Growing Inequality in Life Expectancy and Benefits for the Elderly
The rich live longer than the poor—which means they receive more support from federal programs such as Social Security and Medicare.
Average life expectancy in the United States has been rising for many decades and is expected to continue to rise in the future.  However, the overall average masks large differences across social groups: life expectancy is higher for people who have more education or have higher incomes. Furthermore, these differences have grown in recent decades. In fact, recent gains in life expectancy have accrued largely to those with high incomes or high education, while life expectancy for those with lower incomes and little education has stagnated. This is a very serious problem in itself, but it also has other implications.
A Longer Life Expectancy Earns the Rich More Benefits
A committee of the National Academy of Sciences, which we cochaired, investigated the way these growing disparities in survival affect the amount that different income groups receive from the federal government in lifetime benefits from programs such as Social Security, Medicare, and Medicaid. If higher-income people gain more years of life in old age, then they will also receive benefits from these programs for more years. The committee's report was released a few months ago. Its findings are striking.
We estimated the remaining life expectancy of men and women who were already 50 according to their average earnings levels when they were in their 40s. We found that 50-year-old men born in 1930, with earnings in the bottom 20 percent of the population, would live another 26.6 years, while men in the top 20 percent would live another 31.7 years, for a difference of 5.1 years. For men born in 1960, however, the corresponding numbers are 26.5 years and 38.8 years, for a difference of 12.3 years. The gap in life expectancy, which was already big, more than doubled. Results for women are similar. (See "Remaining Life Expectancy: Men" and "Remaining Life Expectancy: Women.")
How did these disturbing trends in length of life affect the government benefits received by the elderly? To answer that question, we considered a group of people turning 50 in 2010 and asked what benefits they would have received over their lifetimes (above age 50) under the estimated mortality conditions of this generation on the one hand, and on the other hand what they would have received under the mortality conditions of the generation born in 1930, when differences in life expectancy by income were smaller. We also took into account that those with lower life expectancy would generally have worse health and higher disability, because that would mean that they would receive more benefits each year from Medicare and Medicaid. 
For Social Security retirement benefits, we found, unsurprisingly, that men and women in the top 20 percent received more benefits that those in the bottom 20 percent, because they qualified for higher benefits per year and because they received those benefits over more years on average. However, while the gap in lifetime benefits for men under the mortality conditions of those born in 1930 was $103,000, with the mortality of those born in 1960 the gap was $173,000, or $70,000 greater. The comparable increase in the gap for women was $48,000. These larger gaps resulted entirely from the fact that higher-income people gained more years of life in which to receive benefits.
For Medicare benefits, the gap rose by $53,000 for men and by $70,000 for women. Even though people with lower incomes use Medicare more in each year that they are alive, because they are on average in worse health, the greater increases in lifespan for the high-income groups more than offsets their lower usage of Medicare per year.
Medicaid funds long-term care for individuals whose assets fall below a threshold value. Lower-income people receive Medicaid benefits for this purpose to a much greater extent than do high-income people, and women receive Medicaid benefits much more than men because they tend to have more disabilities. The rising gap in lifespan also favors the lifetime Medicaid benefits of higher-income people, in the amount of $7,000 for men and $36,000 for women. The committee also carried out similar calculations for Social Security disability benefits and for Supplemental Security Income (SSI).
We can add together the lifetime benefits for all five of these programs for the elderly to get a comprehensive picture. Comparing the mortality conditions of the 1960 generation with those of the 1930 generation, we find that the gap in lifetime benefits between earners in the top 20 percent and bottom 20 percent grows by $130,000 for men and by $160,000 for women. These are strikingly large numbers. (See "Total Lifetime Benefits: Men" and "Total Lifetime Benefits: Women.")
To be sure, we have thus far only considered the increasing gaps for lifetime benefits. Perhaps the picture changes if we take tax payments into account? When we looked at payments of federal income taxes and both employee and employer payroll taxes after age 50, we found that the mortality context made very little difference to the gap between high- and low-income people—only a $5,000 increase in the tax gap under the mortality conditions of the 1960 generation. When we put all this together to calculate how the widening disparity in lifespan affects net lifetime benefits (that is, after subtracting taxes paid after age 50), we found that the gap increased by $125,000 for men and $155,000 for women.
What Are the Implications for Policy Changes?
Clearly, the widening gap in life expectancy between high- and low-income people has reduced the progressiveness of the federal programs for the elderly in a very important way. It is well known that Social Security, Medicare, and Medicaid will face very serious financial problems as the population ages and as health care prices rise. These systems are not sustainable without policy adjustments. But the same changes in lifespan by income class that swung lifetime benefits in favor of high-income groups may also matter for the policy changes that are being discussed. For example, raising the retirement age or the age of eligibility for Medicare might fall harder on lower-income people whose shorter lives will mean a proportionately bigger cut in lifetime benefits, a topic the committee also investigated. The policy changes considered were raising the early retirement age and the normal retirement age for Social Security, raising the age of eligibility for Medicare, calculating the cost-of-living allowance in a different way, and reducing benefit rates for higher-income groups. For these potential policy changes, we assessed how lifetime benefits would be affected, given the wide disparities in life expectancy. (We were not able to consider another potential change, raising or eliminating the cap on wages subject to the salary cap, because of limitations of our model and data set.) 
The bottom line is that policymakers need to consider the consequences of the widening disparities in health and survival when they evaluate current policies and propose new ones.
Ronald Lee is an economist and demographer at the University of California, Berkeley. Peter Orszag is an economist, author, and chairman of the Financial Strategy and Solutions Group at Citigroup. Contact them at email@example.com
- According to the Office of the Chief Actuary of the Social Security Administration (see the OASDI Trustees Report for 2015, Table V.A3), life expectancy at birth was 71.1 years in 1970, 73.7 in 1980, 75.3 in 1990, 76.7 in 2000, 78.5 in 2010, and, based on provisional data, 79.1 in 2014. This is the average for males and females of period life expectancy.
- The report is titled "The Growing Gap in Life Expectancy by Income: Implications for Federal Programs and Policy Responses." It may be downloaded free at http://www.nap.edu/catalog/19015/the-growing-gap-in-life-expectancy-by-income-implications-for. To make the calculations described in this article, the committee used a microsimulation program called Future Elderly Model, or FEM, which also took into account the health and disability of individuals according to their income class. To calculate the present values of lifetime benefits and taxes, this program used a discount rate of 2.9 percent.
- The payroll tax for Social Security is applied only to earnings up to the cap, which is $118,500 in 2016. One policy change that is often discussed is raising the cap to a much higher level or eliminating it altogether. The payroll tax for Medicare is not subject to any cap at all.
Articles may be reprinted if Communities & Banking and the author are credited and the following disclaimer is used: "The views expressed are not necessarily those of the Federal Reserve Bank of Boston or the Federal Reserve System. Information about organizations and upcoming events is strictly informational and not an endorsement."