Are working-age rural migrants to urban areas really better off? This paper uses data from the Panel Study of Income Dynamics from 1979 to 1997 to answer this question. It builds on literature by Fisher (2005, 2007) on the role of unmeasured characteristics in influencing rural residential choice and economic outcomes. Recursive bivariate probit models of migration and household poverty and two-stage least squares models of migration and household income are estimated for three periods: 1979 to 1985, 1985 to 1991, and 1991 to 1997. The models used in this study suggest that the relationship between rural out-migration and poverty is mixed, while there appears to be no discernable effect of rural out-migration on income in the short-run.