The Opportunity Costs of Informal Elder-Care in the United States: New Estimates from the American Time Use Survey

Abstract

There is an extensive literature studying the phenomenon of informal caregiving for the elderly. In the United States, informal (unpaid) caregiving by family members and friends remains the primary source of long-term care for the elderly population. Arno, Levine, and Memmott (1999) drew attention to the sheer size of this informal care “sector,” estimating that 22–26 billion hours of informal care time were provided in 1997. Recent estimates (National Alliance for Caregiving and AARP 2009) indicate that nearly one in five adults in the United States provides care to an elderly relative or friend older than 50.

While these figures reveal the prevalence and intensity of informal eldercare in the United States, estimates of the costs of this care have been limited by the lack of nationally representative data with detailed information to allow an assessment of the value that caregivers attach to time and time spent providing care. This paper takes advantage of new and unique data from the American Time Use Survey (ATUS) to remedy this deficit, and to provide careful, up-to-date estimates of the costs of informal elder-care in the United States.

In doing so, we update the estimates of Arno, Levine, and Memmott (1999) and Feinberg et al. (2011), while improving on these previous studies by valuing informal elder-care at opportunity cost. Opportunity cost of time, the economic value of activities forgone as a result of providing care, may be thought of as the price of supplying informal care and, as such, is one of the key determinants of how much informal care is provided.

Estimating the opportunity costs of informal care is important from a policy perspective. For many individuals, the opportunity costs of providing care may include lost income due to having reduced work hours to accommodate caregiving (Evandrou, Glaser, and Henz 2002; Henz 2006; Lilly, Laporte, and Coyte 2007; Lee and Tang 2013; Van Houtven, Coe, and Skira 2013). High opportunity costs provide a rationale for workplace flexibility policies, such as those mandated by the Family and Medical Leave Act, as well as newer programs under consideration in a number of states that would allow caregivers to take paid time off (Houser and Vartanian 2012). Relatedly, opportunity costs are an important consideration in the context of programs that allow informal caregivers to be directly compensated, such as Medicare's Cash and Counseling program that provides each eligible long-term care beneficiary with a budget that may be used to pay any caregiver of the beneficiary's choice, including family caregivers (Doty, Mahoney, and Sciegaj 2010). If the opportunity costs of informal caregivers are lower than the cost of formal replacements, this may also result in significant cost savings for Medicaid.

Method

Results

Of the 25,701 individuals in the sample, 5,987 (approximately 23.29 percent) identified themselves as caregivers within the past 3 months, and 1,383 (5.38 percent) reported providing care during the previous 24 hours. The remainder of our analysis will focus on the group that provided care in the 24-hour recall period.

Table​Table11 describes these caregivers, by comparing them to the general population. Age and labor market status are clearly two important dimensions along which caregivers differ from noncaregivers. Although caregivers are on average older than noncaregivers, more than three-quarters of caregivers are under the age of 65, and about three in five caregivers are in the labor force.

In the case of men, caregivers are more likely than noncaregivers to be out of the labor force and less likely to be employed. Even among working-age men (i.e., under the age of 65), caregivers are more likely to report being out of the labor force. In the case of women, there is a smaller difference in labor market status between caregivers and noncaregivers.

Table​Table22 describes care recipients in terms of their relationship with the caregiver and the amount of care they receive, distinguishing between older (over 65) and younger (under 65) care recipients. The majority of care recipients are older than 65. Older recipients are more likely than younger recipients to receive care from their spouses and, unsurprisingly, also receive more care than younger recipients.

We now turn to estimating the opportunity costs of caregiving. In the case of women, the probability of labor force participation falls with the number of children under the age of 6 in the household (the excluded variable in the wage regression), whereas in the case of men, this variable has a positive effect on participation (Table​(Table3).3). In the case of both men and women, the coefficient on the constructed omitted variable (denoted by λ in the table) is significantly different from zero, confirming that sample selection bias was indeed a nonignorable issue.

Table​Table44 summarizes the estimated average and total care time for the population, along with estimated average opportunity cost. We estimate that informal caregiving amounts to 30.06 billion hours annually. Because most caregivers are in the under-65 group, this group accounts for the majority of informal care (22.3 billion hours annually). This group also has higher average wages and accounts for the largest fraction of the opportunity costs of caregiving.

Excluding care given by persons above 65, the opportunity costs of informal care amount to $412 billion annually—the estimate rises to $522 billion annually when we include older caregivers (Table​(Table5).5). We estimate the replacement costs of informal caregiving at the wage rate of an unskilled paid worker (“unskilled replacement”) and at the wage rate of a home health aide (“skilled replacement). The average wage of informal caregivers lies in between the unskilled and skilled paid care wage rates. Accordingly, we find that the opportunity cost estimates lie in between the two replacement cost estimates ($221 billion and $642 billion, respectively).

Discussion

Informal caregiving remains an economically significant phenomenon in the United States, accounting for approximately 30 billion hours of care annually and $522 billion a year in opportunity cost. This is in contrast to total spending on formal long-term care services in the United States, which was estimated at $211 billion in 2011 (O'Shaughnessy 2013).

It is useful to compare our cost estimates with those reported in the existing literature. Arno, Levine, and Memmott (1999) and Feinberg et al. (2011) have estimated the replacement costs of informal care, using similar methodologies. The more recent estimate from Feinberg et al. (2011) used data from the Caregiving in the U.S. survey and the Behavioral Risk Factor Surveillance Survey (BRFSS), and estimated the total annual replacement value of informal care at $450 billion in 2009. Their estimates lies in between our two replacement costs estimates, the difference arising due to differences in both the assumed economic value of time as well as in measured hours of caregiving. First, whereas we have separately considered unskilled and skilled paid care, Feinberg et al. (2011) measured replacement cost as a weighted average of state minimum wage, the median hourly wage of a home health aide, and the private pay hourly rate to hire a home health aide. Second, their estimate of caregiving time is higher, at approximately 40 billion hours per year. Part of this discrepancy may be due to the fact that definitions of informal caregiving vary across different surveys. In addition, surveys also vary in the manner in which caregiving time is elicited: The ATUS is unique in that it measures care time in a fairly precise way (in the context of a time diary), whereas in surveys such as the BRFSS and the Caregiving in the U.S., respondents are asked how many care hours they provide in an average week.

We find that informal care is mainly provided by working-age adults, who consequently bear most of the economic burden in terms of opportunity costs. Our findings underscore the importance of workplace flexibility policies being considered by a number of states that provide paid time off from work for caregivers, as well as programs such as Medicaid's Cash and Counseling program that allows family caregivers to be paid for their assistance. Our estimates indicate that, in the aggregate, although the economic value of informal caregiving is substantially greater than what it would cost to replace this care with paid unskilled care, informal caregiving remains more economical than paid skilled care, suggesting that if the beneficiaries enrolled in Cash and Counseling programs prefer to receive care from family members, this may result in cost savings to Medicaid relative to a situation in which the care is provided by skilled replacements.

Our study has some limitations. First, we acknowledge that treating the wage as a measure of the opportunity cost of time, although it has a long tradition in economics, is not entirely accurate. More refined models of time allocation than the Beckerian model suggest that the true shadow value of time may be different from the wage (Johnson 1966; Oort 1969; De Serpa 1971; Evans 1972). Even in the Beckerian framework, the relevant shadow value of time for a working individual is the after-tax wage, including benefits. This actual wage is difficult to observe. Further, the wage forgone by a nonworking individual may be an underestimate of the value of his/her time. Notwithstanding these critiques, we believe that the measured wage remains a useful, practical approximation to the true shadow value. A second limitation of the study is that we have treated care as a broad category, encompassing many different activities. A more narrow definition that focuses on assistance with physical activities (ADLs, for example) would be useful, but it cannot be estimated using the data available. A third broad limitation of our study is that by focusing on the opportunity costs of care, we have excluded from consideration other well-documented costs associated with providing informal care, such as impacts on the emotional and physical health of the caregivers (Schulz et al. 1995; Pinquart and Sörensen 2003; Vitaliano, Zhang, and Scanlan 2003; Coe and Van Houtven 2009). Including these costs should substantially increase the total costs associated with informal care.

Acknowledgments

References