Family Instability... It's Not Just Mom and Dad

by Kristin Perkins
Postdoctoral Fellow

Children experience many changes in their households while they are growing up. But while we often think about divorcing and remarrying parents as common changes in household composition that affect many children, in new research, I show that changes involving extended family members and nonrelatives are far more common than changes involving a parent. This finding is significant because prior research suggests that it is likely that instability involving nonparental household members affects children's outcomes. It is therefore relevant to assess the extent to which children are exposed to these transitions and how exposure varies by race and family structure.

To gauge the extent of household changes, I used the nationally representative Survey of Income and Program Participation (SIPP) to track a sample of more than 72,000 children and their households over approximately two years. The SIPP interviews households every four months and documents the set of household members at each interview. This allowed me to identify the relationship between each child and each other household member and to determine who exited or entered the households between interviews.

Overall, by the end of two years (after six interviews), about one percent of children experienced a change in household composition involving their mother and five percent experience a change involving their father. In contrast, more than 10 percent experienced a change involving a grandparent, aunt, uncle, cousin, or other extended family member. In addition, more than four percent of children experienced a change involving a nonrelative. This means that if we think only about divorcing and remarrying parents, we miss changes that affect 14 percent of children over a period of about two years.

These rates vary significantly by type of household, by race, and by ethnicity. For example, 18 percent of children living with a single parent and 29 percent of children living with no parents had an extended family member enter or exit their households compared to only seven percent of children who live with two parents. Similarly, over 10 percent of children living with a single parent or no parent experienced a change involving a nonrelative compared to two percent of children living with two parents (Figure 1).

Moreover, 17 percent of black children and 18 percent of Hispanic children experienced a change in household composition involving extended family members compared to only six percent of white children. There is not as much difference by race and ethnicity, however, for changes involving nonrelatives. Rather, the rate clustered around five percent for all three groups (Figure 2).

Taken together, these findings mean that considering only instability involving parents may not uncover differences by family structure, race, and ethnicity that a broader conceptualization of household instability would reveal. This is important because many studies find that divorce has negative effects on children's well-being. My research showing the much more widespread exposure to changes in household composition highlights the need for future research that assesses whether these other changes in household composition are detrimental—or beneficial—for children. For example: do the distraction and stress from instability involving relatives and nonfamily members mean children perform less well in school or have more behavior problems? Or do some types of changes reflect closer relationships with extended kin that are beneficial for children? Such questions, which were beyond the scope of my analysis, clearly merit further attention.

These descriptive findings and future work on the consequences of household instability for children could also have implications for housing policy. If high housing cost burdens and a lack of affordable options contribute to changes in household composition and if those changes are detrimental to children, then expanding the supply of affordable housing and targeting it to specific households and/or specific high-cost or low-income geographies could greatly help those children. Moreover, such policies could have meaningful spillover effects because providing housing assistance to families not only might benefit children in the family that receives the assistance, but could also help children in the households their families otherwise would have joined when they doubled up with extended family members or nonrelatives.