Abstract

For elderly parents, adult children are the most common type of informal care providers. Many studies have examined the effect of informal care provision for a parent on the labor force participation, hours worked, and on the wages of the adult child providing that care. We contribute to this literature by examining the labor supply response of one spouse to the other spouse’s caregiving decisions.

A priori, it is difficult to tell whether caregiving would have a positive or negative impact on the non-caregiving spouse’s labor supply. The spouse could increase work effort to compensate for the caregiver’s reduced earnings or early exit from the labor force, or because of the anticipation of having to spend money on eventual formal care for the parent. Alternatively, if the caregiver drops out of the labor force and either cannot or does not wish to return, even after the caregiving episode ends, the spouse may retire earlier in order to satisfy the desire for joint retirement and leisure time. By considering the effect of caregiving on the labor supply of both members of a married couple, we can more accurately assess the impact of caregiving on household income, wealth accumulation, and retirement timing. In so doing, this paper provides an estimate of the total impact of informal care provision on labor force participation, hours worked, and wages.

Given the current level of informal care provision and its projected growth, this is an important step for predicting retirement behavior and retirement wealth. We use data for married couples with a parent or parent-inlaw alive from the Health and Retirement Survey. We use fixed-effects models to control for time invariant heterogeneity, as well as instrumental variables to address remaining concerns about the endogeneity of caregiving with respect to labor-market outcomes.

Overall, we find that a spouse’s informal care behavior causes very little change in the extensive work margins. Working married women’s hours, however, are responsive to their spouse’s caregiving, just not in the direction suggested by a self-insurance motive. Instead of working more, working women increase their hours per week if they provide care and decrease hours if their husbands provide care. The finding of significant effects on hours of work for both the wife’s and the husband’s caregiving suggests that models that do not consider the caregiving behavior of one’s spouse may lead to incorrect measurement of the effect of caregiving on work for married individuals. If both the husband and wife are caregiving, a common scenario observed in the data, the wife is able to maintain her usual hours of work per week. One possible explanation is that these couples anticipate the future long-term care needs of the parent, and they adjust their respective caregiving efforts to accommodate each individual’s work schedule and maintain their income.

Slides