Quantifying the Cost of Caregiving

Kathleen McGarryThe aging of the population brings with it a host of pressing economic issues. Chief among these is the rising cost of health care and the need to provide care for our elderly population.  While health care costs are high across the age distribution, they are far larger among the elderly.  And among those 65 or older, costs are highest among the oldest old.  For example, in 2014 per capita health care spending for the population ages 19-64 was $7,153, while it was $19,098 for the population ages 65 or old, and $32,903 for those 85 or older.  (Data available at here.)

Much of this cost at older ages stems from the cost of long term care, particularly care for the cognitively impaired. In 2016 the cost of nursing homes and home health care totaled $255 billion. Medicare does not cover most long-term care and few individuals have private insurance to cover the expense. Thus, in contrast to other types of medical care, much long-term care is paid for out-of-pocket. In work with Amy Kelly (MSHS, UCLA) and Jon Skinner (PhD, UCLA), Professor Kathleen McGarry shows that average out-of-pocket spending in the last 5 years of life for those who die from Alzheimer’s disease or related dementias is more than twice that for those who die from cancer. Furthermore, among those with dementia as a cause of death, 56 percent of the out-of-pocket spending was on long-term care.

While the cost of this formal long term care is indeed large, these numbers provide only part of the picture. Missing from these costs is the cost of informal care. Back of the envelope calculations put this number at $565 billion (in 2017 dollars), more than twice the cost of formal care.  These imputed costs are calculated by multiplying the average hourly wage of formal home health providers by the number of hours of informal care provided.  However, we know little about the extent to which informal caregivers reduce their efforts in the labor market because of caregiving responsibilities and how costly any reduction may be.

In a second project, Professor McGarry’s investigates the economic impact of caregiving on working age women. Because women provide the majority of care and often care for elderly parents while the women themselves are still in their prime working years, this group is especially at risk for negative consequences.

In joint work with Sean Fahle (PhD, UCLA), McGarry follows a cohort from their 50s through their 70s to assess both the immediate and long-term effects of caregiving.  The authors find that approximately one-third of these women provide care to an elderly parent at some point during this time and the number approaches 45 percent if care for a spouse is included.  Caregiving is associated with a significant decline in working, in hours worked, and in annual earnings. However, the largest effects are when examining changes over a longer period of time.  Those who provide care to an elderly parent at some point start off in a stronger economic position than non-caregivers—they have greater earnings, more labor market experience, and greater household wealth–but by the end of the window of observation, caregivers have lost wealth relative to non-caregivers, are less likely to be employed and work fewer hours.  They are also significantly more likely to report being in poor health.

These results suggest serious long-term effects of caregiving that likely have negative consequences that affect the quality of life throughout retirement, and potentially lead to the dramatically higher poverty rates observed for elderly women relative to men.

Key to any policy analysis is the potential impact going forward. The lower rates of fertility for the coming cohorts of elderly mean fewer children to care for an elderly parent and the greater the burden on any particular child, while higher rates of labor force participation for working-aged women signal a greater opportunity cost for caregiving. Finally, improvements in the treatment of heart disease and cancer have increased life expectancy and the probability that an individual lives long enough to develop a cognitive impairment—suggesting a great demand for long term care.