Why Rural Residents Do Not Migrate: The Hidden Welfare Costs of Rural-Urban Migration

A persistent puzzle in developing economies is why rural households remain in low-productivity agricultural sectors despite the substantial income gaps with non-agricultural opportunities. While existing studies attribute this gap to market frictions, institutional barriers, and differences in human capital, this paper shifts the focus to household-level welfare trade-offs, specifically, the non-pecuniary welfare losses borne by family members left behind when working-age individuals migrate. We develop a theoretical framework to show how such hidden costs affect labor reallocation and how they can be quantified empirically. Leveraging China's Grain for Green (GFG) Program--a nationwide conservation policy that induced farmland retirement in exchange for subsidies, we show that the policy led to significant increases in migration and non-agricultural labor, especially among women and younger individuals. Using revealed preference logic, we estimate that hidden migration costs amount to 10.5--12.6% of total household income for policy-induced migrants. Drawing on rich survey data, we trace these costs to two key sources: disruptions to children's education and reduced caregiving capacity for elderly household members. These findings highlight the need for policies that ease the burden of migrating with dependents, such as removing restrictions on education and healthcare access in destination areas.