Health Affairs, 22, no. 3 (2003): 168-174 doi: 10.1377/hlthaff.22.3.168 © 2003 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Knickman, J. R. Articles by Kennell, D. L. Search for Related Content PubMed PubMed Citation Articles by Knickman, J. R. Articles by Kennell, D. L. Related Collections Access To Care Elderly Long-Term Care

Health Tracking

TRENDS

Wealth Patterns Among Elderly Americans: Implications For Health Care Affordability

Abstract

 
This paper estimates the ability of the elderly to pay for necessary health care services and emerging technologies. Projections from the Long Term Care Financing Model paint a promising picture of the income and assets that elders in the future will have available to support discretionary, uncovered health care and service costs. Nevertheless, policymakers should pay close attention to the finances of the "Tweeners"—people who are middle class with low levels of discretionary assets available for health and long-term care.

Unless public insurance systems become much more generous in coming years, the elderly will bear the costs of many types of uncovered services. Drug and long-term care costs now top the list of uncovered services.¹ However, emerging elective procedures, perhaps in the area of gene therapies and cutting-edge diagnostic tools, may not be uniformly covered by future insurance programs.

This paper presents projections of income and wealth for the elderly population in 2015 and 2030 using state-of-the art simulation methods. An indicator of "resources available" for uncovered services is also developed. We address important policy and planning questions for 2030 such as (1) How big will markets be for elective, high-cost health and preventive services if insurance does not cover the interventions? (2) How much extra burden can the government expect related to health and long-term care for the poor and the part of the population that becomes impoverished trying to pay for care? (3) Will wealth patterns and emerging technology lead us increasingly toward a multiple-tier medical care system, with the wealthy in the top tier and the poor and middle-income in the lowest tiers?²

Data And Methods

Top Data And Methods Study Results Discussion And Policy... NOTES

 
The projections of the elderly’s income and assets in 2015 and 2030 are based on a simulation model originally constructed by researchers at the Brookings Institution and the Lewin Group.³ This model has been developed and refined since 1986, and the federal government and researchers use it extensively to understand income and asset distributions of the elderly as well as their patterns of disability and use of long-term care services over time.⁴

The model uses data from the April 1993 and March 1994 Current Population Survey (CPS) as the starting point for simulations.⁵ Most behavioral estimates that affect the simulations have been reviewed by experts. Key assumptions of the projections reported in this paper are about how the economy will grow or contract over the next thirty years. In almost all cases, the simulation model uses the economic assumptions used for the Intermediate Scenario in the 1999 Social Security Trustees Report.⁶

The model is designed to capture the interaction of demographic and economic factors that affect the resources of the elderly and their use of acute and long-term care services. Using demographic assumptions primarily from the Census Bureau and the Social Security trustees, the model simulates the changes in family structure (births, death, divorces, marriage) and disability for both the elderly and the nonelderly. The model tracks the effects of changes in cohort size as well as trends in the number and characteristics of people in various sociodemographic groups.

The model simulates the number of disabled widows age eighty-five and older in 2030 based upon the size of that cohort of women who will be age eighty-five and older in 2030 along with assumed trends in marriage, divorce, mortality, and disability. Some of these trends (cohort size and reduced mortality) raise the number of disabled widows, while others (declining disability rates) tend to decrease it. Similarly, the model simulates the work experience of the representative population, their incomes while working, and their entitlement to future Social Security and pension benefits. Based on assumed retirement rates, the model simulates Social Security, pension income, and retirement wealth.

All of the projections we present are measured in 2000 dollars to adjust for inflation’s effects on purchasing power. The assumptions about inflation rates used for the projections mirror those used for the Intermediate Scenario of the Social Security Trustees Report: 3.3 percent, on average, for 2007 and beyond.

Study Results

Top Data And Methods Study Results Discussion And Policy... NOTES

 
Income and assets. Real income among the elderly will increase greatly between 2000 and 2030 (Exhibit 1). In particular, fewer elderly will be in the lowest income bracket, and the percentage with real incomes exceeding $80,000 per year will almost double.

The long-term care resource estimate for single people is often an overestimate because people would still need some income to pay for day-to-day expenses if they were living in the community. However, we wanted to be as conservative as possible in considering who might have inadequate resources and thus require public resources. Perhaps the most appropriate perspective is from the vantage point of a single person entering a nursing home: Such a person can devote all available income and assets to covering expenses associated with a long, terminal nursing home stay.

For married couples, the estimate divides a couple’s assets and income using the principles adopted by many states for determining Medicaid eligibility.⁸ The principle is to leave the healthy spouse who remains in the community with enough resources and assets to support an adequate standard of living.

Distribution. Exhibit 3 presents the distribution of our indicator for resources available for catastrophic events. We divide the elderly population into three categories that characterize their ability to handle costs.

1. The Financially Independent: This group represents the market for expensive discretionary health and preventive services, and should be able to afford most long-term care episodes with current income and savings. Their available resources would cover a two-and-a-half-year-stay in most nursing homes in the country.⁹
   
2. The Tweeners: These people often spend down to Medicaid levels if they have a catastrophic health or long-term care need but could have afforded private long-term care coverage if they had been encouraged to purchase it during their working years.¹⁰
   
3. The Medicaid Bound: Generally, this group does not have discretionary resources for catastrophic events and will need to depend on public programs.
   

Adjustments for inflation. When categorizing the elderly into these three subgroups, we increased the resource categories by 20 percent in 2015 and by 40 percent in 2030 to account for the expectation that health and long-term care prices will rise faster than the general inflation rate. The 40 percent adjustment reflects an average 1.12 percent annual increase in health and long-term care prices in excess of inflation over the thirty years.

Projections. The projections in Exhibit 3 are positive, as would be expected from the estimates of growing income and assets. Perhaps most striking is the decrease in the Medicaid Bound in 2030 compared with 2000. Symmetrically, the percentage in the Financially Independent grouping, most able to take care of themselves with out-of-pocket resources, increases from 27 percent to 38 percent. The Tweener category (people at high risk of someday needing public assistance for services) actually increases slightly from 2000 to 2030.

The oldest-old, disabled, and single elderly. One problem with the estimates for the entire elderly population is that most elderly are not at high risk of a catastrophic health event until they are in their seventies or eighties.¹¹ In addition, elderly people who are single generally are at high risk of needing services in later years.¹²

The patterns over time presented in Exhibit 4 mirror the patterns reported in earlier exhibits: The population age seventy-five and older, the disabled elderly, and the single elderly are much more likely to be Financially Independent and much less likely to be Medicaid Bound in 2030 than in 2000. However, all three of these subpopulations have smaller percentages of Financially Independent members in both years than the entire elderly population or the married elderly population.

Top Data And Methods Study Results Discussion And Policy... NOTES

Without major changes in the way we pay for health care, however, multiple-tier medicine will become more and more pronounced. There will be a substantial market among the elderly for high-tech, expensive, discretionary care. It seems unlikely—given current political and social dynamics—that all or even most of these new interventions will be covered by insurance, especially Medicaid.

The data also show promise, but some challenges, for the public sector concerned about the future growth of Medicaid costs. The Medicaid Bound will decrease substantially in percentage terms. Of course, given the doubling of the absolute number of elders, Medicaid can expect slightly higher numbers of eligible elderly in 2030 than in 2000.¹³

The Medicaid Bound, of course, are at risk in another way: They likely will remain dependent on the ups and downs of publicly financed health care. How generous the government can be to this group will depend in part on who bears the burden of care for the Tweeners. It is the interplay of personal and social responsibilities for the care of these two large subsets of the population that will define philosophical and political debates about health care financing over the next thirty years.

The Medicaid program faces additional, large potential liabilities if the Tweeners are not urged or forced to put money aside during their working years to pay for expected service costs in their retirement years. The Tweeners represent an important population that could only hope to afford long-term care insurance if they purchase it in their forties or fifties, when premium rates are at low levels. If many of the Tweeners do not find ways to avoid Medicaid, the total costs of Medicaid will likely grow quickly over the next thirty years.

It is the Tweeners who face the greatest risks of being left behind in a multiple-tier medical system. This is the group that most needs to be educated about the complexity of long-term care financing and about their lack of current coverage for long-term care. Educational efforts should emphasize to the Tweeners (and the elderly in general) that retirement resource planning needs to consider retaining some assets for service needs or the use of some assets to buy insurance for expensive items such as long-term care services.

Important factors in predicting change. To assess the believability of the relatively positive simulation findings, it is important to get "inside" the model to understand which factors drive the positive findings. Five factors seem to be most important: higher wages during working years, higher education levels, higher women’s labor-force participation, more dual-income families, and the maturation of the pension system.

In most years since 1960, wages have grown more quickly than inflation, and this translates to more wealth, greater likelihood of pension income, and greater likelihood of receiving higher Social Security payments for future elderly. Some of this real wage growth is undoubtedly attributable to the higher education levels of baby boomers compared with earlier generations: Close to 90 percent of baby boomers graduated from high school, compared with 66 percent of the current elderly cohort, and college education also has increased dramatically.¹⁴

The future elderly will see their wealth and retirement income increase in part because of greater female labor-force participation and the associated increase in families with two income earners. In 1950 just over 30 percent of women were in the labor force, compared with near 60 percent in 2000.¹⁵ Dual-income families increased from about 25 percent in 1960 to 37 percent in 2000.¹⁶

While pension coverage levels and perhaps generosity have stagnated in the private sector over the past twenty years, many more elderly of the baby-boom generation will have worked most of their careers under meaningful pension plans than today’s elderly will have done. The current stagnation of pension benefits probably will affect the wealth of elders of the second half of the twenty-first century more than it will the elderly of the next thirty years.

Factors that could threaten future wealth patterns. Forecasts include uncertainty and generally synthesize what might occur if life goes on much the way it has in past years. Five key changes could make our forecasts far less positive in the future.

1. Medical care or long-term care costs could grow much more quickly than inflation over the next thirty years. If higher costs lead to cutbacks in Medicare funding and out-of-pocket costs increase dramatically, then wealth and retirement income could quickly become more constrained.
   
2. The economy could enter a long-term slowdown, resulting in wages’ growing more slowly than inflation and health costs. While this has rarely happened over the past forty years (really only in the 1970s), some observers fear that the current recession could signal more enduring economic problems than are typical in our cyclical economy.¹⁷ Employment opportunities also could erode, especially for the cohort ages 50–70. If combined with a decline in equity markets in the years just before retirement, this could greatly affect total retirement wealth and income.
   
3. Government policies could change economic factors related to retiring. These changes could include extending retirement ages beyond what is already scheduled; changing spousal impoverishment rules, which might leave less income and assets for surviving spouses; and changing Medicare and Medicaid benefits.
   
4. Pension plans could continue to become less generous. While most pension policies affecting baby boomers are pretty much set by now, unexpected pressures could lead to unexpected changes in the way that pensions get paid over the next thirty years. Social Security payments, which represent 40 percent of retirement income for the average person, also could increase more slowly than expected if current laws and political pressures change.
   
5. Elderly baby boomers could spend their retirement wealth more quickly than projected, leaving fewer resources for service needs at the end of life. The simulation model assumes that some people spend part of their wealth each year of retirement; some live only on interest, pensions, and Social Security; and some continue to save in retirement years. On balance, the model projects a slight net dissaving rate among the elderly in the early years of retirement. Consumption-oriented baby boomers could spend retirement wealth more quickly than earlier generations did, which would result in fewer resources’ being available for medical and long-term care needs.¹⁸
   

The current dynamics for elderly baby boomers are promising. If the economy grows at a moderate rate and if public and private forces keep health prices somewhat in line with general inflation, the outlook for private wealth as a source of resources to handle a large share of service costs in retirement years is positive: The elderly will be much wealthier and better able to handle health-related financial shocks in 2030 than they were in 2000.

Editor's Notes

 
James Knickman is vice-president of research and evaluation at the Robert Wood Johnson Foundation (RWJF). Kelly Hunt is a researcher in the RWJF Research and Evaluation Unit. Emily Snell is a doctoral candidate at Northwestern University. Lisa Alecxih is vice-president of the Lewin Group’s Center on Long Term Care. David Kennell is president of Kennell and Associates Inc., a public policy analysis firm, based in Falls Church, Virginia.

The authors thank Steven Schroeder, Robert Hughes, and Stephen Isaacs for their advice and suggestions. The interpretations and opinions are the authors’ and may not necessarily reflect those of the Robert Wood Johnson Foundation, Northwestern University, the Lewin Group, or Kennell and Associates.

NOTES

Top Data And Methods Study Results Discussion And Policy... NOTES

 

1. Henry J. Kaiser Family Foundation, "The Medicare Program: Medicare at a Glance," October 2002, www.kff.org/content/archive/1066/1066-04v2.pdf (12 November 2002); J. Feder, H.L. Komisar, and M. Niefeld, "Long-Term Care in the United States: An Overview," Health Affairs (May/June 2000): 40–56; J.A. Poisal and G.S. Chulis, "Medicare Beneficiaries and Drug Coverage," Health Affairs (Mar/Apr 2000): 248–256; and M.A. Laschober, et al., "Trends in Medicare Supplemental Insurance and Prescription Drug Coverage, 1996–1999," 27 February 2002, www.healthaffairs.org/WebExclusives/Laschober_Web_Excl_022702.htm (12 November 2002).
2. U.E. Reinhardt, "Turning Our Gaze from Bread and Circus Games," Health Affairs (Spring 1995): 33–36; and U.E. Reinhardt, "A Social Contract for Twenty-first Century Health Care: Three-Tier Health Care with Bounty Hunting," Health Economics 5, no. 6 (1996): 479–499.[Medline]
3. N. Gilbert and K.G. Troitszch, Simulation for the Social Scientist (Buckingham: Open University Press, 1999).
4. J.M. Wiener, L.H. Illston, and R.J. Hanley, Sharing the Burden (Washington: Brookings Institution, 1994).
5. U.S. Census Bureau, March 1994 Current Population Survey (Washington: Census Bureau, 1994). These data were chosen because an exact match to the Social Security earnings histories of those in the March 1994 sample was available. The use of actual earnings histories to project future earnings and Social Security income means that the model has significant advantage over those that only have information for a point in time.
6. Social Security Administration, 1999 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds 1999, www.ssa.gov/pub/oact/tr99.pdf (21 February 2003). This report is the basis for projections of future flows of money into and out of the Social Security system. The Intermediate Scenario assumes that wages will grow annually over the thirty-year period approximately one percentage point more than inflation.
7. J.R. Knickman and E.K. Snell, "The 2030 Problem: Caring for Aging Baby Boomers," Health Services Research 34, no. 4 (2002): 849–884.
8. Ibid.
9. G.S. Gabrel, An Overview of Nursing Home Facilities: Data from the 1997 National Nursing Home Survey, Advance Data from Vital and Health Statistics, no. 311 (Hyattsville, Md.: National Center for Health Statistics, 2000); and G.S. Gabrel, Characteristics of Elderly Nursing Home Current Residents and Discharges: Data from the 1997 National Nursing Home Survey, Advance Data from Vital and Health Statistics, no. 312 (Hyattsville, Md.: NCHS, 2000).
10. T. Smeeding, "Nonmoney Income and the Elderly: The Case of the ‘Tweeners’," Journal of Policy Analysis and Management 5, no. 4 (1986): 707–724.
11. K.G. Manton, L. Corder, and E. Stallard, "Chronic Disability Trends in Elderly United States Populations: 1982–1994," Proceedings of the National Academy of Sciences 94, no. 6 (1997): 2593–2598.[Abstract/Free Full Text]
12. K. McGarry and R.F. Schoeni, "Widow Poverty and Out-of-Pocket Medical Expenditures at the End of Life" (Unpublished paper, University of California, Los Angeles, and University of Michigan, 18 October 2001).
13. Because Medicaid eligibility rules are different from the criteria for the Medicaid Bound in the analysis, the model’s simulation of Medicaid long-term care users is higher.
14. U.S. Bureau of the Census, "Years of School Completed by People Age 25 and Older, by Age and Sex: Selected Years 1940 to 2000," Table A-1, 19 December 2000, www.census.gov/population/socdemo/education/tableA-1.txt (21 February 2003).
15. U.S. Bureau of Labor Statistics, Civilian Employment, Series LFS1600002, www.bls.gov/webapps/legacy/cpsatab1.htm (interactive site; specify the data series).
16. Authors’ calculations from U.S. Bureau of the Census, Current Population Survey, Annual Demographic Supplements, "Type of Family (All Races) by Median and Mean Income: 1947 to 2001," Table F-7, 30 September 2002, www.census.gov/hhes/income/histinc/f07.html (11 March 2003); U.S. Bureau of the Census, Current Population Reports, "America’s Families and Living Arrangements: March 2000," Series P20-537, June 2001, www.census.gov/prod/2001pubs/p20-537.pdf (11 March 2003); and U.S. of the Census, Current Population Reports, "Married Couples by Labor Force Status of Spouses: 1986 to Present," Table MC-1, 29 June 2001, www.census.gov/population/socdemo/hh-fam/tabMC-1.txt (11 March 2003).
17. For example, the High Cost assumptions used by the Social Security trustees reflect much lower real-wage growth assumptions. SSA, 1999 Annual Report.
18. Alternative assumptions about dissavings have been modeled as sensitivity analyses, and, as expected, if dissaving becomes the norm in earlier years of retirement, the simulation model predicts fewer Financially Independent elderly and more Medicaid Bound than reported in Exhibits 3 and 4. However, the general pattern of increased wealth and income presented in the paper—although more modest than reported—remains the same under all reasonable assumptions about dissavings in retirement years.

                      What's this?