LTC Bullet:  LTCI Lapses Reconsidered

Friday, May 6, 2016


LTC Comment:  LTCI lapses are conceptually slippery.  Some clarification by Claude Thau after the ***news.***

*** WE PUBLISHED the following LTC Clipping a few weeks ago.  Today’s LTC Bullet conveys long-term care insurance expert Claude Thau’s thoughts on the meaning of LTCI lapses in the form of comments on the same article.

4/13/2016, “LTCI: Which insureds have the toughest grip on their policies?,” by Allison Bell, LifeHealthPRO                                                 

Quote:  “The research team found that, on average, insureds with higher benefits were more likely to keep paying their premiums than insureds with lower benefits, and that insureds who went through a full medical underwriting process were more likely to keep paying.  In the first three years a policy was in force, fully underwritten insureds were only about half as likely to let an LTCI policy lapse as other insureds were. But insureds who bought their LTCI policies through career agents were only about half as likely as other policyholders to let a policy lapse as other insureds.

LTC Comment:  Interesting data and insights regarding lapse rates worth considering in the context of the Boston College Center for Retirement Research claims that over a third of LTCI policies lapse.  See LTC Bullet:  Another LTCI Hit Job? ***



You may find it helpful to read the article referenced above before going on to what follows:  “LTCI: Which insureds have the toughest grip on their policies?,” by Allison Bell, LifeHealthPRO, April 13, 2016.

Having seen an email with some comments by Claude Thau, we asked him if we could share some of his insights with our readers.  He agreed that the following could be helpful to readers.

First of all, he wanted to note that he looks forward to each of these studies by the Society of Actuaries because they provide so much useful information and stimulate so much thought.  He respects Allison greatly because she is a tireless and prolific dispenser of valuable information.

If lapse data is different between X and Y, the key question is “Why are the lapse rates different for X vs. Y?”  Sometimes there are other factors involved which correlate with X and Y and skew the results.  That is, it may be true that the data varied by X vs. Y, but X and Y might not be causing that impact.

For example, what if I were to tell you that heart condition victims are taller than the average person?  Would that surprise you?  Would you conclude that taller people are more exposed to heart conditions than shorter people?  Clearly, it is true that people with heart conditions are taller because a higher percentage of such victims are adults than in the general population.  Because looking at heart conditions screens out a much higher percentage of children than adults, those victims are taller than usual.

Quote from article:  "On average, policyholders who went through a full medical underwriting process and those who bought their coverage from career agents seem to be more likely to keep their coverage than other policyholders."

Comment:  Both of those statements are true, but perhaps not because of the medical or because of the career agent rather than broker, or at least certainly not entirely because of those.  A key factor above is that group policies have higher lapses than individual policies for the following reasons:

1.    When a group has core coverage paid for everyone by the employer, most employees do not buy-up.  Those who don't buy-up, usually lapse when they terminate employment.

2.    Voluntary work-site LTCi has a higher lapse rate, in general, because when people terminate they may forget to pay premium or may not be able to afford the premium because they are temporarily unemployed.  A lag might cause them to have to pay more than one-month’s premium to continue the policy.  If there was a core benefit, they have to absorb the cost of the core benefit prospectively as well.

3.    Executive carve-out policies probably have a higher-than-average lapse rate upon termination of employment because some executives may not want to pay the premium that has been paid for them previously.

  • Individual policies dominate the full medical UW [underwriting] data and group policies dominate the simplified UW cases.  The fact that the medical UW category excludes group policies is probably the primary cause for the difference in lapse rate experience.
  • Brokers dominate the group sales, hence have higher lapse rates.  If someone were to analyze the difference between broker and agent client persistency looking only at individual policies, they might find a significantly different result.

Quote:  "Actuaries classify an LTCI lapse as ‘voluntary’ if a living policyholder stops paying the premiums for some reason other than death. Actuaries contrast that kind of ‘voluntary lapsation’ with lapsation due to the policyholder's death."

Comment:  That's true but a very important caveat is that lapses are overstated in the study because some deaths are coded as lapses.  Another important caveat is that the SOA study addressed full lapses but did not consider partial lapses.  Because of large price increases on inforce policies, there have been a lot of partial lapses, which are not reflected in this data.

Quotes:  "The voluntary individual lapse rate fell to 2 percent, from 2.7 percent in 2005-2007."  "The voluntary group lapse rate fell to 4.5 percent, from 6.4 percent in 2005-2007."

Comment:  Both true, but don’t forget that the percentage of first year policies dropped.  As first-year policies have the highest lapse rates, the overall lapse rate would drop if the percentage of first-year policies drops, even if the lapse rate had remained unchanged for each individual policy year.

Quote:  "The research team found that, on average, insureds with higher benefits were more likely to keep paying their premiums than insureds with lower benefits"

Comment:  I would expect this to be true to some degree, but I suspect that the results are largely distorted because:

  • Group policies (which have higher lapse rates) generally provide less coverage.
  • The size of coverage has probably increased over time (not necessarily monotonically) causing a higher percentage of the large policies being in the first few policy years, when lapse rates are higher.

Quote:  "The research team also found that two opposing forces eventually cause the lapse curves of individual LTCI insureds and group LTCI insureds to cross.....But, because the individual insureds tend to be much older than the group insureds, death becomes a major cause of individual policy terminations starting around the ninth policy year."

Comment:  The first sentence is about lapse rates.  The second sentence is about total termination rates.  Readers might get confused.  The actuaries were trying to report lapses, exclusive of mortality.  But, as noted above, some mortality creeps in because of deaths miscoded as lapses.

Quote:  "Meanwhile, group insureds seem to start getting serious about paying their premiums after holding their coverage for about eight or nine years."

Comment:  There are a number of factors at play here.  Employees have higher turn-over rates at younger ages.  When you get out 8-9 years, you're looking at an older, more established employee who is less likely to terminate employment, hence less likely to lapse.  Also the percentage of core coverage reduces the older the inforce block; as the people who are left are more likely to have paid some or all of the cost, they are likely to be more persistent.

Special thanks to actuary and LTCI industry icon Claude Thau (formerly Chairman of the Center for Long-Term Care Financing’s Board) for permission to publish his comments on lapses.  Reach Claude at 800-999-3026, x2241 or email him at