Is Informal Care Dynastic? Exploring Family Trends in Caregiving Behavior

Presenter: Norma Coe, Boston College Center for Retirement Research

Abstract

The evidence is clear that providing informal care can cause adverse emotional and physical health effects on elderly spousal caregivers, at least in the short term, and even up to five years after a caregiving experience (Schulz et al 2001). While less in known about the health effects of caregiving on adult children, Coe and van Houtven (2008) find that there are negative mental health consequences to providing care for elderly mothers. Married children show an increase in depressive symptoms, which is an effect that persists over the long-term for married men.

Furthermore, we know that adult children will become an increasingly important source of informal care as the baby boomer generation ages, the number of divorces increases, and the differential life expectancy between men and women results in a larger number of widowed elderly women. Thus measuring the effects of providing care for elderly parents is imperative for policy formulation. This project asks whether informal care, and the health and wealth effects of such care, is limited to a one-generation transfer of time or is dynastic, persisting over more than one generation. Because informal care can be a substitute for long-term care (Van Houtven and Norton, 2004; 2008), knowing the persistence of informal care behavior across generations will help improve the accuracy of predictions on long-term care costs. Uptake of long-term care insurance is used to test cross-generational informal care behavior.

Using the Health and Retirement Survey (HRS), we have a long panel of caregivers and potential caregivers, and we examine both the start and the end of the caregiving episode. We use an instrumental variables approach in order to address the endogeneity, using family and sibling characteristics, as well as the death of a parent as instruments. Coe and van Houtven (2008) find that these are strong identifying instruments perfectly suited for this study.

We examine what caregiving behavior does to adult children’s’ own perceived risk for long-term care, and what effect this might have on choices for their own long-term care provision. Adult children who provided care to their parents may be more likely to expect the same behavior from their children, thus the long-term effects of caregiving would be cumulative over the generations. Alternatively, providing care to an elderly parent might make one more likely to use formal health care or purchase private long-term care insurance, all else equal, which implies that any effects of informal care provision will be limited to one transitional generation per family. This suggests very different long-term care arrangements in the future, with an increased uptake of private long-term care insurance and an increased reliance on formal long-term care.

It is important to know how informal caregiving affects the subsequent uptake of informal care for themselves and long-term care insurance in order to be able to predict the future of formal long-term care demand, and the costs associated with it.