Senior Class: America's Unequal Retirement


Senior Class: America's Unequal Retirement

One of the cruelest manifestations of widening inequality happens in life's final quarter.

April 24, 2015

You may also like

This article appears in the Spring 2015 issue of The American Prospect magazine. Celebrate our 25th Anniversary with us by clicking here for a free download of this special issue.

Inequality has been increasing in multiple ways. But one little-appreciated form is the inequality of retirement time. That’s the number of years between retirement and death. It’s not surprising that divergent retirement time should reflect other forms of growing inequality. The poor have lower earnings and often work longer out of necessity, not choice. They are less likely to have decent pensions or private savings. On average, they suffer poorer health and tend to die younger. On all counts, the affluent get to enjoy more years of retirement in relative comfort.

This is just what one would expect in a market economy of rising inequality. What is surprising, however, is that there was once a broadly equal distribution of retirement time across divisions of class and race. This greater equality was a product of several egalitarian policies and institutions created in the mid-20th century, all of which are now under assault—Social Security, Medicare, pension plans, and disability policy. Thanks to these measures, as recently as the 1980s, low-income people who spent roughly the same number of years in the workforce as high-income people obtained approximately the same years of retirement.

In addition, many blue-collar and service-sector workers, especially those in unionized jobs that required hard physical and tedious labor, like miners, autoworkers, truckers, and nurses, had bargained for defined-benefit plans, many of them so-called 30-and-out plans. They could retire at a substantial fraction of their final earnings after 30 or so years on the job. But today, most workers have lost the traditional pensions that once covered about half the workforce. Meanwhile, the affluent are more likely to be healthier as well as wealthier in their golden years—all of which adds up to the elderly rich being able to control their quality of life to a far greater degree than middle-class and poor elders.

(Illustration: Victor Juhasz)

As traditional pension coverage has declined, those who still have adequate retirement packages are likely to be professionals, business executives, and the wealthy. To the extent that middle-class and poorer Americans have access to retirement plans at all, they are typically 401(k) or similar plans that depend on optional contributions from workers and employers. For most workers, these ersatz pensions are inadequate. The median retirement account balance is just over $110,000 for people near retirement, aged 55 to 64. That’s sufficient for just a few years of retirement. And the majority of workers in the bottom two-thirds of the income distribution don’t even have any retirement savings. Moreover, because of the lopsided way that 401(k)s and their cousins, Individual Retirement Accounts, are treated under the tax code, the contributions and earnings of the highest paid participants are subsidized with more favorable tax breaks.

Social Security makes up only part of the gap. The average Social Security benefit is only about $1,200 per month. That’s why the idea of raising the retirement age or extracting other forms of cuts in Social Security is perverse at a time when voluntary employer pensions, home equity, and personal savings—the other layers of the retirement system—are withering.

The Growing Gap in Life Expectancy

The remarkable increase in the average American’s lifespan since World War II conceals a growing racial and class disparity in life expectancies. Social Security economist Hilary Waldron, in a 2007 study, compared two generations of male workers. For men born in 1912, who reached their prime earning years when the pension system was still healthy and the income distribution was more equal, the top half of those who survived to age 60 lived only 1.2 years longer than those in the bottom half. But for men born in 1941, who hit peak earning years around the turn of the 21st century when inequality was widening and pensions were decaying, the top half lived 5.8 years longer than the bottom half. The longevity gap between the top and bottom had dramatically widened. 

In 2014, a Brookings Institution study by researchers Barry Bosworth and Kathleen Burke confirmed Waldron’s findings that class longevity gaps are growing. Men born between 1920 and 1940 who survived to age 55 and were in the bottom tenth of the income distribution could expect to live to age 79.2, while men in the top tenth of the income distribution could expect to live much longer, to age 89.3. Low-income women had longer life expectancies than men with equally low incomes. But though women’s life expectancy has generally exceeded men’s, men and women in the top tenth of the earnings distribution remarkably had the same life expectancy at age 55; they were expected to live until age 89.3. Class trumped even the long-standing fact of greater female longevity.

The retirement gap continues to widen. For men born in 1920, according to the Brookings study, the class difference in how long people could expect to collect Social Security benefits was considerable, due to divergent mortality. Men in the bottom tenth could anticipate collecting benefits for 16.6 years, while men in the top 10 percent of lifetime earnings were expected to collect for 20.7 years. But among men born just 20 years later, in 1940, the gap had widened considerably. Those at the bottom could expect to collect benefits for 18.2 years, but those at the top expected to collect for 26.4 years. In less than a generation, a 4.1-year gap turned into an 8.2-year gap.

Life expectancy is actually declining for women in the bottom 10 percent—and rising for those at the top. Women born in 1920 who remained in the bottom 10 percent of lifetime earnings could expect to collect benefits for 21.2 years, but those born in 1940 and also in the bottom 10 percent could expect to collect for only 19.2 years because their average lifespan had decreased. In contrast, lifespan increased for the top 10 percent: Those born in 1920 had 24.5 benefit years, and those born in 1940 had 27.5 years. So the class difference in benefit years increased from 3.3 years for those born in 1920, to 8.3 years for those born in 1940.

Not surprisingly, retirement inequality is also widening by race. America once had greater racial equity in one very narrow respect. In 1950, both blacks and whites who managed to survive to age 65 were expected to live to age 77. By 2010, white men at age 65 were projected to live almost two years longer than black men and white women a year longer than black women. Blacks, then and now, are much more likely than whites to die before age 65. Today, less than 70 percent of black men survive until age 65, compared to more than 80 percent of white men. The racial gap between black and white women is not as great but is still significant; 81.3 percent of black women survive to 65 compared to 88.3 percent of white women.

The increasing divergence in life expectancy between whites and blacks, and between low- and high-income people, matters—not just for its own sake but in the debates over Social Security. Any policy that cuts Social Security benefits, such as by raising the “normal” retirement age—at which covered workers can collect their full Social Security benefits—will hurt blacks and lower-income people more because both groups die sooner than whites and higher-income people. African Americans and lower-income people also tend to suffer poorer health, which affects both their capacity to work and their quality of life.


Widening Inequality in Health

Health gaps by income and race are their own strand of this story of who gets to control the pace and quality of their life before they die. Lower-income people have more difficult jobs in many ways. They are more likely to be subordinate at work; they face higher risks of losing pay, hours, and jobs, all of which are sources of social and psychological stress that damage health. According to a 2010 Canadian study, poverty by itself doubles the risk of diabetes. Smoking is notoriously correlated with class, as is the availability and quality of treatment for heart disease. Class disparities in health—caused in part by differences in access to quality health care—interact with retirement disparities in another respect: Older people in ill health, though desperately needing to retire, increasingly can’t afford to stop working. Yet lower-income older workers in poor health are less desirable to employers and often have trouble finding jobs.

Once, traditional defined-benefit pension plans combined with disability payments provided a bridge—a source of income until people could begin to collect Social Security at age 62. Some people can use disability payments to retire early, but qualifying for disability benefits is increasingly difficult. It takes two years of not working, and legal and medical experts have to declare that a person is completely unable to do any work for pay—which is an extremely high bar.

Like the other inequalities of aging, these disparities have a racial dimension. In 2013, the Centers for Disease Control and Prevention reported the average African-American man can expect to live 61.1 years free of activity limitations caused by chronic conditions. By comparison, the average white American man can expect to live until age 67 without such limitations. If blacks retire in their early 60s or later, a majority will spend all of their retirement with some form of activity limitation. Among people of the same race at comparable ages, those with lower incomes have worse health.

Elders Who Must Work and Those Who Can Afford Not To

If the groups with shorter average lifespans could be accommodated by more generous early retirement and disability benefits, the nation could keep one aspect of a fair retirement system. Ideally, people who start work at older ages because of more schooling—and who live longer because life treats them better—would retire at older ages. Workers with less education, earlier entry into the labor force, and shorter life expectancies would be able to retire at earlier ages.

Yet the opposite is occurring. Because of the erosion of pensions and the attack on retirement, the vast majority of Americans over 65 in the bottom 90 percent of the income distribution will likely have to work longer than the affluent, even as they also expect to die sooner. This is not a fair retirement system. Yet it seems that the racial and class gaps in retirement time are already baked into retirement policy. The most widely used model to project retirement income, the “MINT Model,” maintained by the Urban Institute and funded by the Social Security Administration, takes the unequal distribution of work as a given. The model projects that 30 years from now,middle-class retirement-age Americans will receive 24 percent of their income from working—compared to 67-year-olds in 2002, who got only 14 percent of their income from work.

Currently, the Social Security Administration assumes that middle-class 67-year-olds in 2040 will get less of their income from assets and retirement plans and almost twice as much of their income from working as they did 40 years earlier—because they can’t afford not to work. But for those in the top 20 percent, the SSA model assumes the younger cohort will not have to work more. The top 20 percent’s share of income coming from wealth, defined-benefit plans, and retirement accounts will grow substantially, from 63 percent to 73 percent. And increased work is also assumed for the non-rich well past retirement age. Social Security economist Mitra Toossi projects that by 2022, about 14 percent of men and 8 percent of women aged 75 and older will work, compared to 7 percent of men and 2 percent of women in 1992. The bottom line: Increased work is proposed as the solution to reduced retirement income for the middle-class workers, but not for the affluent.

But complicating the projected solution of increased work for middle-class elders who have inadequate pensions are the realities of employer demand for older workers. Older workers often have difficulty finding employment. Age discrimination in hiring, training, and wages and salaries, though illegal, is on the rise. This is the price we pay as a society for the absence of a comprehensive pension system, consistent norms for the proper age to retire, and decent pensions above the essential Social Security.

As 401(k)-type plans replace traditional defined-benefit pensions, and as other forms of employer-based retirement plans stagnate, pension income is expected to fall for middle-class as well as lower-income workers. Workers in a 401(k) world, which now is only available to about half of the labor force, will need to work longer and cut their retirement time. Unlike Social Security, which is deliberately redistributive, 401(k) assets and income mirror the growing inequality of wage and salary income. As the 401(k) system matures, one of the key equalizing mechanisms of the American retirement system, which let people who died sooner retire sooner, will be lost. Retirement time and retirement income will become more unequal.

In 1992, the elderly in the top 20 percent had almost three dollars in retirement income for every one dollar the middle-income retiree had. By 2010, retirees at the top had almost five dollars for every dollar a middle-class retiree had. The ratio of the middle-class retiree’s average income to the top quintile’s average was 35.7 percent in 1992 but only 22.4 percent in 2010. This gap will only widen if we continue the same set of policies, and it will worsen further if Social Security benefits are cut.

A more subtle form of inequality in the years after 65 reflects why people in different social classes choose to remain in the paid labor force. There are those who work for love, and those who work for money. People in the elite professions are likely to keep working because they gain satisfaction, prestige, or influence from their jobs. People with routine jobs may be eager to retire, but increasingly find themselves still in the labor force because they need the money. The unequal income distribution among retirees only means that the difference between those who work past age 65 for love and those who work for money will get wider.

The Remedy: Expanded Social Security and Universal Pensions

Currently, Social Security–eligible workers may start collecting benefits at any age between 62 and 70. The earlier they take benefits, the smaller their checks are. If people can wait until age 70 to collect, either because they have other sources of money or can work, there is a huge financial gain. The Social Security benefits at age 70 are 76 percent higher than the benefits collected at age 62. However, the person waiting to collect at age 70 chose to forgo a check for eight years—either because he or she did not need the money, or because of a wager on a long life expectancy. If everyone lived the average lifespan, then the system would be actuarially fair; the worker who collected at age 62 would have the same lifetime benefit as one first claiming at age 70. But people with lower incomes have worse job prospects in old age and tend to die sooner. So it makes a great deal of sense for them to collect at earlier ages—and many do.

By contrast, those more affluent people who defer collecting until age 70 will likely live longer than average; they are more likely to have jobs they enjoy; they may well have investment and pension income—and they also get larger Social Security checks. Lucky them. The policy goal embedded in the current system is to ensure that, based on average life expectancy, people who take a lower benefit early receive about the same amount in total benefits over their lifetimes as those who wait to receive higher monthly benefits later. But the system doesn’t achieve that because it assumes away life-expectancy gaps of class, educational attainment, and race. A higher normal retirement age would only compound those inequities.

Proposals to raise the full retirement age to 70 (and even 76) from age 67 (which it will be in 2022) would drastically reduce benefits collected at earlier ages. Discussions about the retirement age are typically based on averages. But increases in average longevity obscure the wide differences in life-expectancy gains by race and class. Reducing benefits by raising the age at which one can collect full benefits would only intensify economic and racial inequalities. It would make low-income elders even more vulnerable by cutting their income and forcing many to compete in the low-income labor force.

The period at the end of one’s life in which one can live in peace, free from the toil of work, has become an essential part of the American social contract. Until recently, our system delivered on that contract. Raising the Social Security retirement age would serve to diminish the quality of life for all older Americans. For blacks, women, and the bottom 90 percent of income earners, the reduction would be more severe.

Rather than cutting Social Security, we need to increase it, as Senators Elizabeth Warren and Sherrod Brown and others have proposed. In addition, we need to complement Social Security with a fully portable retirement plan for all American workers, with subsidies for low earners. The combination of enhanced Social Security and a universal pension plan would need to offset the inequality of earned income during working life, not reinforce it.