The likelihood that an adult child will reside with parents declines very sharply after age 18:In 1990,74% of 18-19 year olds lived with parents,compared to 40% at ages 20-24 and 16% at ages 25-29 (US Bureau of the Census 1992a).A substantial minority of the unmarried,however,continues to live at home into middle age.Three recent studies using different data sets asked what percentage of over-65 parents have coresident adult children; all arrived at figures in the range of 13-15% (Aquilino 1990,Ward et al 1992,Speare & Avery 1993).Some of these children are young adults who have not left home yet,others are temporary returnees,but some are 35 or 40 and home to stay.Although coresidence declines with child's and parent's age,it remains a significant phenomenon across the life course.
If we take a longer perspective,there is even less evidence that this generation is slow to establish residential independence.Historical evidence suggests substantial diversity in unmarried children's homeleaving behavior in the West:although many adolescents from rural and working-class backgrounds left home at very early ages to enter an apprenticeship or domestic service,others stayed home until relatively late marriage (Wall 1978).Like other dimensions of the transition to adulthood,the major historical change in home leaving is the extent to which it is now concentrated in a relatively narrow age range.In past centuries (and in Latin America today),children left home at any age between 10 and 30 (Wall 1978,DeVos 1989); in western Europe and North America today,the end of coresidence occurs rather abruptly between ages 18 and 21.
The larger story,and the one on which we have much less information,is change in parent-child obligation and its relationship to living arrangements.In the past,unmarried children who stayed home were expected to be net contributors to the household division of labor and to submit to their parents' authority (Mayer & Schwartz 1989).Those who lived away from home were more likely to send a portion of their wages home than to receive parental subsidy for their privacy and independence.In Caldwell's terms (1976),wealth and deference flowed up the generational ladder.The changes observed in household composition may obscure more dramatic changes in dependency and obligation between coresident parents and children.
The family life cycle perspective assumed a fixed set of developmental stages through which families moved from birth to death.In this perspective,the empty nest stage is inaugerated by the marriage of the lastborn child rather than by home leaving,a transition not treated separately.The family life cycle model has proven a useful device for organizing demographic data about average experiences of historical cohorts,but has been criticized for its failure to incorporate the experience of those who divorce or who never marry.As a result,many scholars have turned to the life course perspective,a more flexible model that emphasizes sociohistorical variability and focuses on transitions rather than stages,a wider variety of destinations,more variability in pathways,and the possibility of countentransitions (Hohn & Mackensen 1989).It is still,however,very much a structural and normative framework.According to Hagestad & Neugarten (1985:36),the life course perspective "concentrates on age-related transitions that are socially created,socially recognized,and shared" (p.35,emphasis in original).Thus,changes in social norms and in the transition to adulthood are considered here as part of the lifecourse perspective.