Friday, October 9, 2015

Why Retirees Let LTC Insurance Lapse

A reader recently asked my opinion on long-term care (LTC) insurance policies. My position is that many retired households, perhaps most, will not be able to afford LTC premiums and will have no decision to make. Wealthy households will be able to self-insure. That leaves the households in between with a decision to make. For those households, purchasing a long-term care insurance policy can be the lesser of two evils.

The problems with LTC insurance are well known (see An Economist Explains the Dangers of Long-Term Care Insurance). Many carriers have found the policies unprofitable and have simply gotten out of the business. According to a recent Wall Street Journal article, five of the 10 largest LTC policy sellers, including MetLife and Prudential Financial, have sharply reduced or discontinued sales entirely since 2010. Buying insurance when many insurers are abandoning the market for that insurance is risky.

Some Boomers have been encouraged to buy LTC policies because they worked well for their parents, but this is not your father’s LTC policy. Rates increased substantially after insurers realized they had initially underpriced policies and would need to raise rates substantially if they were to make a profit. You're unlikely to get the deal your parents got.

Perhaps the greatest problem with LTC insurance is the possibility that an insured retiree will let his or her policy lapse and, after making large premium payments for years or even decades, not be covered when LTC insurance is actually needed.

While insurers can't increase premiums for a specific policy, they can increase premiums for classes of policyholders. Jane Gross, a retired correspondent for the New York Times and author of the excellent blog on aging, NextAvenue, recently wrote in a post entitled, " Reasons to Worry – and Agitate – About Your Financial Security", 
"Next up is my long-term care insurance policy, which now costs $1,357.85 a year but, this letter tells me, is raising its premiums by 48 percent — unsurprising but still breath-taking.
It would be way more than that, the MetLife representative told me by phone, but for the fact that New York State has one of the nation’s most stringent insurance commissions. . .
In 2019, MetLife told me, the state commission will again allow the insurer to raise its prices. When that happens I’ll reduce my benefit duration from three years to two. Is there a point when I’d flush down the toilet 13 years of premiums already paid? I haven’t a clue."
Although the probability that a retiree will need some amount of long-term care is significant, the probability of a financially catastrophic stay in a long-term care facility is relatively small, as I described in an earlier post. Many stays will be quite brief and some can be paid out-of-pocket.

Long-term care expenses can range from informal care in the home to expensive nursing facilities, from short stays that can be paid out-of-pocket to lengthy stays with catastrophic costs.

The following table (click to enlarge) is from a study entitled, "Long-term care over an uncertain future: what can current retirees expect?" Notice that while 69% of people over age 65 required some form of long-term care, 31% required none at all. Another 29% required stays of two years or less. The scary number is the 14% that required more than 5 years of in-facility care. The odds of that happening are somewhat low, but the magnitude of the risk can be financially catastrophic and that's what we need to prepare for in some way.

Medicare does not cover most long-term care costs. Medicaid may, but only after most of the retiree's financial resources have been spent. It is intended to cover indigents. To use it for long-term care, you must become one.

A recent brief published by the Center for Retirement Research at Boston College entitled, “ Why Do People Lapse Their Long-Term Care Insurance?”, researches this important issue. According to the report, “At current lapse rates, men and women age 65 have, respectively, a 32- and 38-percent chance of lapsing prior to death, assuming that lapse rates remain at the same levels observed for recent cohorts.”

Let me repeat that. About 30% to 40% of 65-year old's will eventually allow their LTC policy to lapse, forfeiting all benefits after having paid years of premiums. Insurers bemoan low lapse rates of 1% to 2% for these policies as a justification for rate increases, suggesting they are on the hook for underpriced policies issued years ago that buyers are unwilling to relinquish. This is an annual lapse rate, however, and a 1.5% annual lapse rate over a 30-year retirement results in about a third of policies lapsing over a 30-year retirement.

The brief's other key findings are:

1. Lapses could be due to the burden of insurance premiums, a strategic calculation that care use is less likely, or poor decisions due to declining cognitive ability.

Why do retirees allow their LTC coverage to lapse? Insurance premiums could increase and render the policy unaffordable, or they could remain the same while the retiree's ability to pay declines. That doesn't mean the premiums would have been spent totally in vain, because risk was protected prior to the lapse and that has value. But it does mean that the retiree has a potentially huge uninsured risk going forward. It would be a poor outcome, indeed, to pay premiums for decades and then suffer devastating long-term care costs after the policy lapsed.

2. The analysis finds support for both the “financial burden” and “cognitive decline” explanations.

The research finds that many LTC policy lapses are the result of either premiums that the retiree can no longer afford, or the retiree losing the mental acuity needed to maintain the policy. They might forget to make premium payments or simply decide, incorrectly, that they no longer need the policy.

3. The consequences of lapsing are significant, as lapsers are actually more likely than non-lapsers to use care in the future, partly due to cognitive decline.

Interestingly, the study found that retirees who let their policies lapse are more likely to need long-term care in the future than those who don't lapse. This is partly explained by the fact than impaired retirees are more likely to let policies lapse and impaired retirees are also more likely to need long-term care.

4. Thus, for some lapsers, having insurance could be counterproductive as they buy it to protect against risk but drop it just when the risk becomes more likely.

In other words, if you’re going to allow the policy to lapse, you’re probably better off not buying it in the first place. Of course, when you buy the policy, it is probably your intention to keep it in force.

Joe Tomlinson, whose opinion I respect most on retirement insurance issues, suggests that purchasing the policies, despite their known faults, is the lesser of two evils. In AdvisorPerspectives, Joe expressed his preference for standard LTC policies over hybrid policies (life insurance or an annuity with an LTC rider) or self-insurance.

Retirement advisor, Dana Anspach, wrote a nice piece, "What Happens When You Don’t Have Long Term Care Insurance?", explaining the downside of not purchasing LTC insurance. I agree with all of it, perhaps with the exception of identifying "the worst case scenario [as] where one spouse remains healthy and retains most of the ongoing costs of living independently and the other spouse needs care in assisted living or nursing home." I think paying premiums for decades and then letting a policy lapse just prior to incurring huge end-of-life costs has a strong claim on that title.

I believe that you should forego these policies if you can’t afford the premiums, and potentially large future premium increases, or if you are wealthy enough to self-insure.

If you fall in between, I don’t believe there is a single clear winner in the numbers. There are several strong arguments in favor, and several against. Your decision will largely depend on your unique financial situation and your risk tolerance. For some, insuring the risk of catastrophic end-of-life medical expenses is worth a lot. It lets them sleep at night.

I hope that explaining the issues involved will help with your very difficult decision.


  1. It's interesting that you discuss lapse rates. In recent years, the LTC insurance companies have been citing lower-than-expected lapse rates as one justification for raising their premiums. And apparently, they have discovered that relatively few people actually abandon their policies when premiums increase, so it appears that many people who purchase LTC policies are not tightly financially contrained. This gives the LTC companies even more incentive to raise premiums down the line...

    On the issue of affordability, I currently pay just under $5K annually for a 4-year shared-care policy (currently up to $10K/mo in benefits) with 5% compound inflation protection. This policy was purchased 6 years ago (at my age 55). If I purchased the same policy today from scratch from the same company (Genworth), my insurance agent said that it would cost over $20K per year. At the latter rate, I would probably just forgo LTC entirely, even if I could afford that amount. I think everybody reaches a point -- perhaps irrational -- where the price tag is just too high regardless of the potential long-term benefits.

    1. Good points all. Especially agree with your last sentence. Worse if that decision point comes after making years of premium payments.

      I'm looking further into lapse rates. Carriers have long cited low lapse rates, but the more recent research referenced above (and not from the insurance industry) says it's quite high. I'll keep you posted.

      I believe Genworth is the largest LTC insurer, but they are also the one that I most often read is in trouble.

      Great comments. Thanks for sharing!

  2. Hi Dirk,

    I really like reading about the topic of LTC since I struggled with the decision so much before purchasing a LTC policy. In our case the choice was really the best alternative we could find at the time. My wife could not qualify for a traditional LTC policy (like I believe your blog discusses), but she could qualify for a hybrid LTC policy (whole life with a LTC rider). While, I too read Joe Tomlinson and respect his opinion and thus know such a hybrid policy in general provides less pure LTC coverage than a traditional policy this was our best option at the time so rather than have no coverage we chose it. The policy required a sizable deposit (67K) and ongoing annual premiums of $5639. For this one gets (upon qualifying) $6000 per month for 25 mos (after 60 day waiting period). This is basically return of my deposit (at about a 3.2% return if assume need arises 25 yrs after purchase). Then after 25 mos the rider kicks in and we get the $6000 inflated at 5% from the date of purchase or say a little over 20K per month for life (for mos 26 on). The 25 mos is combined months between my wife Ann and me. The annual premium does not increase, and should neither of us need any sort of LTC coverage then our estate would receive a 150K death benefit. I struggle with paying the annual premium (the amount seems high though we can afford), but like an article Michael Kitces wrote

    I believe we are basically self-funding anywhere between 25-50% of the first 2 yrs of LTC cost (and insuring anything beyond 2 yrs), and hoping to insure that the tail risk of an extended stay beyond 2 yrs is covered. The gap between cost of care at time of need and the $6000 payout for the first 25 mos is the gap we must self-insure. This seems like a reasonable approach given our means and the LTC options available to us at the time we were looking. In the end I have found solace for our decision by realizing that we have attempted to protect each of us from a potentially financially ruining situation (though I don't disagree the odds of such an event are low, but this is the type of risk meant for insurance low probability high impact) where one of us needs an extended nursing home etc stay. Also, the annual premium will not likely change our living standard and will not rise. We like that the 25 mos is a combined/joint 25 mos, and that the policy has a 5% inflation rider and lifetime coverage. And of course we like that the annual premium won't increase, and there is a waiver or premium during the time we are exercising benefits. Also, should we change our mind at any time we can get the cash value of the policy, but of course the LTC insurance would be gone at this point. And the cash value does not grow very much over time from the original deposit. The company has an A+ (Superior 2nd highest of 16 ratings) Best Rating, and AA- (4th highest of 22 ratings) from S&P.

    I do worry about the policy lapsing from cognitive decline as your blog mentions. Our policy had an option where we could list one person who would get notified if the premium was not paid on the due date. So this is some safeguard, and there is a grace period of 31 days. We also get a bill every year in the mail to prompt payment. But I would like an automatic payment draft option (and will call about this).

    Great article Dirk. I guess bottom line I still think the annual premium is a little high at $5639 (I look at the $6000 benefit for the fist 25 mos as just return of my 67K deposit). The annual premium basically covers any LTC need we would have after 27 mos (2 mos waiting period and 25 mos of cost sharing). I have tried to check whether the annual premium cost is reasonable by pricing (as close as I could) a similar benefit on the Federal LTC prgm website (premium calculator option) and Genworth LTC pricing site.

    Thanks again for a great article Dirk. Brad

  3. What asset or income level in retirement would qualify as being "Wealthy" where you say: "Wealthy households will be able to self-insure."

    1. Good question! Unfortunately, this is one of those "how much is enough" questions that's really difficult to answer. I've read recommendations from $700K per spouse to $2M per spouse. $700K seems on the low side to me and $2M sounds a little high. My guts says $1M to $2M per spouse before you should consider self-insuring if you can afford the insurance.

      That's the problem with not purchasing the insurance – you can never know with certainty if you'll have enough savings to cover a really poor outcome.

      Unfortunately, I think the only answer to "how much is enough" might be "the more the better." I'm sure neither of us finds that extremely helpful.

  4. Enquiry from Britain here: do any of your insurers provide cover in exchange for a lump sum payment up front? That would avoid the risk of annual payments being increased, or of muddle leading to payments not being made.

    Personal remark: I shouldn't dream of taking out an insurance for the rest of my life that allowed the supplier to increase my annual payment. Only a fixed payment would appeal to me.

  5. I think uncertain premiums are a big hurdle for most people and part of the reason only 13% of Americans buy LTC policies.

    I am not aware of any insurers that offer straight LTC insurance here for a single payment, but then I'm not that familiar with that market. I have read articles suggesting that would make the policies more attractive, but I am unaware of any carrier offering them presently.

    You can buy life insurance with an LTC rider for a single premium payment, but Joe Tomlinson finds them more expensive than standard LTC or self-insuring.

    BTW, my British friends and I thoroughly enjoyed this viral post from Scott Waters. If you haven't seen it, enjoy.

    And thanks for writing!

  6. I bought John Hancock LTC policies 7 years ago for myself and spouse at ages 56 and 50 respectively. We opted for a policy that had a compound inflation rider tied to the CPI and have had no rate increases. I have noted that JH policies of similar vintage with 5% compound riders have seen considerable rate increases. I wonder if the premiums for policies with 5% compound riders proved to be unsustainable in the low interest rates environment of recent years.

    As I have seen appreciation of our investments and some peak earning years since buying the policies I think we could afford to self-insure but am suffering from the sunk cost fallacy, particularly since I know that with these policies you substantially overpay for the low level of risk during the early years.

  7. Ouch! That sunk-cost fallacy is indeed part of this problem. It bites us all in so many ways.

    Thanks for contributing to the discussion!

  8. Here's a great piece by Michael Kitces. Michael is another writer whose opinion on retirement insurance issues I highly value. (He co-authored a book entitled, "The Annuity Advisor.")

    In this post, Michael explains when hybrid life insurance policies with an LTC rider might make sense. The most common situation will be for purchasers who can't qualify for the more stringent underwriting of traditional LTC policies. He also questions whether the "guaranteed" premiums of hybrid policies are a mirage.

  9. Dirk, here is another article by Michael Kitces explaining why the terms of LTC insurance needs to change (so as to make it more amenable to insurance again ie. like catastrophic coverage) as the nature of LTC claims has changed. He suggests individuals self-insure for a longer elimination period, and insure for the possible extremely long LTC stays. This would reduce the LTC insurance cost making it more affordable, but he admits many may not be able to afford the longer elimination period (1-2 yrs). As it stands now many LTC policies and many buyers want shorter elimination periods which ends up raising the costs when it might be more beneficial to lengthen the elimination period during which time a person self-insures while leaving insurance to do what it is intended to do cover truly catastophic LTC stays. Thanks, Brad.

    1. Thanks, Brad. I agree with Michael. LTC insurance should cover catastrophic costs. Same with health insurance and car insurance. Take the longest elimination period (deductible) you can afford.

  10. Hi Dirk,

    A discussion about Long Term Care should also include the ability to retain assets up to the amount of care a Partnership policy has paid for under the Federal Deficit Reduction Act of 2005 which expanded the Partnership program nationwide ( ).

    Kiplinger has a short article on them

    and Kitces mentions them in point #5 in this benefit design post

    current State status may be found here
    and one can Google their State Long Term Care Partnership to get details in their State.

    Moral of the story is that now one can retain and protect more assets with a Partnership plan than with a hybrid or standard non-partnership plan.

    1. Thanks, Larry. This has turned into a broader discussion of LTC insurance, which is what you hope for when you write a blog post.

      When Larry mentions asset protection in this context, he's talking about protecting assets from having to be spent to qualify for Medicaid.

      Typically, before Medicaid will pay LTC expenses you have to spend down nearly all of your assets. In states that have passed Partnership laws (see Larry's link), you can protect an amount of your personal assets (you don't have to spend them down) equal to the amount of your insurance benefit.

      For example, if your LTC insurance provides $250,000 in benefits, then you could get Medicaid benefits when your insurance benefits run out and keep $250,000 in assets instead of needing to spend them all before Medicaid kicks in. (Hybrid policies don't qualify.)

      You hope, of course, that LTC insurance covers it all and you never need Medicaid, but this is catastrophic planning.

      Larry's links are your extra credit reading assignment for this post.

      Fun discussion. Thanks!

    2. Yes extra credit Dirk ... and standard LTC insurance policies also don't qualify (ref your italics note near the bottom of your comment) ... policies have to be specifically labeled a Partnership policy for benefits to qualify for the asset protection. Just wanted to point this caveat out so people don't argue with their State over their non-Partnership policies.

    3. I said that hybrid policies don't qualify, but some standard LTC policies won't, either. According to Kiplinger, most – but not all – LTC policies will qualify.

      Thanks for clarifying that. Larry.

    4. Larry has correctly pointed out that my response still might be misinterpreted. We both understand that to qualify a policy must be certified by your state. (Some states have reciprocal agreements if you later move, but some don't.)

      My reading of the Kiplinger piece is that most LTC policies issued in the future will qualify, not that most existing policies qualify.

      Bottom line – make sure the policy is certified by your state. If not, it won't qualify as a Partnership policy. Nor will a hybrid life insurance/LTC policy or Annuity/LTC policy.

  11. Dirk, one thing we discovered with the Partnership Plan in Indiana was while assets are protected the income from those assets (which is what we wanted to live on) is not protected. Also, one must still apply and qualify for Medicaid - it is not automatic. Brad

    1. Thanks. I believe most if not all states have both an asset test and an income test for Medicaid and these partnership policies protect the assets. You still have to pass the income test.

      My response said "could get" Medicare benefits, though you might not. A clearer statement is "then you can apply for Medicaid benefits."

      I'm not terribly excited about partnership LTC because if you never use Medicaid there is no benefit. If you're not wealthy and your plan to deal with high end-of-life medical expenses is to use Medicaid from the beginning, then you wouldn't buy any LTC policy.

      Still, LTC policies that protects assets are better than LTC policies that don't when the cost is the same. If you are married and end up on Medicaid, you could protect significant assets for your spouse.

      Thanks, Brad!

  12. In a recent post, Michael Kitces answered a question I had found puzzling. The LTC industry reports a low (1% to 2%) lapse rate for their policies, but the study referenced in my posts says 32% of men and 38% of women let policies lapse. The explanation is that the insurance industry quotes an annual lapse rate, which will result in a about a third of policyholders lapsing over 29 years (at -1.5% per year).

  13. This speculation might apply more to Britain than the USA, but here goes.

    For me and my wife LTC insurance would be more useful for the first of us to go into care rather than the second, because in the latter case our house could be sold to fund the LTC. I'd imagine that such a policy would be substantially cheaper than cover for both of us. Are such policies available in the market?

    1. I'm not sure I understand the question. Are you asking if you can purchase an LTC policy in the U.S. that covers only one spouse? If so, according to the American Association of Long Term Care Insurance, some insurers offer discounts when only one spouse is insured, but several don't. You can read more here.

      Don't know how it works in the U.K., but every time I visit, I realize that I don't know how anything works in the U.K. I needed help buying a ticket for the Tube with my non-chip credit card last time! I finally did get the hang of the trouser press, though.

    2. What I meant, Dirk, is whether anyone sells an LTC policy that would pay out whenever the first of a couple needs LTC, irrespective of whether that person is the husband or wife.

    3. No, I'm thinking of keeping the cost down by having only one of the couple covered, to wit the first one to enter long term care. My logic is that if the other spouse needs LTC he or she could pay for it by selling investments or property. You really don't want to sell your house to fund the first of the couple to enter LTC. When/if the second one enters LTC, the house becomes redundant anyway, so sell it.

    4. I think I understand what you are trying to do, but I'm not sure there is a policy described in that way. I believe a joint policy would allow you to buy one $100K policy, for example, and pool the benefit. The first spouse to need LTC could use up to the full $100K, but any benefit not used would then be available to the second spouse. Wouldn't that accomplish the same thing plus give the second spouse access to unspent benefits that a "first-spouse-to-need" policy would not? You could purchase a pool equivalent to the maximum benefit you want for the first spouse with need for LTC.

    5. Thanks, Dirk. I think my idea ought to be cheaper because the insurance company wouldn't have to pay over any unused benefit, but it certainly wouldn't be cheaper if no firm offers it! Maybe one should try it?

      I'm just trying to think of any way to reduce the cost while still giving the essential cover. Obviously one has to start from the assumption that the couple are trying to provide for themselves, not for any potential inheritors.

    6. Agreed – what you describe should be cheaper. I'm just not aware of anyone who overs such a policy, but it's a big country with a lot of insurers. Perhaps another reader knows of one.

  14. Thanks for this post and the discussion. There needs to be more published on this topic by unbiased (non-insurance) sources. I will definitely read all of the articles linked here.

    1. Paul, there are many articles out there, pro and con, from both insurers and unbiased third parties. Many have good arguments. That's a sure sign that there is no single answer and that either path is risky.

      Here's the problem, as I see it. When you buy car insurance or home insurance, you're paying the insurer to accept your risk. You pay a year at a time for a year of coverage and there isn't a lot of risk that the insurer won't pay a covered loss or raise your premiums to the point that you can no longer afford them.

      With LTC insurance, there is significant risk that you might pay a lot for insurance, suffer a claim, and not have it covered, for various reasons. So, the insurer hasn't really accepted all of your risk in exchange for that premium. I think both insuring and not insuring have risk and that is why there is no great answer.

      Unless and until insurers figure out pricing that is both affordable and profitable, the insurance is going to be risky. But so is not insuring.

      Thanks for writing!

  15. Great article, Dirk. You're an exceptional writer.

    Unfortunately, the brief written by the Center for Retirement Research (CRR) uses some very old and some sketchy data. Their study only looked at people who may have owned LTC insurance 13 years ago (way back in 2002). And their sample size was very small--about 124 people who may have lapsed their LTC insurance sometime between 2003 and 2006.

    The biggest flaw, however, is that CRR’s brief did not mention the federal law that protects seniors from losing their long-term care coverage.

    Due partly to this federal law, the percentage of seniors that lapse their LTC insurance each year has dropped by 68% since the year 2000.

    You can read more about it here:

    Thanks again for a great discussion.


    1. Thanks, Scott. I enjoyed our discussion, as well, and I have thoroughly reviewed the LTC industry response to the research from the Center for Retirement Research at Boston College to which you refer.

      I will restate my original comment, since you were kind enough to join the discussion here at the Retirement Cafe.

      All research is flawed in some way and I'm sure CCR's research can be questioned. They suggest that retirees have about a 35% chance of letting LTC coverage lapse.

      The federal law to which you refer was not in effect for the policies studied, but when I queried CRR, they told me their research has found the law ineffective.

      The LTC industry quantifies lapses differently. They estimate a 1% to 2% annual lapse rate. If the rate turns out to be 2% (and, as I noted, the industry's inability to accurately forecast lapse rates has been a huge part of the problem), then a retiree who lives 30 years after purchasing the policy will have a lifetime lapse probability of 1 - 0.98^30, or 45%. Should the rate turn out to be 1.5%, her lifetime probability is 36%. At 1% a year, she will have a 26% probability of lapse 30 years after purchasing the insurance.

      So, despite all of the challenges to the validity of CRR's research, I can use your numbers or theirs and come up with about the same probability that an elderly retiree's policy will lapse (25% to 35%).

      Thanks for reading my blog, and thanks again for the discussion!

  16. Hi Dirk, very informative article. I hope table 2 (2005) can be updated to reflect current statistics. As a single person with one adult son do you use the same general rule for someone who is neither wealthy nor poor to consider buying ltc? best wishes

    1. I'm sure it will be updated as time goes by and more data is collected, but I think there are two things to keep in mind that won't change as that data changes.

      First, this is data regarding what has happened in the past and the more important question is what will happen in the future. Will LTC policies be improved to correct the current problems? At the other extreme, will the trend for LTC insurers to get out of that market continue? Past data won't answer those questions today.

      Perhaps a more important consideration is that no matter how likely it is or becomes that you will need long-term care, the fact is that you might. And the cost could be catastrophic. We don't forgo homeowners insurance simply due to the evidence that most houses don't burn down because we want to avoid a low-probability, high-magnitude risk. (And because our mortgagor makes us.) We buy it largely because we consider homeowners insurance affordable and safe. There is little risk that insurers will stop selling homeowners insurance or raise our premiums to the point of unaffordability.

      Some households (certainly not most) could pay $250K of $500K in end-of-life costs and still support a surviving spouse. Given the known problems with LTC policies, those households might well self-insure.

      I suppose a married person could depend more on the possibility of receiving care from his or her spouse, but a surviving spouse can't. If your son is financially independent, then planning on the Medicaid route in the unlikely event of catastrophic health care costs might be a reasonable choice. If you were quite wealthy, I'd say self-insure. If you can't afford the premiums, you really have no choice but to go the Medicaid route. If you're in between, as I said, I don't think there is a clear answer.

      You have to lay out the possible outcomes, their costs and the cost of insurance and see which one is most palatable for you personally.

      Thanks for writing!