At the Joint Economic Committee’s hearing on social capital in America, Harvard political scientist Robert Putnam highlighted a looming problem that has flown under the policy radar. Many people know that we face a challenge of providing elder care for baby boomers. But what is less recognized, according to Putnam, is the burden that deficits of social capital in this generation will place on paid forms of care. The boomers “will almost certainly require substantially more paid eldercare per person than their parents’ generation.”

Earlier retirement and longer lifespans are part of the problem, but federal spending projections already take those into account. We ignore the social capital problem, however. Elder care is provided by both informal sources (e.g. family and friends) and paid sources (e.g. nursing homes). Putnam argued that boomers have fewer sources of informal care than previous generations, and that paid sources of care (including those financed by Medicaid) will therefore have to fill in the resulting gap. Based on preliminary estimates that should be considered tentative, Putnam calculated that boomers “are entering retirement with one third less social support than their parents had at the same stage of life.” Compared to the birth cohort of 1930, boomers will enter retirement with lower rates of living with a spouse (12 percent lower), fewer children (36 percent lower), fewer close friends (30 percent lower), and fewer community ties (40 percent lower).

What is more, having fewer social relationships to lean on will have implications for boomers’ quality of life. As Putnam notes, “[s]ocial isolation is widely recognized as a strong predictor of morbidity and mortality, especially among the elderly.”

These large and currently underappreciated future costs of the decline of social capital put into stark relief new research on the future of “kinlessness.” By 2060, over 21 million Americans over age 50 will be without a living partner or child, up from 14.9 million in 2015. That’s according to a paper published in the Proceedings of the National Academy of Sciences by Ashton M. Verdery of Pennsylvania State University and Rachel Margolis of the University of Western Ontario. These kinless Americans would outnumber those expected to have Alzheimer’s (13.8 million), diabetes (16.9 million), or arthritis (20.6 million). Moreover, the number of Americans without a living partner, children, or parents or siblings is expected to jump from 1.8 million to 6.3 million over the same period.

The growing number of older adults without living family members means that more and more Americans will face old age with fewer sources of support and care. Family members provide the bulk of long-term care to the elderly, and not surprisingly older adults who are without a spouse or children are more likely to be placed in a nursing home. Loneliness in later life is a significant risk factor connected to poorer physical and mental health.

Several factors are driving the increase in kinlessness. The largest factors the researchers emphasize are the decline of marriage and the increase of childlessness. Between 1990 and 2010, “Gray divorce”—divorce in older age—doubled, which also adds to the number of unpartnered older Americans. Smaller family size is also part of the equation. Adults have, on average, fewer siblings—or are more likely not to have any siblings—compared to previous generations, and are also less likely to have a sibling that is several years younger. Thus, older Americans will be less likely than in the past to have a living sibling in old age.

Of course, an increase in cohabitation or other types of non-marital unions, like long-term dating relationships, could potentially offset some of the predicted increase in kinlessness attributable to marriage decline. However, even when the researchers took into account the growth of cohabitation and dating relationships, and assumed the same degree of stability in these types of relationships as in marriage (currently, marriage is far more stable), they still predicted a major growth in kinlessness.

Overall population growth and the increase in the elderly population as a whole also plays into the growth of kinlessness in later life. The percentage of Americans 50 and older is projected to increase from 34.6 percent in 2015 to 41.5 percent by 2060, and population growth means that the U.S. will have roughly 60 million more older Americans than it currently does.

The researchers note that kinlessness is expected to be greater among black Americans. They predict that the percentage of older white Americans without a living spouse or child will remain roughly the same between 2015 and 2060, at approximately 8 to 10 percent for both men (from 6.6 million to 8.2 million) and women (from 6.3 million to 7 million). But for black men and women, those over age 50 without a spouse or child are predicted to increase: from 9.7 percent to 12.6 percent for black men (from 1 million to 2.7 million), and from 10.5 percent to 15.1 percent for black women (from 1.1 million to 3. 3 million).

When it comes to the broader measure of kinlessness—those without a living partner, children, siblings or parents—the authors estimate that the percentage and numbers for both blacks and whites are expected to climb. For white men and women, the researchers project that the percentage will double, from 0.8 percent to 1.9 percent for white men, and from 1.1 percent to 2.2 percent for white women. For black men and women, they project a much larger increase: from 1.7 percent to 5.6 percent for men and from 2.2 percent to 7.3 percent for women.

Those in later life often require care as their physical and cognitive functioning decline. However, an increasing number of older Americans than in past generations will be without the support of close family members as they experience this stage of life. The time to start planning for the problem has long passed, but better late than never; how will we take care of the aging boomers?