| The banner ad and navigation bar to the right is placed there by Geocities. We don't have any control over what ad is placed there. |
Revised 9/16/99 -- See fb-changes.html for what's new and a revision history.
Index Of All Web Pages (1)
Master Table of Contents (2)
Index page (fb-index)
<<###PREVIOUS(fb-81)
NEXT(fb-90)###>>
A sharp decline in the ratio of workers to dependents has occurred and is projected to continue to occur. This is a consequence of the low fertility rate (non237) and modest immigration rate (non238), and the growing life expectancy (non239); these topics were all discussed previously.
As shown in the below table, excerpted from the larger table in section non55, the number of OASDI (Social Security) covered workers per OASDI beneficiary used to be 16.5 in 1950, but now it is a much smaller 3.4. It is projected to decline from 3.4 in 2000 to 3.1 in 2010, 2.1 in 2030, and 1.9 in 2075. The precipitous drop between 2000 and 2020 is due to baby boom retirements.
-------(245)-------- Table T244.1 Year ............. 1950 1960 1980 2000 2020 2040 2060 2075 Covered Workers Per Beneficiary .. 16.5 5.1 3.2 3.4 2.5 2.0 1.9 1.8 Source: March 1999 Social Security Trustees Report (TR99) Table II.F19
Looking at it in inverse: currently, each worker is supporting 0.2941 beneficiaries (1/3.4 = 0.2941). In 2040 and beyond, each worker will be supporting about 0.5000 beneficiaries (1/2.0 = 0.5000). That is an increase from 0.2941 beneficiaries per worker to 0.5000 beneficiaries per worker, which is a 70% increase in the number of beneficiaries per worker ( {0.5000-0.2941}/0.2941 = 0.700 ). In other words, the number of beneficiaries a worker will have to support will grow by 70% over the next 40 years.
In 2075 each worker will be supporting 1/1.8 = 0.5555 beneficiaries. That is an increase from 0.2941 beneficiaries per worker today to 0.5555 beneficiaries per worker in 2075, which is a 89% increase.
It is argued by some that the demographics aren't as bad as they seem. They point out that during the baby boom, the total dependency ratio was even higher than what is expected in the future. The total dependency ratio includes dependent children as well as dependent seniors.
The below table shows dependency ratios based on the population size of three age brackets. This table is considerably different than Table T244.1 - a short form of TR99's Table II.F19 (besides being the inverse). Table T244.1 dealt with OASDI covered workers and OASDI beneficiaries. Whereas Table T247.1 below (from TR99's Table II.H1) deals with populations: those under age 20 (which I'll call "pre-adults"), those between ages 20 and 64 inclusive (which I'll call "working-age adults"), and those aged 65 and over (which I'll call "seniors"). The tables are not comparable for a number of reasons, such as {1} some working - age adults are not working and/or are dependents under the DI (Disability Insurance) program of Social Security. {2} And some seniors are both working and beneficiaries. {3} And a considerable number of "pre-adults" are working and paying into the system. And so on. However, Table T247.1 (II.H1) is useful in illustrating broad demographic trends.
Table T247.1
(From Table II.H1 of the March 1999 SS Trustees Report {TR99})
Social Security Area Population as of July 1 and
Dependency Ratios, by Alternative and Broad Age Group,
Calendar Years 1950-2075
Population (in thousands)
Dependency
Population In Thousands Ratio
---------------------------- ------------
65
Calendar Under and
Year 20 20-64 over Total Aged Total
-------- ----- ------ ------ ----- ----- -----
A B C D E F G
Historical data:
1950 53,895 92,739 12,752 159,386 0.138 0.719
1960 72,989 99,842 17,250 190,081 .173 .904
1965 80,134 104,833 19,093 204,059 .182 .947
1970 80,685 113,194 20,921 214,800 .185 .898
1975 78,438 122,862 23,266 224,566 .189 .828
1980 74,570 134,431 26,149 235,151 .195 .749
1985 73,250 144,899 29,065 247,214 .201 .706
1990 75,161 152,964 32,004 260,128 .209 .701
1995 79,037 159,812 34,281 273,130 .215 .709
Intermediate Forecast:
2000 81,571 168,509 35,383 285,463 .210 .694
2005 82,247 177,919 36,747 296,912 .207 .669
2010 82,151 186,102 39,742 307,995 .214 .655
2015 81,831 191,402 45,639 318,872 .238 .666
2020 82,514 193,580 53,013 329,106 .274 .700
2025 83,377 193,298 61,437 338,112 .318 .749
2030 83,881 193,342 68,442 345,665 .354 .788
2035 83,909 195,831 72,028 351,767 .368 .796
2040 83,903 199,470 73,271 356,643 .367 .788
2045 84,210 202,476 73,957 360,643 .365 .781
2050 84,689 204,030 75,424 364,144 .370 .785
2055 85,133 204,672 77,700 367,505 .380 .796
2060 85,435 204,828 80,678 370,941 .394 .811
2065 85,651 205,856 82,942 374,450 .403 .819
2070 85,900 207,228 84,712 377,840 .409 .823
2075 86,228 208,348 86,355 380,931 .414 .828
The Aged Dependency Ratio (column F) is
(population aged 65 and over) / (population aged 20-64)
The Total Dependency Ratio (column G) is
( population aged 65 and over + population under age 20 )
/ (population aged 20-64)
Another way to state the above Total Dependency Ratio formula is:
Total Dependency Ratio =
the number of "pre-adults" (under age 20)
PLUS
the number of "seniors" (age 65 and over)
ALL DIVIDED BY
the number of "working-age adults" (ages 20 to 64)
The below table is just the above table, but only showing the aged
dependency ratio column:
Aged
Dependency Ratio
----------------
Historical data:
1950 0.138
1960 .173
1965 .182
1970 .185
1975 .189
1980 .195
1985 .201
1990 .209
1995 .215
Intermediate Forecast:
2000 .210
2005 .207
2010 .214
2015 .238
2020 .274
2025 .318
2030 .354
2035 .368
2040 .367
2045 .365
2050 .370
2055 .380
2060 .394
2065 .403
2070 .409
2075 .414
In the below, I speak in very rough and general terms, i.e. of working - age adults supporting pre-adults and seniors, even though there are considerable exceptions such as those discussed above (e.g. there are working - age adults not working and/or are beneficiaries etc.). I do this to express this table's statistics in a similar way to how I expressed Table T244.1's statistics.
Looking at the aged dependency ratio, and very roughly speaking: currently, each working-age adult person is supporting 0.210 seniors. In 2030, each working-age adult will be supporting 0.354 beneficiaries. That is a 69% increase in the number of seniors per working-age adult. In 2075 each working-age adult will be supporting 0.414 beneficiaries. That is a 97% increase between now and 2075.
So we can see that the picture is gloomy regarding the aged dependency ratio. And that the aged dependency ratios expected in the future (0.354 in 2030, 0.414 in 2075) are much higher than what have been experienced in the past (0.138 in 1950, 0.173 in 1960, 0.185 in 1970, 0.195 in 1980, and 0.210 currently).
However, the Total Dependency Ratio (ratio of pre-adults and seniors to working-age adults) shows a much different picture.
It shows that the total dependency ratios expected in the 2030 to 2075 period, between 0.788 and 0.828, are no worse -- actually they are considerably better -- than what we've already experienced between 1960 and 1975 (0.828 to 0.904). So, if each working-age adult was able to support so many dependent - age people in the 1960-1975 period, why is there so much concern about supporting even fewer dependent - age people in the future?
The below is from the above table, but only showing the Total Dependency Ratio column:
Total
Dependency Ratio
----------------
1950 0.719
1960 .904
1965 .947
1970 .898
1975 .828
1980 .749
1985 .706
1990 .701
1995 .709
Intermediate Forecast:
2000 .694
2005 .669
2010 .655
2015 .666
2020 .700
2025 .749
2030 .788
2035 .796
2040 .788
2045 .781
2050 .785
2055 .796
2060 .811
2065 .819
2070 .823
2075 .828
Here are some ways that the cost of supporting the elderly is higher, and is perceived to be more burdensome (because it involves supporting anonymous elderly through higher taxes), than the cost of supporting one's own children:
** Most of the cost of supporting the elderly is done through involuntary taxes, while most of the cost of supporting children is done privately and voluntarily in one's own family.
** Paying higher involuntary taxes to support anonymous elderly is less gracefully accepted than spending privately and voluntarily on one's own children.
** Investing vs. Consumption: Spending on one's children is regarded by many as an investment in their children's future (and in their own future). Spending on benefits for the elderly is considered pure consumption and charity,
** Public (all levels of government) per-capita spending on the elderly is much higher than on children ((I think it was on the Concord Coalition site that I saw the figure of 3 to 1. This ratio includes local public spending on children's education)).
** Federal per-capita spending on the elderly is nearly 10 times as much as spending on children
** Total (public and private) per-capita spending on the elderly is greater than on children. Children (below 18 years of age) don't expect to live in their own homes or apartments, but the elderly do as long as they are able. And medical costs for the elderly are much higher than for children.
Here are some quotes from Gray Dawn that make some of the same points above:
[Begin quotes from Gray Dawn p. 108-111:]
Back in the 1950s, to be sure, the relative number of dependent children was a lot higher than it is today or will be in the future -- especially in the U.S. and a few other postwar "baby boom" countries such as Canada and Australia. But that was in an era in which the developed world did not socialize much of the cost of dependency. Today it does -- and it does so for the old to a much greater extent than for the young. Even including cash and in-kind benefits from all levels of government, per capita public spending on the elderly (aged 65 and over) relative to spending on the non-elderly ranges from a low of 5 to 1 in Canada to a high of 11 to 1 in the U.S. [What about the ratio of spending on the elderly to spending on children? Why did he specify the ratio of spending on the elderly to spending on the non-elderly?] Within the U.S. federal budget alone, each elder consumes on average nearly ten times as much in benefits and services as each child ($15,600 versus $1,700 in fiscal year 1995). The bottom line is that elder dependency is far more costly to public budgets than youth dependency. This fact is entirely obscured by the "total dependency" concept.
Sometimes the dependency theorists argue that what matters is total spending on dependents, private as well as public. Yet even by this broader measure, the typical old person consumes more than the typical child. Apart from the raw numbers, moreover, this measure assumes that it makes no difference to the economy whether income is transferred privately within families or publicly through government. It also implies that any income that workers spend on anyone other than themselves represents an equivalent burden -- regardless of whether the spending is a voluntary sacrifice for one's own children or a tax-and-transfer payment to some anonymous retiree. Most fundamentally, it ignores the distinction between investment and consumption. Money spent on senior benefits is essentially pure consumption. Money spent on children is an investment. In the Washington Post, Richard Leone of the Twentieth Century Fund [now The Century Fund at www.tcf.org] expresses puzzlement that Americans await the coming age wave with anxiety but didn't consider the 1960s an era of "deprivation," even though the the U.S. total dependency ratio was higher in 1960 than it will be in 2030. The difference, of course is that thirty-five years ago adults were sacrificing through families to build the future, while thirty-five years from now they will be sacrificing through government to reward the past. What happened then was investment; what is scheduled for our future is consumption.
[End quotes from Gray Dawn p. 108-111]
See also the Medicare, Medicaid and Healthcare section, non115.
"Old Old" refers to those aged 85 and over, also denoted "85+". The rapidly growing proportion of the elderly that are "old old" may not be too much of an issue with Social Security, as one's Social Security pension is the same whether one is aged 65 or 105. Well, except that if someone outlives their financial resources, they can become eligible for supplemental Social Security benefits.
But, the growing size of the old - old population certainly affects Medicare and Medicaid costs. That's one thing to keep in mind when someone -- such as Rush Limbaugh during the "Medi-scare" debate of 1995 or 1996 -- said that the Republican plan was not cutting Medicare because the plan contained an increase in benefits of 6% per eligible recipient per year. (Back then inflation was about 2-3% if I recall, so that 6% per year increase represented a real (after-inflation) increase of 3-4% per year). Well, one reason that costs could increase at more than a 3-4% per beneficiary per year real rate is because Medicare costs have been increasing at about a 5% per beneficiary per year real rate. Another reason for a higher growth rate is because of the rapid increase in the proportion of elderly that are old old. So a continuation of the historic growth rate is certainly not unreasonably pessimistic.
Here are some excerpts from the Gray Dawn book about the old old issue:
[Begin Gray Dawn excerpts:]
[85+ in millions, p.42:] In 1900, U.S. residents aged 85+ numbered a mere 374,000. Today, they number nearly 4 million, and by 2040 they will exceed 13 million.
[young-old vs. 85+ growth rate, p.42:] By the year 2040 in the U.S., the Census Bureau projects that the population aged 65 to 74 will grow by roughly 80% -- but the 85+ age group will grow by 240%.
[80+ in millions, p.43:] U.S population aged 80+, per graph, source: Census ( 1996 )
1970: 3.7 million 1995 8.1 million 2040: 26.2 million[100+ in millions, p.42:] Centenarians are projected to grow from 63,000 to over 834,000 between now and 2050, a 13-fold increase over the next 50 years.
[Life expectancy at 80, p.44:] Life Expectancy at age 80 (years of life remaining) - reading from a graph:
1950: 6.6 1970: 7.2 1990: 8.5 1995-6: 8.5[Longevity assumptions in official forecasts assume a slow-down in longevity growth. Examples of new medical treatments, p.59:] Far from overstating global aging, today's official projections may well underestimate its true future magnitude. On longevity, for example, none of the official fiscal projections assume that life expectancy will continue to rise at its historical pace (let alone speed up); instead, they all assume it will grow more slowly in the decades to come. This assumption sometimes leads to almost surreal results. In the U.S., for instance, newspaper headlines report unflagging progress against major killer diseases, such as the announcement that in the 1990s the incidence of cancer has been declining for the first time since the 1930s. Yet the Social Security Administration continues to assume that U.S. longevity at age 65 will grow 60% more slowly in the future than it has over the past quarter-century -- meaning that Americans by the year 2050 will have a life expectancy no greater than the Japanese already have today.
[SSA longevity assumptions are far lower than many others. Medical advances cited, p.60:] There is no sound basis for this slow - growth longevity prognosis. The U.S. Social Security Administration now projects that the number of Americans aged 85 and over will grow from 4 million in 1998 to 15 million by 2050. But this stunning growth trend is actually slower than the findings of most other academic demographers and government agencies (including the Census Bureau), who project from 20 to 40 million or even more in 2050. To support the higher numbers, these experts point to the historical trend as well as the recent progress on the genetic origins of illness (especially cancer) and on the biochemistry of human aging. In the 1990s, entire new industries, specializing in everything from organ replacement to gene splicing, have emerged. Firms are poised to take advantage of what some entrepreneurs are calling a "golden age" of biomedical discovery. ... it seems foolish to suppose a future in which medical advances slow down.
[Health care spending per capita on 85+ is much higher than on young old, p.45:] Total per capita health spending in the U.S. on the old old (85+) is three times as much as that on the "young-old" (aged 65 to 74). For specific types of health care, this multiple varies greatly -- from twice as much for hospital spending to over twenty times as much for nursing homes.
[Institutional care much higher for 85+ than for young old, p.45:] In virtually every developed country, the share of the elderly receiving institutional care rises from less than 5 % among those under age 75 to between 20 and 50% for those aged 85 and over. Such long-term care is extremely costly -- and now averages about $40,000 annually per person in the U.S., an amount exceeding the median household income. According to one study, the number of nursing-home residents in the U.S. will roughly double by the year 2030 while the total real cost of nursing homes will grow almost fivefold. By then, the U.S. will be spending more on nursing homes (in real dollars) than it spends on Social Security today.
[Near end - of - life health care costs, p.46:] Some experts claim that this cost growth will slacken in future decades as our "health spans" lengthen along with our lifespans and the ills of old age are relegated to a brief (and inexpensive) period of declining vigor in the final years of life. Other experts disagree. According to the "failure of success" hypothesis, the main effect of modern medicine has been to increase the number of "marginal survivors" in the population. Longer lifespans, they say, will therefore be accompanied by a rising incidence of chronic illnesses, including arthritis, osteoporosis, deafness, blindness, heart and circulatory disorders, and various forms of senile dementia, including Alzheimer's -- which afflicts 5 percent of elders under age 80 but 20 percent of elders over age 80. These chronic illnesses have been relatively resistant to cure, are relatively expensive to treat, and often require labor-intensive personal care for which there are no technological substitutes. Although it is still unclear which medical model is correct, one thing is for certain: The share of old old receiving institutional care continures to grow rapidly in most developed countries, and the cost per resident continues to climb at least as fast as worker pay.
[Medicaid pays about half of nursing home stays, p.46:] In the U.S., Medicaid pays for roughly half of all nursing home stays even while many middle-class families are bankrupted by the cost they must pay out of their own pocket.
[Medicare cost increasing 4%/year faster than wages, p.68:] In the U.S. Medicare outlays per enrollee have grown at least 4% per year faster than wages during every decade since the program began.
[Long-term care (primarily nursing home) expected to rise to 4% of GDP, p.68:] The biggest uncovered item is long-term care for elders who don't or won't qualify for public assistance. And this is a big item indeed. In the U.S., the growth in total spending on formal long-term care over the next 40 years is expected to amount to roughly 4% of GDP. Government now expects to pay for only half of this cost increase -- and provide no help for the even greater increase in informal care within ordinary families. In an aging world full of two - earner couples and short of potential caregivers, government may be forced to pay a much larger share of the long-term care burden.
[National health spending has increased by 5% of GDP in last 20 years, p.106:] Over the last two decades, a benign demographic period in America in which boomers have only progressed from youth to middle age, U.S. national health spending has already risen by a staggering 5 percent of GDP. Now imagine what will happen as they transition from middle-age to elderhood if Americans remain unwilling to tolerate any restraint on health-care spending.
[HCFA arbitrarily hopes that real percapita Medicare spending will decelerate from a 5% growth rate to a 1% growth rate, p.106:] The HCFA projects that the growth in real per capita Medicare spending will decelerate from its historical rate of 5 percent per year to just 1 percent per year by the 2020s. (See Note 9 a few lines down). HCFA does not point to any specific change in technologies or policies that might explain this cost slowdown. It simply assumes that someday, somehow, costs must be controlled if Medicare is not to consume the entire economy.
[More on percapita Medicare spending, From graph p. 107:] Real Medicare costs from 1970 to 1995 have risen on average 5%/year/percapita. The HFCA, in the "intermediate projection" assumes it will fall in a straight line to about 1.3% in 2020 and 1% in 2025 and after.
[Medicare costs to rise from 2.7% of GDP to 5.9% of GDP, or maybe 14% of GDP, Note 9 p. 251:] Even so, HFCA estimates that Medicare costs will rise from 2.7 percent of GDP today to 5.9 percent of GDP by 2030. Compare this to projections run by researchers at the employee - benefits firm of Watson Wyatt Worldwide, who assume that future per-capita cost growth will be closer to historical experience. These projections show Medicare costs rising to just over 14 percent of GDP by 2030.
[End Gray Dawn excerpts]=================================================== Graph page 61: "The number of "old old" could far exceed the official projections. Americans aged 85 and over in 2050, according to alternative projections (in millions): Population Aged 85+ Millions Projection --------- ---------------------------- 14.6 SSA Projection, 1996 (assumption used in offical U.S. fiscal projections) 18.2 Projection by Census, 1996 21.4 Projection by Ronald Lee 39.0 Projection by James Vaupel 48.7 Projection by Kenneth Manton (Ronald Lee, James Vaupel, and Kenneth Manton's research were funded by the National Institute on The Aging)===================================================
(hits since June 22, 1999)
Get your own Free Home Page