The existence of a growing gap in longevity between the rich and the poor is clearer than ever: Between 1930 and 1960, men at the top of the economic ladder saw an eight-year increase in life expectancy, while men at the bottom saw virtually no change. That’s just one finding from the National Academy of Science, which has documented how the growing inequality of wealth and income in the U.S. has been accompanied by, and perhaps is actually causing, an increasing gap in life expectancy between the wealthy and the working class.

For men born in 1930 who lived in the bottom 20 percent of income distribution, life expectancy at age 50 was 76.6 years; for those born in 1960, it was mostly unchanged at 76.1. For men who lived in the top 20 percent of the income distribution, it was a different story: Their respective life expectancy numbers jumped from 81.7 to 88.8. (A similar gap grew between white and black people, and the trends for women were about the same, but not as dramatic.)

Certainly, retirement has become more democratic since Social Security was passed in 1935, guaranteeing a stable income to all retired Americans. And cleaner air, cleaner water, and anti-smoking campaigns have greatly increased life expectancy. But while the National Academy’s research doesn’t establish causal links, it demonstrates an insidious new form of inequality that has serious implications for retirement security—including Medicare, Social Security, and pensions—and long-term care.

This gap is not just about overall longevity—it’s about the quality of life the elderly will have at the end of their days. Today, the wealthy live longer, and additionally tend to have well-educated children who can navigate health-insurance paperwork and help advocate and pay for better health care.

Recently, a friend who is an expert in health-care policy and the regulation of care work wrote to me saying that she is dealing with end-of-life care for her father. She asked, “How do people with fewer resources (of all kinds) manage this?” She and her sister had two Ph.D.s between them, yet all of the medical, financial, and legal details have taken her “nearly to the limits of my intellectual capacity … My heart goes out to families with fewer resources and less familiarity with U.S. social policy than I have!”

My friend could not have summarized the research better. The growing class, education, and racial gaps in longevity go along with corresponding gaps in quality of end-of-life care. People with savvy children tend to live longer and die in the presence of loved ones. Those who attended college are about 1.5 times as likely to choose hospice care for their parents than those who never went. And people who live in neighborhoods with a higher median income were more likely to use hospice than people from lower socioeconomic backgrounds.

The idea that everyone should work longer since everyone is living longer is one used to justify policy proposals such as cutting Social Security benefits. But that idea is a misleading oversimplification.

The Retirement Equity Lab at The New School (a project I direct) has pointed out that the growing class and racial gaps have dire implications for retirement policies. A cut to Social Security benefits—which raising the retirement age, an oft-suggested proposal, essentially is—would induce people without means to work in old age. This would produce an unseemly form of inequality: The people who live the longest will be able to retire, and the people who have to work longer will be the same people who are losing at longevity. The poorer will work and the richer will play in old age, a class divide we’ve already seen in the 19th and early 20th centuries. If post-work benefits are not shored up, this disparity will only get worse.