Why Families Move

A model of migration in which the family rather than the individual is the decisionmaking unit is tested using longitudinal data on households from the University of Michigan's Panel Study of Income Dynamics. The study employs several methodological features new to migration research, such as imputing wages that could be earned in alternative destinations and explaining family-level choice of destination. Principal findings are that (1) heads of households who are unemployed or looking for another job are more likely to move than those satisfactorily employed, (2) high local unemployment rates prompt the unemployed to migrate, (3) families are more likely to move again if they have moved in the recent past, and they are most likely to return to places where they have lived previously, (4) families who have moved several times are more likely to move again than are families who made one or no recent moves if the last of those multiple moves was not to a place lived in previously, and (5) wives appear to have a significant influence on the family's decision to move and where, and families with working wives are not necessarily less likely to move than are families with nonworking wives. Important policy implications are discussed.