This is a question I’ve been thinking about lately while working on another project, and I want to walk through my thinking so far to see if any readers want to provide any additional insight.
Long-term care is very expensive, but not everybody needs it
Due to search-engine optimization and paid ad placement this information is actually annoyingly difficult to find (by which I mean I had to click to the second page of search results), but according to the federal Administration for Community Living, the average cost of a private room in a nursing home nationally in 2016 was $7,698 per month, and the average cost of care in an assisted-living facility is $3,628 per month. This is obviously a lot of money, between $43,536 and $92,376 per year depending on your state and the intensity of the care you need.
That is, if you need it, which not everybody does. This is the fundamental logic behind many kinds of insurance: when it is certain that an unknown future subset of a population will face a ruinous economic blow, the entire population can pool their assets in advance with the understanding that some of the population will “waste” their premia, some will “make it back,” and some will receive a benefit far out of proportion to their payments. The entire population or subset of the population (coworkers, union members, age cohorts, etc.) pays the same amount for the same certainty of protection should they be the one to suffer.
It seems to me people often get confused on this point. For example, it makes no sense to sell “primary school insurance” to pregnant women or new parents, since virtually all newborns in the United States attend primary school. Buying insurance “just in case” your children attend primary school would be absurd, since almost all the insured would be receiving their own money back — there would be no pooling of risk, and the downside of a costly new level of administration. Parents who want to make sure they have money for school supplies just save money for school supplies, and everybody else takes their chances having to pay for pens and pencils out of their current earnings or from charity.
Long-term care insurance is expensive too
On the same federal site, I learned that in 2007 (yes, I’m also annoyed at the fact this datapoint is from a decade earlier than the one above, but it is what it is), the average long-term care insurance policy cost $2,207 per year and provided $160 in daily benefits for a maximum of 4.8 years. Note that once you have a qualifying event and begin to receive benefits, in most cases you are no longer required to pay a premium.
This makes it easy to calculate the maximum value of the average long-term care insurance policy in 2007: $280,320. Of course, since this is an insurance product, only a small number of customers will receive the maximum benefit. Some will allow their policies to lapse before they need care, some will die without needing long-term care (not necessarily by dying young — they might simply die a natural death in old age while still able to care for themselves), and some will end up needing long-term care, but for less than the maximum term, or at a lower cost than the maximum benefit.
Does long-term care insurance really “protect your savings?”
Long-term care insurance is often sold as a way for wealthy people to “shield” savings they may wish to leave to a spouse or heirs, out of fear that a few years of expensive care at the end of their lives will drain away their estate leaving their family or favorite charitable causes with nothing.
That’s all well and good — after all, insurance isn’t supposed to be an investment with a predictable rate of return, it’s about buying piece of mind. Long-term care insurance, however, is such a novel product that it doesn’t seem to me to have undergone the kind of scrutiny that older products like life insurance have. It’s a cliche, for example, that it’s better to buy term life insurance to protect your dependents, rather than the much more expensive whole life insurance, and invest the cost difference in assets that are likely to outperform the piddling interest rates the savings component of whole life insurance offers.
A similar principle seems to apply in the case of long-term care, for two reasons. First, people buy long-term care policies far too late in life, typically in their 50’s or 60’s, which means they pay very high premia. According to the American Association for Long-Term Care Insurance, the average time elapsed between policy purchase date and eligibility date is just 14 years. Think about it this way: people self-insure for long-term care for the first 50 or 60 years of life, despite having lower earnings and assets and longer possible time horizon of care, and then overpay for insurance when they have the most money and shortest life expectancy.
Second, framing the costs of long-term care as requiring you to “spend down” your savings ignores the fact that savings and investments also produce income which can be used to pay for the cost of care. If your care costs $67,956 per year (the average of an assisted living facility and a nursing home), and you receive the average Social Security old age benefit (in 2015) of $16,536, then you need to cover the difference of $51,420 per year. But that expenditure will not reduce your assets by that amount. Instead, it will reduce your assets only by the amount not covered by income generated by your assets. Assuming a permanent (extremely conservative) 3% rate of return over your working life, retirement, and period of care, you would need to accumulate $1.7 million in order to generate that difference purely out of income indefinitely, that is to say, without spending down your assets at all, with no long-term care policy. Moreover, since most long-term care policies have maximum benefit amounts, the more you spend on long-term care insurance, the more exposed you are to the risk of outliving the maximum benefit and having inadequate savings to cover ongoing care.
In other words, you need insurance during your working years while aggressively saving, in case your savings plan is interrupted before you’re able to reach the point you can self-finance your care indefinitely.
The private insurance market is getting in the way
The problem, of course, is that while all of the above is true, it’s also impossible. If you’re young, getting the kind of long-term care policy I’m describing is in fact prohibitively expensive! That’s for two reasons. First, so few young people do buy long-term care policies that insurance companies have too little data to price plans appropriately, and naturally err on the side of expensive. Second, insurance companies are acutely aware of the risk of adverse selection: why would a young person buy a long-term care policy if they didn’t at least suspect they were likely to use it?
Group plans through an employer are an option for folks whose employers offer them, but obviously it’s up to the employer whether to do so and what kind of plan to offer, so just because your employer offers a plan doesn’t mean it’s the best option for you. If you do buy a plan through your employer you should be able to keep both it and your original age and health rating after separating from the employer, although since you’ll no longer be paying for the plan through payroll deductions be sure to keep an eagle’s eye out for billing information to keep the plan from lapsing.
Conclusion
All of which brings me back to the point of this post: why is there a private long-term care insurance market in the first place? The obvious solution to long-term care costs is a robust public insurance program, paid for with a nominal tax on the entire working population. Young people with a small likelihood of needing long-term care could pay the same tiny percentage of their income as older people with the highest likelihood of needing care — but only if the entire population participates.
As long as the only way to buy long-term care insurance is from for-profit companies marketing to older people most likely to use it, then long-term care insurance will remain expensive, inadequate, and uncertain.
Leave a Reply