Long-term-care insurance is not a very popular product. It can require many years of expensive premium payments, and any care that is provided may not occur until very late in life. While the easy choice may be to not buy insurance, that may not be the best choice.

Although long-term-care insurance has been available for more than 30 years, only about 10% of the elderly population is covered. Reasons include:

* Premium expense.For an individual, the cost of comprehensive coverage can run more than $3,000 a year, depending on age. Couples can get discounts, but total premiums may still run close to double those for individuals. Costs are also uncertain. Last year, for example, John Hancock, a major long-term-care insurer, asked regulators for premium increases averaging 40% on existing policies.

* Lack of need. Many individuals never need significant amounts of long-term care. Although it's been estimated that two-thirds of 65-year-olds will have a need, less than half will need care for 90 days or more. In many policies, the insured pays for the first 90 days of care before benefits begin.

* Government help. For those without sufficient assets to pay for long-term care, Medicaid provides support for many services. The Medicare program also pays some long-term-care costs, but only on a limited basis. Last year's federal health care overhaul contains a provision known as the Community Living Assistance Services and Supports Act that's expected to provide modest benefits to those who enroll and pay required premiums.

* Advisor reluctance.Long-term-care insurance is complex and filled with options. Many advisors lack the knowledge to help clients choose among them, or even to recommend an appropriate specialist.

The most common strategies for long-term-care planning involve either trying to build up enough savings to meet future needs, or hoping that family support or government help will be available if needed. It's not easy to commit to years of sizable premium payments for a vague and distant need.

But the risks should not be ignored. The potential impact of a serious long-term-care need can be devastating financially, and insurance sales presentations often focus on such possibilities.

For example, the illustration might show the potential cost if one member of a couple needs to spend three years in a nursing home. Such a scenario is both plausible and worth worrying about.

MODELING COVERAGE

Yet the problem with such illustrations is that they do not indicate the likelihood of needing long-term care as well as the amounts of care that might be required.

Yet the problem with such illustrations is that they do not indicate the likelihood of needing long-term care as well of the amounts of care that might be required. Consider the range of possible outcomes using an example of a 55-year-old purchasing long-term-care coverage with a 90-day elimination period before benefits begin, a benefit of $225 a day, a five-year benefit period and inflation protection.

The cost of such a policy is estimated at $3,400 a year. This policy cost may seem high compared with some advertised rates, but it is based on an individual in average (not excellent) health with a daily benefit set high enough to likely provide full reimbursement of projected costs.

For expenses, there are four scenarios with claim periods ranging from zero to five years. Cost levels assume a mix of home care and facilities care, with claim costs averaging $57,000 a year in current dollars and annual cost increases of 3.5%.

The individual is assumed to live to 85; any long-term-care costs are incurred in the last years of life. By 55, this individual is assumed to have $250,000 in savings, and a further assumption is that income sources (from work, Social Security and pensions) will be just enough to cover regular living expenses, including any needs for short-term care lasting less than 90 days. In the absence of additional expenses, the $250,000 will grow and pass to heirs as a bequest at death. The level of long-term-care expenses affects the bequest values with no insurance.

For scenarios with insurance, coverage is assumed sufficient to pay for any expenses beyond the 90-day elimination period. The probability assumptions applying to the different scenarios are key to this analysis.

Advisors could use the gains and probabilities to do a rough cost-benefit analysis. In this case, there's a 60% chance of incurring a $60,000 loss-the present value of the premiums by purchasing the insurance.

But such a loss does need to be weighed against amounts and probabilities of potential gains. On average, an individual expects to lose money ($12,000, by this estimate) by buying long-term care insurance because insurers strive to cover risks and expenses, and make money.

INSURANCE VS. INVESTMENT RISK

It can be difficult to weigh different probabilities and loss amounts, so it would be useful if there were a way to condense results into a single measure. One way to do this is to relate the long-term-care risk to the more familiar concept of investment risk.

Advisors can begin by making the artificial assumption that the individual in the example takes a super-safe approach to all risks by buying long-term-care insurance and investing all $250,000 of the savings in risk-free assets. Under this scenario, there is no variability in bequest values and the expected gain from dropping coverage is estimated.

Standard deviation is used for the present value of bequests as the variability measure. Variability also is estimated for a scenario where, instead of dropping coverage, this individual achieves the same increase in expected financial gains by investing some savings in stocks.

The expected gain from dropping coverage is $12,000, which results in the standard deviation of bequest outcomes rising to $78,000 from zero. This individual could achieve this same estimated gain of $12,000 by investing about 5% of the $250,000 in stocks.

However, investing this much in stocks produces far less bequest variability than dropping coverage - $13,000 versus $78,000. For this particular case, going without long-term- care insurance carries six times the risk of achieving the same expected gain by taking some investment risk. This result depends on numerous assumptions, but some testing indicates the differential will be substantial under any reasonable assumptions.

THE IMPLICATIONS

As the data shows, long-term-care insurance is worth considering, and the following points are worth keeping in mind:

* Insurance may be more valuable for couples than for individuals because the financial impact will fall on the spouse or partner before affecting bequest values.

* If bequest values are an important consideration, insurance can help reduce risk by narrowing the range of possible outcomes.

* Even if family considerations or bequest values are not major concerns, having insurance can reduce the need to set aside a large amount of retirement savings for the possibility of a significant long-term-care need.

* Those concerned about managing investment risk need to recognize that self-insuring for long-term care carries an even bigger risk than equity market risk.

* Those who hope to use financial or gifting strategies to be able to rely on Medicaid need to recognize that the program, which currently limits choice and services, will probably become even more constrained because of fiscal pressures.

KEY CONCERNS

What about substantial rate increases? Given that some companies have requested premium increases of up to 40%, there's certainly cause for concern about future jumps. However, the past few years have been a perfect storm for the industry.

Interest rates have fallen, lapses have decreased, longevity has improved, care costs have increased faster than general inflation and claims have increased in both frequency and duration. New product pricing reflects these factors, so there should be less need for premium increases going forward.

To the extent that any future premium increases reflect higher long-term-care costs, those costs will affect both insured and uninsured individuals. Clients cannot escape increases in the cost of care by not buying coverage.

Won't the Community Living Assistance Services and Supports Act replace the need for long-term-care insurance? No. The provisions of the act are still in the design stage, but the plan is only to provide benefits of $50 to $75 a day-way less than costs that may run $250 a day or more. Also, the law aims to be both self-sustaining and open to everyone regardless of health status, so the possibility looms of higher premiums for coverage than for private insurance that's underwritten.

Why not buy catastrophic long-term-care insurance to cover only long-duration claims? It might make sense, based on general insurance principles, to self-insure any claims of up to, say, two years, and buy less expensive coverage solely for claims of a longer duration.

The problem with this approach is that insurers are concerned about the financial risks of offering open-ended benefit periods, so policies with especially long elimination periods tend to be steeply priced, if offered at all. It certainly makes sense to keep premium costs down by opting for a 90-day elimination period, but it may not be worth choosing policies with longer elimination periods.

What about hybrid policies as a lower-cost alternative? A few companies are beginning to offer policies combining coverage and investments, where benefits do not kick in until investment funds are depleted.

Such products do offer consumers a form of catastrophic coverage. A key consideration will be the level of expense charges on the investments bundled into these products compared with alternatives.

It requires a big investment of time for advisors to learn enough about long-term-care insurance to be helpful to clients. It may seem easy to avoid the effort and recommend that clients self-insure. But those advisors won't likely assess the overall risks faced by their clients. 

--This story first appeared in Financial Planning magazine.