Aging of the Elderly: An Intragenerational Funding Approach to Long-term Care

4.1. Intragenerational Models

There are two generic approaches to intragenerational funding models considered here. Neither of these is intended to replace the existing PAYGO programs; by supplementing these programs, these models offer the possibility that more of the costs are actually carried by the old as a group:
•    Model 1. The old as a group pay for their long-term care insurance by way of a reduced state pension.
•    Model 2. The old as a group purchase an additional annuity to supplement the state pension that has a long-term care add-on insurance component.
The first approach would be designed so that low-income retirees who are not able to afford to pay for any insurance product are fully subsidized as they are now by the state. Middle-income groups, however, might be attracted by a reduced rate of state pension that would be exchanged for the state meeting part of future long-term care costs. We consider below how this might work in the United States.
The second approach is designed specifically for middle income groups as a life annuity plus long-term care insurance purchased with a single premium at 65 or 70. This could be made attractive and might capture a wide pool of annuitants. Those who die early and do not need care, along with those who live into old age but do not need long-term care (the vast majority of those who survive), subsidize the ones who need care. The younger the age of purchase and the greater the numbers who purchase, the more the risk would be shared. Those whose health status makes them poor risks for long-term care insurance are good risks for life annuities, so that linking the two risks is likely to increase long-term care coverage of the population and reduce the need for medical underwriting, and adverse selection in the annuities market.
Surprisingly there has been little literature to date devoted to exploring the potential of pooling risks of longevity (requiring lifetime annuities) with the risk of needing long-term care. Murtaugh, Spillman & Warshawsky (2001) propose a method for linking the two risks in a single product in a voluntary market that has the potential to be cheaper by reducing adverse selection, and provide coverage for more people. This theme is developed in a recent contribution where it is argued that the combination of a life annuity and long-term care insurance “…has the potential to make them available to a broader range of the population, with minimal underwriting and at lower cost” (Warshawsky, Spillman, & Murtaugh, 2002).
There is some interest from some annuity providers emerging worldwide. For example, preliminary modeling for the UK by actuarial consultants Watson Wyatt Worldwide shows that worthwhile income increases could be paid once long-term care became necessary for modest reductions in the initial annuity. They see the demand for purchases for such annuities arising later in retirement, at above 70 years (Watson Wyatt Worldwide, 2002).
There are several issues to consider in designing a life annuity that includes a rider for long-term care insurance.
•    The age at which the policy is to be purchased. The role of deferral of purchase.
•    The nature of the costs to be covered: the policy may either indemnify the actual costs or pay a specified amount for an assessed condition. For the latter, once the highest level of dependency is diagnosed, the annuity increases by a given factor regardless of the nature of the care chosen.
•    The size of policy and whether maximums should apply. This may be important if there are significant subsidies or government guarantees to this product.
•    The kind of inflation adjustment that applies and who pays for it.
•    The source of the purchase price. Can it include home equity and if so on what basis?
It is likely that any advanced annuity product such as envisaged here would not be forthcoming except with a strong involvement from the state. The unlocking of home equity in helping to pay for this kind of annuity may make it very attractive. A possible way this might work is considered for New Zealand.

4.2. Applying Model 1 in the United States

As earlier pointed out, in the United States, funding for these services relies more on personal (out-of-pocket) payment and public welfare (Medicaid) but less on social insurance and private insurance (Chen, 2001; Johnson & Uccello, 2005). This method is akin to sitting on a two-legged stool, which is unlikely to be stable at best and unsustainable at worst, because it tends to impoverish many people and thereby severely strains Medicaid budgets nationwide.
Out-of-pocket payment — sometimes called self-insurance — fails to use the insurance principle of pooling risks. Self-insurance, by definition, is assuming the risk by oneself, rather than with others in a large group of persons exposed to the same type of risk. Medicaid is a joint federal and state program that provides medical assistance for the poor and disabled.
The program is administered by the states with federal matching grants for a portion of the cost of medical benefits, solely from general revenues. Some analysts have regarded Medicaid as a public insurance program, but it is not insurance because it lacks risk pooling. Labeling Medicaid — a welfare program — as insurance appears to use the term in a vernacular sense (“something to fall back on”), rather than in its actuarial sense, in terms of risk pooling among a large number of persons exposed to the same type of risk (Chen, 2005).
It is possible to propose a new model in which social insurance and private insurance will pay for the bulk of the costs, supplemented by personal payments. Medicaid would be used for the truly poor, its original purpose. It may be called a “three-legged-stool” funding model (Chen, 1993; Chen, 2005).
Since new public and private resources for long-term care seem scarce, the proposed model suggests using our existing resources more efficiently by trading resources dedicated for one purpose for another purpose. It may be called the “trade-off principle.”
Applying the trade-off principle in the public sector, we could divert, say, 5% of a retiree’s Social Security cash benefits (not payroll taxes) to fund a social insurance program that provides basic long-term care. This may be called a “Social Security/Long-Term Care (SS/LTC) Plan.” With this plan, retirees themselves are trading some income protection for some long-term care protection. The trade-off would enhance a retiree’s total economic security. Low-income beneficiaries, though covered by the program, will be exempt from the trade-off. This program could pay for one year of nursing home or two years of home care (Chen, 1993).
Since the social insurance program would provide the basic coverage, private long-term care insurance would cost less than it does now and thus become more affordable to more people. The visibility of the SS/LTC plan could, in addition, serve as a catalyst to increase awareness of the need to prepare for long-term care. And people would finance additional care out of pocket.
Applying the trade-off principle in the private sector, we could purchase private long term care insurance by linking it to life insurance policies or annuities; to employment-based pensions; to personal savings such as 401(k) plans or IRAs; and/or to home equity conversion products like reverse mortgages.
The trade-off principle is already being used in the private sector. For example, a person could buy an insurance policy that combines life insurance and long-term care, which pays for long-term care expenses, if needed, by commensurately reducing life insurance benefits. Although available, this type of combination policy is not very popular. Perhaps there is a role for the government to encourage it.
The three-legged-stool funding model may be regarded as a policy approach that would simultaneously foster self-reliance (by means of private insurance and personal payment) and collective assistance (in the form of social insurance). Participation in the SS/LTC plan could be mandatory with an opting-out provision. That is, upon receipt of Social Security retirement benefits, people would be enrolled into the SS/LTC plan automatically, but they may opt out of it within a specified time frame. Once opted out, an individual may not opt in again, however. Or, to avoid adverse selection against the state, people may be given a one-time opportunity to join SS/LTC plan, for example, at age 62, the earliest age eligible for reduced Social Security benefits.
Some may feel that the risk pool for SS/LTC, based on the current cohorts of Social Security recipients, might not be sufficiently large. However, as the older generation grows in number in the next decades, the risk pool they compose will enlarge commensurately, making it a more viable group for pooling risks.

4.3. Applying Model 2 in New Zealand

Until 2007 New Zealand was in a unique position to offer limited subsidies for a combined life annuity/ long-term care product given that there had been no expensive and regressive tax-driven subsidies to the accumulation phases of retirement saving. In 2007, a new saving scheme, KiwiSaver, began with generous subsides. Over time the lump sums generated by this scheme at age 65 will grow and it is very important that some thought is given to how these lump-sums are used (St John, 2009).
The current married rate of the New Zealand state pension is about $NZ 15,000 net for a single person. This, together with another $30,000 is sufficient to meet the annual capped fee cost of long-term care. Suppose a retiree’s private saving, including Kiwisaver, is used to buy an inflation-adjusted annuity of $10,000,pa, an insurance rider could provide that this annuity would treble on the diagnosis of needing long-term care (St John, 2005) Based on the probabilities of needing care, a purely actuarial calculation at age 65 (averaged for male and female) assuming a real rate of interest of 2%, suggests an annual inflation-adjusted premium of about $500 or equivalently an additional capital sum for the annuity purchase of $11,500 (St John, 2004a).
This amount is tentative and based on a purely actuarial costing, but compares favorably with the costs of setting up and running trusts to hide assets to avoid the asset test for long-term care.
It must be emphasized the $500 pa or the $11,500 single premium does not itself purchase full coverage. The insurance operates on the original annuity, so that the costs of long-term care $45,000 would be paid for from $15,000 net state pension plus $30,000 enhanced annuity.
This voluntary option could be offered from 2011 to the band aged 65-74 with the state operating the scheme as social insurance. The implicit premium for long-term care in the annuity provision could be used to help pay for the current costs of long-term care (and thus be a PAYGO scheme) or used to build a trust fund to be drawn on later. Figure 1 shows how the numbers in the younger “old” population are expected to rise.

Figure 1: Population aged 65-74

Source: Statistics New Zealand (Statistics New Zealand, 2004).

Under current medium assumption projections there will be around 316,000 aged 65-74 by 2011. Assuming that one half of the top six deciles use their cash saving, perhaps with a home equity share to buy a capped inflation-adjusted annuity of $10,000, a sizeable fund of more than NZ $1 billion could be generated in 2011.9 From this, the state social insurance programme would pay an annual annuity of $10,000 to the annuitant, and a further $500 each year to an earmarked long-term care fund. To encourage participation, the asset and income test for long-term care should be strengthened by tightening up on ways to avoid the test through creation of family trusts.
As successive cohorts enter retirement, there will be a growing number of people providing these long-term care contributions, and some of the funds could be applied on a PAYGO basis to care for the existing growing frail population.

5. Conclusion

Since older people, as a group, have in past decades improved their income and wealth positions, they as a group appear more able to pay for some of the support they need during old age. Further, financial ability of older people could also be expected to increase from continued work, part-time or full-time, owing to better health for at least some members in this group.
It is possible to think past the old models of social insurance, which impose costs directly on the working-age population, and the old models of private annuities and private long-term care insurance, which are not working well if at all. Concern about intergenerational equity is likely to become an increasingly important issue as the population profile of each country begins to change rapidly in the next decades. Intragenerational risk sharing may lessen concerns about possible intergenerational conflicts because the support for the older generation will fall more on older persons themselves.
The two models considered here in particular would shift the risks in the retired generation itself from those who live longer and need income over a longer period to those who do not live as long, and from those who are less healthy (or more dependent) to those who are healthier (or less dependent). A combined private/public insurance approach is needed, recognizing the limitations of both pure privately-funded insurance and pay-as-you-go social insurance funded by payroll taxes.
By encouraging the older age group to fund more insurance needs themselves, more resources may be freed to meet the increased demands of an aging population. It is suggested that such intragenerational risk sharing can improve both the perceptions and the reality of intergenerational equity. And it bears emphasis that the intragenerational funding is suggested to supplement, but not replace, the intergenerational funding already in use.

9 These tentative calculations are based on St John (2004a) where the costs for a man and woman of an inflation adjusted $10,000 annuity assuming 2% real interest is estimated to be $136,000 and $156,000 respectively with long-tern care rider and $142,000 and $173,000 with the long-term care rider.

Pages: 1 2 3

Tags: , , , ,