Volume 11 Issue 1 - February 2010
Q&A: Proximity and Coresidence Among American Families
The University of Michigan Retirement Research Center interviewed researchers Robert Pollak and Janice Compton to discuss the findings from their 2009 Working Paper, titled “Proximity and Coresidence of Adult Children and their Parents: Description and Correlates."
MRRC: Your study finds that most Americans live within 25 miles of their mothers. Were you surprised at those results?
RP: I think there’s a misperception about how spread out families are geographically and I think this provides a clearer picture. The best predictor of proximity of adult children to their mothers is education. Adult children with college degrees are much less likely to live with or near their mothers.
JC: What’s most striking about this research, in my view, is shown in the accompanying table: 50% of couples where neither of them have a college degree live close to both mothers and 50% of couples where both have a college degree live far away. It’s almost a mirror split. The numbers are almost exactly reversed.
Couples Near Mother by Education: http://www.mrrc.isr.umich.edu/?id=243
MRRC: Are there any differences with regard to who lives closer to their parents by birth order or the age of the child?
RP: We found no influence of the birth order on how close adult children live to their parents. In general, we found that younger adult children are more likely, and older adult children are less likely, to live with or near their mothers. Also, we also found that only children are more likely to live closer to and with their mothers.
MRRC: Why do you focus on mothers?
RP: If both parents are alive and living together, then the mother represents both parents. If both parents are alive and living apart, we focus on proximity to mothers rather than fathers, because mothers are more likely than fathers to provide and receive hands-on care. Grandmothers are more likely than grandfathers to provide care for grandchildren, and elderly mothers are more likely than elderly fathers to receive long-term care from adult children.
MRRC: What is the percentage of households where the parents no longer live together?
JC: According to the National Survey of Family Health, approximately 11.6% of mothers are not living with the child’s father. Of this 11.6%, approximately 40% (5% of the total sample) are remarried.
MRRC: Does greater proximity of adult children and their parents indicate that those families who live closer spend more time helping each other than those who live farther away?
JC: Most definitely. You can see a fall-off in helping each other between families that live 5 miles from their mother to those who live 10 miles away. This includes such things as shopping, errands, transportation, housework, yard work, car repairs and other help around the house. The closer you live to your mother, the more likely you are to transfer help to her and the more likely you are to receive help.
MRRC: Does this fall-off in support with distance also apply to emotional support?
JC: Emotional support does not fall off as steeply with distance. But, you do also see more emotional support between those who live closer.
RP: The most significant activities in time transfers are child care, on the one hand, and care of the disabled elderly on the other, which is what you would expect. And that kind of hands-on care depends on proximity or coresidence.
Families that are more spread out geographically are also more likely to have more education, more income, more resources, and in that sense, almost certainly more access to market substitutes for family provided care.
Americans live closer to their families than academics might have imagined, looking around at themselves and their colleagues. The explanation for that is that academics are atypical because they have a tremendous amount of education and that’s correlated with proximity in a negative way. Higher education levels mean that you tend to live farther away from your family.
MRRC: Your paper states that couples who have more education are less likely to have mothers who live close to each other. Could you say a little bit more about that?
JC: If you go to college, you’re more likely to marry someone who’s not from your hometown and not from your state. Power couples – where both people have a college degree – are less likely to have been born in the same state than couples with less education. It’s a different decision altogether for couples whose mothers are from different states about caring for their family and whom to live near.
RP: Mothers may be a proxy for networks of other family members and friends. You shouldn’t be misled into thinking that it’s always mothers who are the only factor driving this. The location of mothers is going to be correlated with uncles, aunts, cousins, and with friends, to the extent that we’re picking up where people went to high school. So there are other kinds of networks involved.
MRRC: How close do single people tend live to their mothers?
RP: Adult children over 25 who are single live closer to their parents than married couples. The median distance is 15 miles for singles and between 20 and 30 miles for married couples depending on the characteristics of the couple.
MRRC: If adult children are living with their mother or parents, does that usually mean that the elderly are disabled?
JC: Looking at it from the point of view of the adult child, if the adult woman is married, then she might be more likely to live with her mother if her mother is old, widowed, and in need of care. But it’s a very small percentage who actually live with and take care of their mother.
Gender affects coresidence differently by marital status. Single men are more likely to live with their mothers compared to single women, while married men are less likely to live with their mothers compared to married women. When these groups are separated, we find that single men are less likely to live with mothers who are in need of care, while married women are more likely to live with mothers who need care.
We don’t get into the connection between care and coresidence directly in this paper. However, the evidence shows that elder care isn’t coming through coresidence as much as it used to.
MRRC: One might think that American families are not very cohesive since they don’t live together as much as they did in the past, but the proximity data tells a different story.
JC: Now remember that we’re looking at it from the viewpoint of the adult children. So if most adult children live within 25 miles of their mother, from the standpoint of the elderly mother, the chances that she has at least one child nearby are higher.
MRRC: Are single adult men actually living in their parents’ homes?
RP: What’s striking about the data is how little coresidence there is of married couples with either his mother or her mother. However, almost 20% of single men coreside with mom. If you exclude coresidence, median distance of single men to their mother’s home is 15 miles. It would be interesting to know how many of the single men have never left home. Which puts a different twist on the question of who’s taking care of whom.
MRRC: When adult children take care of their elderly parents, whose house are they living in?
JC: If you’re living with your child and you’re not the head of the household, then you might well be disabled. Your children live with you when you’re healthy, but you live with your children when you’re not healthy. Again, these are very small percentages.
RP: The dataset that can address that question is the Health and Retirement Study. And that study shows that, predominantly, when there is coresidence, the adult children own the house rather than the disabled elderly parent.
MRRC: You also found that compared with whites, Hispanics are no more likely to live in close proximity to their mothers, but twice as likely to live with their mothers. And Blacks are more likely than whites to live near their mothers and more likely to live with their mothers. Why do you think this is so?
JC: There are certainly cultural issues surrounding coresidence and family proximity that we do not comment on in the paper. For both Hispanics and blacks, it is likely that neighborhood effects may be stronger than for whites, due to the existence of ethnic enclaves. These groups may also have stronger cultural norms towards coresidence and close proximity.
Finally, if a larger percentage of these groups are immigrants or first-generation Americans, then the higher coresidence and closer proximity may be reflecting an increased need of the mothers due to language barriers and their lower ability to access outside (i.e., market) child care. It would make for an interesting follow-up paper to consider the race and ethnicity differences more closely.
(MRRC Working Paper 2009-215 was supported by a grant from the Social Security Administration through the University of Michigan Retirement Research Center.)