990. The list of contributing factors is long: electric pumps; mechanical threshers; chemical pesticides, herbicides and fertilizers; new cultivars; and finally (though post-dating 1990) new rice systems that did away with the need to prepare seedbeds and then transplant seedlings. Output increased, which in itself should have lowered the marginal productivity of labor [given unchanged labor supplies] while technology cut the number of labor days by over 50%. The Household Responsibility System allowed families to reallocate labor themselves (rather than waiting upon the decisions of the commune and regional hierarchies). So vegetable production, brick production and (where relevant) cotton rather than rice all rose. Farmers were better housed (local TVEs producing bricks), better dressed and better fed; real incomes and hence average incomes were up. So we have a great divergence between rising average incomes at the family level despite falling marginal productivity (and therefore implicity wage).
So again, assume w=0. In our class context of analyzing migration, that says that if a family member can earn ANY (net) income, then they will migrate. As long as daughter “Xiaojie” doesn’t outright starve – well, from the family’s perspective, it means one less mouth to feed, so there can be a not-so-subtle “push” involved. Quite realistically, even the remote prospect of an urban wage might be sufficient to offset initial moving and living expenses.
We thus find the Tondo in Manila, nasty barriors in Mexico City, horrific favela in Rio de Janeiro, and less familiar locations in New Delhi, Nairobi and elsewhere, teeming with the un- or informally-employed from the countryside. This matches the “push” perspective. (The seminal paper is Michael Todaro . “A Model for Labor Migration and Urban Unemployment in Less Developed Countries.” Amer Econ Rev 59:1, 138-148.)
But while we find migrant neighborhoods in Chinese cities, filled with people packed into tight quarters, there’s nothing approaching the urban cesspools common to other developing countries. Cities are not packed with beggars, the homeless and indigent children. That’s one more indication of something on the migration front that begs for explanation.