Claude Thau

The 2021 Milliman Long Term Care Insurance Survey is the 23rd consecutive annual review of stand-alone long-term care insurance (LTCI) published by Broker World magazine. It analyzes the marketplace, reports sales distributions, and describes available products.

More discussion of worksite sales, including a comparison of worksite sales distributions vs. non-worksite sales distributions will be in Broker World magazine’s August issue.

Unless otherwise indicated, references are solely to U.S. stand-alone LTCI sales, excluding exercised future purchase options (FPOs) or other changes to existing coverage. “Stand-alone” refers to LTCI policies that do not include death benefits (other than returning premiums upon death or waiving a surviving spouse’s premiums) or annuity or disability income benefits. Where referenced, “combo” products provide LTCI combined with life insurance or annuity coverage. “Linked benefit” policies are combo policies which can allow more than the death benefit or annuity account value to be used as LTCI.

Highlights from this year’s survey

Participants
Eight carriers participated broadly in this survey. Four others provided sales information so we could report more accurate aggregate industry individual and multi-life sales.

We estimate our statistical distributions reflect about 75 percent of total industry policies sold (85 percent of premium) and about 15 percent of worksite policies sold (30 percent of premium). Total industry sales and total worksite sales (the denominators) include authors’ estimates of sales for three companies.

Our worksite statistical distributions can vary significantly from year-to-year because insurers focusing on particular worksite markets may be over- or under-represented. Our worksite statistical analysis is overly weighted toward executive carve-out programs. The carriers which provided 2020 statistical data had an average worksite annual premium of $4,218 whereas the insurers which did not provide statistical data had an average worksite premium of $1,634.

CalPERS, Northwestern and the six insurers displayed in the Product Exhibit provided broad statistical information. Auto-Owners, Country, LifeSecure, and Transamerica contributed total and worksite sales (new premium and lives insured) but did not provide other information. However, in a few places we were able to reflect some of their product designs in our statistical distributions. We estimate approximately $10 million of sales for those insurers which did not contribute their sales data.

Country Life, Genworth and CalPERS stopped selling stand-alone LTCI in 2020; however, Genworth has since resumed sales and CalPERS intends to resume sales. MassMutual and Transamerica discontinued selling stand-alone LTCI early in 2021 but Transamerica continued to accept new worksite applications until mid-year and may be experiencing last minute sales as a result.

Sales Summary

• After having reported an increase in new premiums in 2019 (for the first time since 2012), twelve insurers reported $139,562,096* in 2020 annualized new premium sales (including exercised FPOs) and 41,440 new policies, 9.7 percent less premium and 11.9 percent fewer policies than the same insurers sold in 2019. However, some insurers with limited 2020 activity did not report 2020 sales to us. Including estimated sales for those insurers we peg total stand-alone LTCI industry sales at about $150 million*, which is nearly 13 percent less than the 2019 sales of $171,634,536 (restated upward by nearly $900,000 from our report last year). As noted in the Market Perspective section, sales of policies combining life insurance with LTCI or Chronic Illness benefits dropped eight percent in the first half of 2020 and corresponding premium dropped 19 percent. *Single premium sales are counted at ten percent for the annualized premium calculations herein.
• Although we are not able to quantify the magnitude, the COVID-19 pandemic could have had a material impact on sales.
• Three insurers sold more new premium than in 2019.
• The pandemic may have had a greater impact on worksite sales. We estimate worksite new annualized LTCI premium dropped 15 percent in 2020 (18 percent fewer new sales).
• With FPO elections included in new premium, Northwestern retained the number one spot in sales. Mutual of Omaha was a strong second and again had a large lead in annualized premium from new policies sold. Together, these two insurers combined for 60 percent of new premium including FPOs, compared to 59 percent last year.

The number of policies inforce increased each year through 2014 but has decreased each year since, decreasing 0.9 percent in 2020. However, inforce premium continued to increase (1.9 percent) despite fewer policies inforce and more policies in paid-up status. Inforce premium rises due to sales, price increases, and benefit increases (including FPOs), and reduces from lapses, reductions in coverage, deaths, and shifts to paid-up status for various reasons. A major carrier did not report inforce data, likely suppressing the percentage of inforce premium increase.

Collectively, the seven participants which reported claims saw claims increase only 1.6 percent in 2020, partly because two insurers reported a reduction in claims. The other five insurers experienced a 5.2 percent increase in claims in 2020. Overall, the stand-alone LTCI industry incurred $12.9 billion in claims in 2019 based on companies’ statutory annual filings, raising total incurred claims from 1991 through 2019 to $154.4 billion. (Note: 2019 was the most recent year available when this article was written.) Most of these claims were incurred by insurers that no longer sell LTCI. The reported 2019 incurred claims are 11.2 percent higher than the $11.6 billion of incurred claims reported in 2018.

The placement rate dropped from 59.2 percent in 2019 to 57.8 percent in 2020, which is also lower than the 59.0 percent and 58.8 percent placement rates in 2017 and 2018, respectively. Difficulty in collecting underwriting information during the pandemic may have contributed to lower placement rates. Table 23 shows that three-fourths of the reduction in placement rate can be attributed to suspended and withdrawn applications and Table 26 shows that underwriting averaged about 3 days longer. The longer average underwriting time was also likely caused by the pandemic. However, due to 2020 pending applications that will be reported in 2021, as well as the 2019 applications that were pending at the end of that year, some of our statistics such as underwriting processing time do not fully reflect the impact of the pandemic.

About the Survey
This article is arranged in the following sections:

Highlights provides a high-level view of results.
Market Perspective provides insights into the LTCI market.
Claims presents industry-level claims data.
Sales Statistical Analysis presents industry-level sales distributions reflecting data from 8 insurers.
Partnership Programs discusses the impact of the state partnerships for LTCI.

Available here at www.BrokerWorldMag.com:

• Product Exhibit shows, for 6 insurers: financial ratings, LTCI sales and inforce, and product details.
• Product Details, a row-by-row definition of the product exhibit entries, with some commentary.
• Premium Exhibit shows lifetime annual premiums for each insurer’s most common underwriting class, for issue ages 40, 50, 60, and 70 for single females, single males, and heterosexual couples (assuming both buy at the same age), based on $100 per day (or closest equivalent weekly or monthly) benefit, 90-day facility and most common home care elimination period, three-year and five-year benefit periods or $100,000 and $200,000 maximum lifetime buckets, with and without Shared Care and with flat benefits or automatic three or five percent annual compound benefit increases for life. Worksite premiums do not reflect any worksite-specific discount, though some carriers offer this.
• Premium Adjustments (from our Premium Exhibit prices) by underwriting class for each participant.
• Distribution by underwriting class for each participant
• State-by-state results: percentage of sales by state, average premium by state and percentage of policies qualifying for Partnership by state.

MARKET PERSPECTIVE (more detail in subsequent parts of the article)

“Both buy” and married discounts are reaching stable points. In the past year, some insurers lowered couples’ discounts significantly, to the 15 to 20 percent range. One-of-a-couple discounts are also dropping, to the zero to five percent range. The histories of these discounts are related.

When all LTCI had unisex prices, insurers’ single-person prices reflected that single buyers were predominantly female. Couples who bought policies were nearly 50 percent male; insurers did not have to charge them two prices weighted toward females, so they raised the couples’ discount. When the industry converted to gender-distinct pricing, the justification for couples’ discounts decreased substantially; however, large couples’ discounts continued to be common until now.

As an example, a single person price of $1,500 led to a couples’ combined price of $1,800 when couples’ discounts were 40 percent. If insurers took away the couples’ discount when one spouse was declined, consumers would be disappointed that the spouse about whom they were most concerned was declined, and the price for the healthy insured would be 5/6 of their combined price (5/7 if the discount was 30 percent). Fearing that the healthy spouse would refuse the policy, insurers typically cut the discount in half, rather than removing it completely. In the above example, the healthy spouse would have been charged $1,200 to $1,275, depending on whether the couples’ discount was 40 or 30 percent, respectively.

In the 2019 article “Is Your Spouse Contagious?,” Milliman analyzed the additional cost of LTC incurred by people whose spouses already had LTCI claims. As people who are LTC caregivers are likely to have worse LTCI claims experience than single people, a discount is inappropriate if the spouse is uninsurable. Of course, sometimes only one spouse buys even though both are healthy; for example, perhaps one spouse already has coverage. Nonetheless, their results make it hard to justify an across-the-board one-of-a-couple discount.

We summarized the “Is Your Spouse Contagious?” findings in our survey questionnaire, asking what the industry can do to lower claim costs after the claim or death of a spouse. The same five insurers responded to each question and their recommendations mostly cut across both situations. They cited proactive outreach to provide supportive services to help clients age in place, including preventive measures and care management, examples being home safety assessments and technology to monitor an insured’s health if they live alone. Following the death of a spouse, one carrier suggested informal care options with low daily limits which would permit relatives to be paid caregivers and another suggested a new plan of care, answers that seemed to anticipate that the survivor was already on claim. There were some concerns, however, with one insurer indicating that “cost will be a barrier” and another commenting “if appropriate”.

When one spouse is on claim, two insurers suggested offering respite care options to support the non-claim spouse.

Some specific steps the respondents may have had in mind in their general comments include brain fitness training; helping the healthy spouse marshal and manage resources to help the care recipient; access to cognitive tests that provide early detection of cognitive slippage; lifestyle suggestions including diet and sources of prepared meals; home modification projects; document organization; end-of-life planning; tele-medicine; information to access available government benefits and local services; help in finding and evaluating potential commercial caregivers; products and information about reducing risk and/or facilitating improved caregiving; etc.

We noted that the 10-year Treasury yield as of January 1, 2021 was 0.93 percent and asked what insurers projected it would be in 5 and 10 years. The range in five years was scattered between 1.5 to 3.0 percent and in ten years was two to three percent, with four of the five answers being 2.9 to 3.0 percent. Although all envisioned rising interest rates, only two mentioned that it would ease pressure on profit margins.

The June issue of Broker World magazine had an article about the impact of the pandemic which included data from this survey. Rather than repeat that data here, we refer readers to the June issue.

Government-provided LTCI programs continue to be a topic of discussion.

The Affordable Care Act (2010) included a LTCI program (the “CLASS” Act). Criticisms of the design were validated when, after passage of the ACA, the government dropped the LTCI program because it could not find a way to make it work.

Since then, there have been state efforts to create LTCI programs. On January 1, 2022, the Washington Long-Term Care Trust (WLTCT) will begin collecting 0.58 percent payroll tax to fund a $36,500 lifetime pool of money starting in 2025 (the pool of money is intended to inflate according to the Washington consumer price index). The program provides coverage for vested individuals receiving care in the state of WA. Government-provided LTCI programs will be watched closely for their potential impact on LTCI sales.

Since 2009 (varies by jurisdiction), if an insurer concludes that a claimant is not chronically ill, the insurer must inform the claimant of his/her right to appeal the decision to independent third-party review (IR), which is binding on insurers. As shown in our Product Exhibit, most participants have extended IR beyond statutory requirements, most commonly to policies issued prior to the effective date of IR. It is hard to get data relative to IR. Regulators in some states have not set up the required panel of independent reviewers and regulators do not collect IR statistics. Insurers often do not track IR. It is complimentary of the industry’s claims processing that there has not been a clamor for IR; that there have been significantly fewer requests in the past two years; and that insurers have consistently been completely upheld 70 percent or more of the time based on data we have received over the years from insurers and from Steve LaPierre, President of LTCI Independent Eligibility Review Specialists, LLC (LTCIIERS). In other cases, the reviewer might agree with the insurer partly (e.g., that the claimant did not satisfy the triggers initially, while concluding that the claimant satisfied the triggers later).

Current prices are more stable than past prices, partly because today’s prices reflect much more conservative assumptions based on far more credible data and lower assumed investment yields. Three participants have never increased premiums on policies issued under “rate stabilization” laws. The others reported no increases on policies issued since 2015, 2014 (2 insurers), and 2013 (2 insurers). Nonetheless, many financial advisors presume new policies will face steep price increases, and hence may be reluctant to encourage clients to consider LTCI.

As shown in Table 23, the placement rate for the past 8 years has remained at about 60 percent. Improving the placement rate is critical to encourage financial advisors to mention LTCI to clients. The industry may be able to improve placement rates as follows.

• Utilize E-applications for faster submission and reduced processing time, thereby increasing placement.
• Pre-qualify an applicant’s health. Effectively and efficiently increasing the percentage of applicants who are pre-qualified will decrease declines.
• More effective education of distributors by insurers, such as drill-down questions in on-line underwriting guides and eApps.
• Require cash with the application (CWA). CWA led to about five percent more of the apps being placed according to our 2019 survey.
• Continue to improve messaging regarding the value of LTCI and of buying now (rather than in the future). Such messaging would increase the number of applications and improve the placement rate by attracting younger and healthier applicants.

Once again, more than 80 percent of our participants’ policyholders exercised their FPOs (future purchase option, a guarantee that, under specified conditions, a policyholder can purchase additional coverage without demonstrating good health). As both the additional coverage and unit price increase over time, FPOs become increasingly expensive, even more so with the price increases that the industry has experienced. The high election rate demonstrates the importance of LTCI to policyholders and the effectiveness of annual (as opposed to triennial) negative-election FPOs. (Negative-election FPOs activate automatically unless the client rejects them, as opposed to positive-election FPOs which activate only if the client makes a request.) At least in some markets and with some designs, policyholders reliably exercise FPOs when they must do so to continue to receive future offers. Considering such FPOs and other provisions, we project a maximum benefit at age 80 of $305/day for an average 58-year-old purchaser in 2020, which is equivalent to an average 2.9 percent compounded benefit increase between 2020 and 2042. Purchasers may be disappointed if the purchasing power of their LTCI policies deteriorates over time.

Linked benefit products are attractive because even if the insured has no LTC claim, their family will receive benefits and because they often have guaranteed premiums and benefits. According to LIMRA, 516,809 stand-alone LTCI and combination (life and LTCI and life and Chronic Illness) policies were sold in 2019 and were distributed as follows, with stand-alone LTCI’s 11 percent market share the same as 2018. (ADB= “Accelerated Death Benefit)

The stand-alone column and the linked-benefit column are both much more significant in terms of LTC protection than indicated above, for the following reasons:

1. The Accelerated Death Benefit (ADB) provisions do not increase over time. By the time the average claim payment is made, the stand-alone policies’ maximum monthly benefit (on plans with benefit increase features) will have risen significantly. The linked-benefit portion would also have increased significantly.
2. The stand-alone and linked-benefit policies have significantly lower lapse rates than the policies with ADB. Thus, they are much more likely to be around when care is needed.
3. The stand-alone and linked-benefit policies may have lower mortality rates because of their better persistency and because they are purchased with LTCI in mind whereas the ADB policies are generally purchased for their life insurance and the ADB benefit may be incidental in the minds of the buyers.
4. The ADB policies are less likely to be used to pay for care, as it may be preferable to have the death benefit be paid to the beneficiary.

A small offset is that the stand-alone policies are more likely to be reimbursement-based benefits, which are less likely to result in the full benefit being paid each month.

According to LIMRA, in the first half of 2019, combination life sales dropped slightly, but then rebounded strongly in the latter half of the year to finish ten percent higher than 2018 in terms of premium (although, one percent lower in terms of the numbers of policies). However, in the first half of 2020 (the most recent data available as this article was being written), premiums were 19 percent lower and new policies eight percent lower than the weak first half of 2019. In addition to the pandemic, LIMRA noted that the new 2017 CSO Mortality Table and low interest rates led to higher premiums which depressed sales. The policies with the most significant LTCI benefits (linked benefits) dropped the most (27 percent in premium; 21 percent in policy count).

Only four participants offer coverage in all U.S. jurisdictions; no worksite insurer does so. Insurers are reluctant to sell in jurisdictions which have unfavorable legislation or regulations, restrict rate increases, or are slow to approve new products.

All but one of our participants use third party administrators (TPAs) and five of the eight participants use reinsurers. The number of insurers using TPAs for the following functions is shown in parentheses: Underwriting (6), Issue (5), Billing (5), Commissions (3), and Claims (6). Two insurers noted they use their underwriting and claims TPAs for only some functions and another, which responded that it does not use TPAs for underwriting and claims, noted that it sometimes receives help from its TPA, but the TPA never makes the decision. We thank CHCS, DMS, LifeCare Assurance, Long-Term Care Group, RGA, and Wilton Re for their contributions to the LTCI industry. Other reinsurers and TPAs may support insurers not in our survey. In some cases, affiliated companies provide reinsurance or guarantees.

CLAIMS

• Seven participants reported 2020 claims. As some companies are not able to provide detailed data, some statistics are more robust than others.
• The insurers’ combined claim payments on individual policies rose 1.6 percent in 2020 over 2019. Two insurers reported a decrease in claims, while the other five insurers reported a 5.2 percent increase.
• The LTCI industry has had a much bigger impact than indicated above, because a lot of claims are paid by insurers that do not currently sell LTCI or did not submit claims data to us.

LTCI claims paid by insurers no longer selling LTCI may differ significantly from the following statistics as their claimants are more likely to have facility-only coverage, be older, etc.

Table 1 shows the total dollar and number of reported individual and multi-life (not group) LTCI claims. It reflects the same carriers for both years. As noted above, total claims rose only 1.6 percent. The impact of the pandemic is not clear in the data, but some claimants likely died prematurely and/or spurned facility care, particularly nursing home care. Some claimants likely abandoned facilities, thereby losing coverage either because they did not hire commercial home caregivers or because they had a facility-only policy and did not qualify for exemptions that some insurers provided. Claimants and commercial caregivers potentially were hesitant to meet during a large part of the year, likely dampening home care claims.

Table 2 shows the distribution of those claims by venue, which have shifted away from nursing homes over the years (except in 2019) due to consumer preferences and more claims coming from comprehensive policies. The distribution reflects different insurers from year-to-year; 2019’s aberration was caused by the absence of data from an insurer which resumed providing data this year.

The inception-to-date number of claims is surprisingly more weighted to Nursing Home than the dollar of claims. It seems that nursing home claims come from older policies with lower maximum daily benefits.

In the distribution based on number of claims, a person who received care in more than one venue is counted once for each venue, but not double-counted in the total line.
Six carriers reported their number of open individual claims at year-end, ranging between 51 and 81 percent of the number of claims paid during the year, averaging 64 percent overall.

Table 3 shows average size individual claims since inception. Because 41 percent of claimants since inception have submitted claims from more than one type of venue, the average total claim might be expected to exceed the average claim paid for any particular venue. However, individual Assisted Living Facility (ALF) claims stand out as high each year, probably because:

• ALF claims appear to have a longer duration compared with other venues.
• Nursing home costs are most likely to exceed the policy daily/monthly maximum, hence nursing home claims are most likely to understate the cost of care.
• People who maximize the use of their maximum monthly benefits can generally spend as much in an ALF as in a nursing home.
• Although some surveys report that ALFs cost about half as much as nursing homes on average, ALFs often charge more for a memory unit or for levels of assistance that align more closely with nursing home care. Upscale ALFs seem to cost a higher percentage of upscale nursing home costs than the average ALF/nursing home ratio.

Several insurers extend ALF coverage to policies which originally did not include ALF coverage, providing policyholders with significant flexibility at time of claim, but contributing to the insurers’ need for rate increases.

The following factors contribute to a large range of average claim results by insurer (see Table 3), which results in significant year-to-year differences in Table 3 because different insurers contribute data:

1. Different markets (by affluence; worksite vs. individual; geography; etc.)
2. Demographic differences (distribution by gender and age)
3. Distribution by benefit period, benefit increase feature, shared care and elimination period.
4. Distribution by facility-only policies vs. 50 percent home care vs. 100 percent home care vs. home care only, etc.
5. Different lengths of time in the business.

The following factors cause our average claim sizes to be understated.

1. For insurers reporting claims this year, nine percent of inception-to-date individual claims are still open. Our data does not include reserve estimates for future payments on open claims.
2. People who recover, then claim again, are counted as multiple insureds, rather than adding their various claims together.

Besides being understated, average claim data does not reflect the value of LTCI from some purchasers’ perspectives, because the many small claims drive down the average claim. LTCI can provide significant financial return for people who need care one year or longer. A primary purpose of insurance is to protect against adverse results, so the amount of protection, as well as average claim, is important.

Five insurers provided their current individual (excludes group) monthly LTCI claim exposure, which exceeds $5 billion (note: reflects only initial monthly maximum for one insurer). As shown in Table 4, this figure is thirty times their corresponding monthly LTCI premium income and more than 41 times their 2020 LTCI monthly paid claims. Eight insurers contributed data regarding their inforce distribution by benefit period. Treating endless (lifetime benefit periods) as a 15-year benefit period, we found that their average inforce benefit period is 6.85 years. Changing the assigned value of the endless benefit period by one year has an impact of approximately 0.25 years on the average inforce benefit period. With annual exposure thirty times annual premium and assuming an average benefit period of about 6.85 years, we estimate that total exposure is 207 times annual premium.

Three insurers reported their current average individual maximum monthly maximum benefit for claimants, with results ranging from $4,121 to $6,729.

Nursing home (NH) claims are more likely to use the policy’s maximum daily/monthly benefit than ALF claims, because ALF daily/monthly costs are generally lower and because policies sometimes have lower maximums for ALFs. ALF claims correspondingly are more likely to use the policy maximum than are adult day care and home care claims.

STATISTICAL ANALYSIS
Bankers Life, CalPERS, Knights of Columbus, Mutual of Omaha, National Guardian, New York Life, Northwestern and Thrivent contributed significant background data, but some were unable to contribute some data. Four other insurers (Auto-Owners, Country, LifeSecure and Transamerica) contributed their number of policies sold and new annualized premium, distinguishing worksite from other sales, but not clarifying whether FPOs were included in the premium.

Sales characteristics vary significantly among insurers based on market differences (individual vs. worksite, affluence, gender distribution, etc.). Year-to-year variations in policy feature distributions may reflect industry trends but may also reflect changes in participants, participant practices and designs, participant or worksite market shares, etc. The statistical differences between the worksite and non-worksite sales will be reported in the August issue of Broker World.

Market Share
Because new coverage is being issued, we include purchased increases on existing policies in new premium (we call them FPOs and include board-approved increase offers). FPOs increased in 2020, cushioning the reduction in premium from new sales. Removing FPOs spotlights insurers with more new policy sales. Table 5 lists the top 8 participants in 2020 new premium including FPOs and without FPOs. Northwestern ranks #1 including FPOs, with Northwestern and Mutual of Omaha accounting for 60 percent of the market. Ignoring FPOs, Mutual of Omaha is #1. The premium includes 100 percent of recurring premiums plus ten percent of single premiums.

Worksite Market Share
After a strong 2019, worksite premium dropped about 15 percent in 2020, accounting for a lower percentage of total sales than in 2019. It is possible that the worksite market could have produced a bigger percentage of total sales than in 2019 had the pandemic not occurred. However, with Transamerica leaving the market and other changes, the future of the worksite LTCI market (except executive carveout) is unpredictable. LifeSecure, Mutual of Omaha, and National Guardian may pick up market share; linked-benefit worksite products might pick up market share; and/or LTC risk covered in the worksite might drop.

Worksite sales consist of three different markets, the first two of which produce a higher percentage of new insureds than of new premiums:

• Voluntary group coverage generally is less robust than individual coverage.
• Core/Buy-Up programs have particularly young age distributions and modest coverage because a lot of people do not buy-up and are less likely to insure spouses.
• Executive carve-out programs generally provide the most robust coverage. One- or two-couple executive carve-out sales may not qualify for a multi-life discount with some insurers, hence may not be labeled as worksite sales in submissions to our survey.

The amount of worksite sales reported and its distribution among the three sub-markets significantly impact product feature sales distributions. Table 6 is indicative of the full market (including our estimates for two insurers which did not report sales), but this year’s policy feature distributions significantly underweight the voluntary and core/buy-up markets, as most of those markets are not reflected in our statistical data. More information about worksite sales will appear in the August issue of Broker World magazine.

Affinity Market Share
Affinity groups (non-employers such as associations) produced 8.1 percent of new insureds (see Table 7), and 6.0 percent of new business premium. Less than 20 percent of the lower affinity average premium is attributable to the affinity discount. The balance may be due to younger issue age or less robust coverage. It would seem that, over time, an increasing number of associations would have discounts, which might cause the affinity group percentage to increase over time, but our data does not demonstrate such a pattern.

Characteristics of Policies Sold
Average Premium Per Sale
The average new business (NB) premium per insured (Table 8), subtracting FPOs for the insurers that reported statistics, and the average premium per buying unit (a couple comprise a single buying unit) increased six to seven percent each to $2,706 and $3,847, respectively.

Data for 2018 and earlier years included FPOs in these calculations, overstating the average premium per new insured and buying unit.

Average premium per new policy ranged from $1,029 to $3,792 among the 12 insurers.
The lowest average new premium (including FPOs) was in Puerto Rico ($2,511), followed by Indiana ($2,642) and Kansas ($2,915), while the highest was in the District of Columbia ($5,336), followed by Connecticut ($4,820), and New York ($4,656).
For those insurers which reported results for 2019 and 2020, the average inforce premium (reflecting rate increases, FPO elections and termination of older policies) increased from $2,246 to $2,309, a 2.8 percent increase. However, Table 8 shows a decrease due to a change in insurers reporting inforce premium.

Issue Age
Table 9 summarizes the distribution of sales by issue age band based on insured count. The average issue age remained 57.7. All insurers reported an average issue age between 55 and 60. The age distributions for 2016 and earlier had more worksite participants than recent years. Note: one survey participant has a minimum issue age of 40, one will not issue below 30, and one will not issue below 25.

Benefit Period
Table 10 summarizes the distribution of sales by benefit period. For the first time since 2014, endless benefit periods registered a measurable percentage, albeit only 0.2 percent. The average benefit period for limited benefit period policies dropped back to 3.75, but if we treat endless benefit periods as 15 years, the average benefit period was 3.82. Because of Shared Care benefits, total coverage was higher than the 3.75 average suggests. Three-year benefit periods accounted for 52.9 percent of the sales.

Monthly Benefit
Monthly determination applied to a record 83.0 percent of 2020 policies (Table 11). With monthly determination, low-expense days leave more benefits to cover high-expense days. One insurer offers only daily determination; one insurer offers a choice; and the other insurers automatically have monthly determination.

Table 12 shows the largest concentration ever in the $4,500-$5,999 initial monthly maximum range (32.3 percent). The average maximum monthly benefit increased slightly to $4,888.

Benefit Increase Features
Table 13 summarizes the distribution of sales by benefit increase feature. Most notably, only 14.6 percent of the policies were sold without a benefit increase feature.

“Indexed Level Premium” policies are priced to have a level premium, but the benefit increase is tied to an index such as the consumer price index (CPI).

As shown in Table 14, we project the age 80 maximum daily benefit by increasing the average initial daily benefit from the average issue age to age 80, according to the distribution of benefit increase features, using current future purchase option (FPO) election rates and a five percent/year offer for fixed FPOs. The maximum benefit at age 80 (in 2042) for our 2020 average 58-year-old purchaser projects to $305/day (equivalent to 2.9 percent compounding). Had our average buyer bought an average 2019 policy a year ago at age 57, her/his age 80 benefit would be $316/day (equivalent to 2.9 percent compounding). Most policyholders seem likely to experience eroding purchasing power over time if cost of care trends exceed three percent.

FPOs are important to insureds in order to maintain purchasing power, and 82 percent of our participants’ 2019 to 2020 FPOs were exercised. The high election rate is noteworthy, considering that the cost increases each year due to larger coverage increases each year, increasing unit prices due to age, and additional price increases due to rate increases.

One insurer had an election rate of 91 percent, two insurers had 68 to 70 percent, two insurers had 43 to 45 percent, and one insurer had 19 percent. It seems clear that higher election rates occur if FPOs are more frequent (every year vs. every 3 years) and are “negative-election” (activate automatically unless the client rejects them) as opposed to “positive-election” (which activate only if the client makes a request). At least some blocks of business demonstrate that policyholders will exercise FPOs if they must do so to continue to receive future offers.

FPOs are also important to insurers, accounting for at least 22 percent of new premium in 2020. Two insurers had nearly half their new premium come from FPOs.

Elimination Period
Table 16 summarizes the distribution of sales by facility elimination period (EP). Ninety-two percent (92 percent) of buyers opt for elimination periods between 84 and 100 days. The percentage of EPs of 100 days or more was 5.2 percent, lower than any other year in the table.

Table 17 shows the percentage of policies with zero-day home care elimination period (but a longer facility elimination period). For insurers offering an additional-cost zero-day home care EP option, the purchase rate is sensitive to the cost.

Table 17 also shows the percentage of policies with a calendar-day EP. It is important to understand that most calendar-day EP provisions do not start counting until a paid-service day has occurred.

Sales to Couples and Gender Distribution
Table 18 summarizes the distribution of sales by gender and single/couple status.

In 2012, the last year in which all sales were unisex, 54.9 percent of buyers were female. In 2013, the female percentage spiked to 57.2 percent as females purchased unisex pricing that was still available. Since unisex pricing has disappeared entirely except in the worksite, the female percentage has plateaued at slightly above 54 percent. The percentage of females varies significantly based on an insurer’s markets. This year, the percentage of females varied from 45 to 63.9 percent among insurers.

The percentage of accepted applicants who purchase coverage when their partners are declined also varies significantly by insurer based on their couples’ pricing and their distribution system. Few insurers are able to report this data.

Shared Care and Other Couples’ Features
Table 19 summarizes sales of Shared Care and other couples’ features.

• Shared care allows one spouse/partner to use the other’s available benefits if their own coverage has been depleted or offers a third independent pool that the couple can share.
• Survivorship waives a survivor’s premium after the first death if specified conditions are met.
• Joint waiver of premium (WP) waives both insureds’ premiums if either qualifies for benefits.

Changes in distribution between carriers can greatly impact year-to-year comparisons in Table 19, because some insurers embed survivorship or joint waiver automatically (sometimes only in some circumstances) while others offer it for an extra premium or do not offer the feature.

In the top half of Table 19, percentages are based on the number of policies sold to couples who both buy (only limited benefit, for Shared Care). The bottom half of Table 19 shows the (higher) percentage that results from dividing by sales of insurers that offer the feature. For insurers reporting Shared Care sales, the percentage of both-buying couples who opted for Shared Care varied from 8 to 89.7 percent. The corresponding percentage of couples with Joint WP varied from 10 to 100 percent and for Survivorship ranged from 2.1 to 14.1 percent.

Table 20 provides additional breakdown on the characteristics of Shared Care sales. As shown on the right-hand side of Table 20, two- to four-year benefit period policies are most likely (26 to 31 percent) to add Shared Care. Partly because three-year benefit periods comprise 53 percent of sales, most policies with Shared Care (61 percent) have three-year benefit periods, as shown on the left side of Table 20.

Above, we stated that Shared Care is selected by 36.4 percent of couples who both buy limited benefit period policies. However, Table 20 shows Shared Care comprised no more than 31 percent of any benefit period. Table 20 has lower percentages because Table 19 denominators are limited to people who buy with their spouse/partner whereas Table 20 denominators include all buyers.

Shared Care is more concentrated in two- to four-year benefit periods (88.4 percent of shared sales) than are all sales (73.0 percent). Couples are more likely to buy short benefit periods because couples plan to help provide care to each other and Shared Care makes shorter benefit periods more acceptable. Single buyers are more likely to be female, hence opt for a longer benefit period.

Existence and Type of Home Care Coverage
Four participants reported sales of facility-only policies, which accounted for 0.7 percent of total sales. One insurer was responsible for more than 80 percent of such sales. Ninety-six percent (96.4 percent) of the comprehensive policies included home care benefits at least equal to the facility benefit. No participant has reported home-care-only sales since 2018.

Other Characteristics
As shown in Table 21, partial cash alternative features (which allow claimants, in lieu of any other benefit that month, to use between 10 and 40 percent of their benefits for whatever purpose they wish) were included in 52.5 percent of sales, including non-participant sales.

Return of premium (ROP) features were included in 10.4 percent of policies. ROP returns some or all premiums (usually reduced by paid LTCI benefits) when a policyholder dies. Approximately 75 percent of ROP features were embedded automatically in the product. Embedded features are designed to raise premiums minimally, typically decreasing the ROP benefit to $0 by age 75.

Nearly sixteen percent (15.8 percent) of policies with limited benefit periods included a restoration of benefits (ROB) provision, which typically restores used benefits when the insured has not needed services for at least six months. Approximately 90 percent of ROB features were automatically embedded.

Insurers must offer shortened benefit period (SBP) coverage, which makes limited future LTCI benefits available to people who stop paying premiums after three or more years. The insurers able to report SBP sales, sold SBP to 3.7 percent of buyers.

Only one insurer issued non-tax-qualified (NTQ) policies, which accounted for 0.1 percent of industry sales.

“Captive” (dedicated to one insurer) agents produced 53.1 percent of the coverages. Brokers produced 46.3 percent and a direct-to-consumer carrier accounted for 0.6 percent. At one time, “captive” agents who sold LTCI tended to specialize in LTCI. Now many are agents of mutual companies.

Sales distribution by jurisdiction is posted on the Broker World website.

Limited Pay and Paid-Up Policies
In 2020, two insurers sold policies that become paid-up in 10 years or less, accounting for 1.1 percent of sales, the highest percentage since 2014, as shown in Table 22.

Because today’s prices are more stable, premium increases are less likely. One of the key reasons for buying 10-year-pay (avoidance of rate increases after the 10th year) is greatly reduced, while the cost of 10-year-pay has increased, making it less attractive than in the past. Nonetheless, limited-pay and single-pay policies are attractive to minimize post-retirement outflow and to accommodate §1035 exchanges.

Seven participants reported that 3.0 percent of their inforce policies are paid-up, a lower percentage than last year because an additional insurer participated.

PARTNERSHIP PROGRAM BACKGROUND
When someone applies to Medicaid for long-term care services, most states with Partnership programs disregard assets up to the amount of benefits received from a Partnership-qualified policy (some Indiana and New York policies disregard all assets). Except for California, states with Partnership programs grant reciprocity to Partnership policies issued in other jurisdictions. Partnership programs are approved in 44 jurisdictions, all but AK, DC, HI, MA, MS, UT, and VT, but MA has a similar program (MassHealth).

Four states (CA, CT, IN and NY) blazed the trail, legislating variations of the Robert Woods Johnson approach (“RWJ” states), whereas “DRA” states use simpler, more consistent rules developed later, in the Deficit Reduction Act of 2005. For example, RWJ states require a separate Partnership policy form, generally still have more stringent benefit increase requirements and sometimes assess a fee for insurers to participate (none of which apply in DRA states).

Approximately 60 percent of Partnership states now allow one percent compounding to qualify for Partnership, which can help low-budget buyers qualify for Partnership and also enables worksite core programs to be Partnership-qualified. A higher percentage of policies would qualify for Partnership in the future if insurers and advisors leverage these opportunities. However, currently only three insurers offer one percent compounding.
Partnership programs could be more successful if:

1. Advisors offer small maximum monthly benefits more frequently to the middle class. For example, a $1,500 initial maximum monthly benefit covers about four hours of home care every two days and, with compound benefit increases, may maintain buying power. Many middle-class individuals would like LTCI to help them stay at home while not “burning out” family caregivers and could be motivated further by Partnership asset disregard. (This approach does not work in RWJ states with high Partnership minimum daily benefit requirements.)
2. Middle-class prospects were better educated about the importance of benefit increases to maintain LTCI purchasing power and to qualify for Partnership asset disregard.
3. The four original Partnership states migrate to DRA rules.
4. More jurisdictions adopt Partnership programs.
5. Programs that privately finance direct mail educational LTCI content from public agencies were adopted more broadly.
6. Financial advisors were to press reluctant insurers to certify their products and offer one percent compounding.
7. More financial advisors were certified. Some people argue that certification requirements should be loosened. At a minimum, the renewal certification process could be improved.
8. Linked benefit products became Partnership-qualified.

PARTNERSHIP PROGRAM SALES
Participants reported Partnership sales in all 44 authorized states except CT and NY. No participant sold Partnership policies in more than 41 states in 2020. Two had Partnership sales in 38 states, three in 29 to 34 states, one in 1 state, and the other has chosen not to certify Partnership conformance.

Insurance brokers do not have access to Partnership policies in CA, CT and NY and from only one insurer in IN. However, in some of those states, consumers can purchase Partnership-qualified coverage from one or two other entities.

In the DRA states, 55 percent of policies qualified for Partnership status. Minnesota (81.4 percent) leads each year. Georgia, Maine and Wisconsin were also above 70 percent.

The original RWJ states had few Partnership sales. In New Mexico, a recent Partnership state, only 4.7 percent of sales qualified, which should increase in the future. Kentucky had only 25.2 percent.

UNDERWRITING DATA
Case Disposition
Seven insurers contributed application case disposition data to Table 23. In 2020, 57.8 percent of applications were placed, the lowest ever, seemingly impacted by the pandemic. One insurer reported a 76.4 percent placement rate; all others were below the average, with three insurers between 40 and 47 percent.

Decline rates and suspended/withdrawn rates hit records. While we thought the suspended/withdrawn rate would rise with the pandemic, of the six insurers who have reported such data for both 2019 and 2020, two had a noticeably lower suspended/withdrawn rate in 2020 and two were +/-0.1 percent.

Decline rate by carrier varied from 11.9 to 38.4 percent, but all but two insurers were between 23.2 and 29.1 percent, varying based on factors such as age distribution, distribution system, market, underwriting requirements, and underwriting standards. Our placed percentages reflect the insurers’ perspective. A higher percentage of applicants secures coverage because applicants denied by one carrier may be issued either stand-alone or combo coverage by another carrier or may receive coverage with the same insurer after a deferral period.

Table 24 shows that, compared to 2019, the placement rate increased below age 60 and decreased above age 60, which seems consistent with a theory that the lower placement rate was related to the pandemic. This data is a subset of the placement data in Table 23.

Low placement rates increase insurers’ cost per placed policy. More importantly, low placement rates discourage advisors from discussing LTCI with clients. In addition to not wanting to waste time and effort, advisors fear that declined clients will be dissatisfied. In the Market Perspective section, we listed ways to improve placement rates. This is a critical issue for the industry. If readers have suggestions, they are invited to contact the authors.

Underwriting Tools
Six insurers contributed data to Table 25, which divides the number of uses of each underwriting tool by the number of applications processed. For example, the number of medical records was 82 percent of the number of applications. That does not mean that 82 percent of the applications involved medical records, because some applications resulted in more than one set of medical records being requested.

Insurers are trying to speed underwriting to increase placement rates. In the worksite market, insurers are less likely to use some of these tools.

Year-to-year changes in distribution of sales among insurers significantly impact results. Lower maximum ages result in fewer face-to-face exams. Insurers might underreport the use of an underwriting tool because they may lack a good source for that statistic. For example, an insurer might not be able to split phone interviews by whether or not they include cognitive testing.

The biggest change was the drop in Face-to-Face assessments, perhaps because it was hard to schedule them due to the pandemic.

Underwriting Time
Table 26 shows the average processing from receipt of application to mailing the policy time (40 days) reversed the trend of speedier processing. Of the six insurers that reported this data in both 2019 and 2020, the three fastest processors of 2019 all got faster and the three slower processors of 2019 all got slower, expanding the range from 28.9-60.3 to 26.9-65.9.

Rating Classification
Table 27 shows that a lower percentage of policies was issued in the most favorable rating classification, but the highest percentage was placed in the two most favorable classes since 2012. That’s surprising because prior to 2016, we had more worksite business in the rating classification table. For more information, look for the August issue of Broker World magazine where we will separate data by worksite vs. non-worksite business.

Only 7.7 percent of policies were issued beyond the second-best classification, but three insurers placed 11.0 to 12.4 percent of policies in such a class and another insurer placed 19.2 percent of policies in such a class.

Two insurers placed more than 80 percent of their applicants in the best underwriting class. The other five insurers placed 19 to 25 percent in their best class.

Lastly, Table 27 shows the percentage of decisions which were either declined or placed in the 3rd or less-attractive classification is reducing. Hence, we conclude that industry is making more favorable decisions rather than declining applicants in lieu of giving them a substandard rating,

Tables 28 and 29 show the 2020 percentages of policies issued in the most favorable category and decline decisions by issue age. Tables 27 and 28 do not exactly match Table 26 because some participants provide all-age rating or decline data. The percentage placed in the most favorable classification increased for ages 65 and older, likely statistical fluctuation. Although overall decline rates increased, the by-age decline rates decreased, except for under age 30; that anomaly results because only some insurers can provide decline data by issue age.

CLOSING
We thank insurance company staff for submitting the data and responding to questions promptly. We also thank Sophia Fosdick, Margaret Liang, Nicole Gaspar, and Alex Geanous of Milliman for managing the data expertly.

We reviewed data for reasonableness and insurers reviewed their product exhibit displays. Nonetheless, we cannot assure that all data is accurate.

If you have suggestions for improving this survey (including new entrants in the market), please contact one of the authors.

Reference:
Society of Actuaries (November 2016). Long-Term Care Insurance: The SOA Pricing Project. Retrieved May 16, 2020, from https://www.soa.org/globalassets/assets/files/static-pages/sections/long-term-care/ltc-pricing-project.pdf (PDF).

The 2020 Milliman Long Term Care Insurance Survey is the 22nd consecutive annual review of stand-alone long term care insurance (LTCI) published by Broker World magazine. It analyzes the marketplace, reports sales distributions, and describes available products, including group insurance.

More analysis of worksite sales will appear in the August issue of Broker World magazine.

Unless otherwise indicated, references are solely to U.S. stand-alone LTCI sales, excluding exercised future purchase options (FPOs) or other changes to existing coverage. “Stand-alone” refers to LTCI policies that do not include death benefits (other than returning premiums upon death or waiving a surviving spouse’s premiums) or annuity or disability income benefits. Where referenced, “combo” products provide LTCI combined with life insurance or annuity coverage. “Linked benefit” policies are combo policies which can allow more than the death benefit or annuity account value to be used as LTCI.

Highlights from this year’s survey

Participants
Nine carriers participated broadly in this survey. Six others provided sales information so we could report more accurate aggregate industry individual and multi-life sales.

We estimate our statistical distributions reflect about 75 percent of total industry policies sold (85 percent of premium) but only about 12 percent of worksite policies sold (28 percent of premium). Our worksite statistical distributions can vary significantly from year-to-year because insurers focusing on particular worksite markets may be over- or under-represented. Last year, we commented that the worksite business in our statistical analysis was overly weighted toward executive carve-out programs. This year, that imbalance has increased. The carriers which provided 2019 statistical data had an average worksite annual premium of $4,568 whereas the ones which provided only total sales had an average worksite premium of $1,627. The corresponding 2018 averages were $2,015 and $1,561, respectively. In 2017, the difference was in the other direction ($1,015 and $1,441, respectively).

National Guardian Life contributed statistical distributions for the first time, but CalPERS and Genworth did not provide statistical distributions this year. Country Life, which has been a consistent participant in this survey, discontinued LTCI sales effective May 1, 2020.

The insurers displayed in the Product Exhibit all provided broad statistical information, although they were not all able to provide all the data we requested. Northwestern also provided such information. Auto-Owners, CalPERS, Genworth, LifeSecure, Transamerica and United Security Assurance contributed total and worksite sales (new premium and lives insured) but did not provide other information. However, in a few places we were able to reflect some of their product designs in our statistical distributions.

Sales Summary

• The 15 carriers reported sales of 54,563 policies and certificates (“policies” henceforth) with new annualized* premium of $170,770,732 (including exercised FPOs) in 2019, which compared to 2018 industry sales of 56,287 policies with new annualized* premium of $169.7 million (both slightly restated from last year’s report), a 3.1 percent drop in the number of policies but a 0.6 percent increase in new annualized premium, the first increase in new premium since 2012. As noted in the Market Perspective section, sales of policies combining LTCI with other risks dropped slightly. *Single premium sales are counted at 10 percent for the annualized premium calculations herein.
• Seven of the top 10 insurers sold more new premium than in 2018, while four sold more new policies. This difference resulted largely because elected FPOs add premium but not new policies.
• Worksite LTCI rebounded in 2019. The three stand-alone LTCI carriers specializing in worksite each saw sales jump by 62 to 76 percent despite increased competition from a major new carrier and from linked-benefit policies.
• With FPO elections included in new premium, Northwestern garnered the number one spot in sales. Mutual of Omaha was a strong second and had a large lead in annualized premium from new policies sold. Together, these two companies combined for 59 percent of new premium including FPOs, compared to 57 percent last year.
• Based on our analysis, for the fifth straight year (and fifth time ever), our participants’ number of inforce policies dropped, but less than one percent.
• Despite the reduced number of policies and increased number of policies in paid-up status, inforce premium increased 10.0 percent and the average inforce premium per policy rose from $2,208 to $2,426. Inforce premium rises due to sales, price increases, and benefit increases (including FPOs), and reduces from lapses, reductions in coverage, deaths, and shifts to paid-up status for various reasons.

Participants’ claims rose 8.1 percent. Overall, the stand-alone LTCI industry incurred $11.6 billion in claims in 2018 based on companies’ statutory annual filings, raising total incurred claims from 1991 through 20187 to $141.5 billion. (Note: 2018 was the most recent year available when this article was written.) Most of these claims were incurred by insurers that no longer sell LTCI. The reported 2018 incurred claims are 5.5 percent higher than the $11.0 billion of incurred claims reported in 2017.

The placement rate rose to 59.2 percent, higher than 58.8 percent in 2018 and 59.0 percent in 2017. The small increase occurred despite the increase in average age. The more favorable and faster underwriting decisions documented below contributed to the higher placement rate. As discussed in the Market Perspective section, the industry may be able to increase placement rates further, which should make financial advisors more comfortable suggesting that their clients consider LTCI, all else equal.

About the Survey
This article is arranged in the following sections:

• Highlights provides a high-level view of results.
• Market Perspective provides insights into the LTCI market.
• Claims presents industry-level claims data.
• Sales Statistical Analysis presents industry-level sales distributions reflecting data from 11 insurers.
• Partnership Programs discusses the impact of the state partnerships for LTCI.

Available at www.BrokerWorldMag.com:

• Product Exhibit shows, for eight insurers: Financial ratings, LTCI sales and inforce, and product details. Please note that, during the COVID-19 pandemic, some insurers have temporarily discontinued sales that would require face-to-face interviews. Our Exhibit ignores such temporary restrictions.
• Product Details, a row-by-row definition of the product exhibit entries, with some commentary.
• Premium Exhibit shows lifetime annual premiums for each insurer’s most common underwriting class, for issue ages 40, 50, 60, and 70 for single females, single males, and heterosexual couples (assuming both buy at the same age), based on $100 per day (or closest equivalent weekly or monthly) benefit, 90-day facility and most common home care elimination period, three-year and five-year benefit periods or $100,000 and $200,000 maximum lifetime buckets, with and without Shared Care and with flat benefits or automatic 3 or 5 percent annual compound benefit increases for life. Worksite premiums do not reflect any worksite-specific discount, though some carriers offer this.
• Premium Adjustments (from our Premium Exhibit prices) by underwriting class for each participant.
• Distribution by underwriting class for each participant
• State-by-state results: percentage of sales by state, average premium by state and percentage of policies qualifying for Partnership by state.

Market Perspective (more detail in subsequent parts of the article)

• Since 2009 (varies by jurisdiction), if an insurer concludes that a claimant is not chronically ill, the insurer must inform the claimant of his/her right to appeal the decision to independent third-party review (IR). The IR determination is binding on insurers. As shown in our Product Exhibit, most participants have extended IR beyond statutory requirements, most commonly to policies issued prior to the effective date of IR. It is hard to get data relative to IR. Regulators do not collect statistics and insurers often do not track it. It has been a positive sign that few IR requests have been made and that insurers seem to have been upheld approximately 90 percent of the time based on data we have received over the years from insurers and from Steve LaPierre, President of LTCI Independent Eligibility Review Specialists, LLC (LTCIIERS). Mr. LaPierre reported a 40 percent drop in IR requests in 2019, but an increase to approximately 30 percent of cases in which insurers were overruled. Part of the increase may be explained by statistical fluctuation because the number of IRs is low. Generally, if the appeal rate drops, the strongest cases are likely to continue to be appealed, resulting in a higher percentage being overruled. Another factor, in the experience of one of the authors, is claims examiners’ unfamiliarity with old policy forms in which standard wording was modified to fit a state-specific requirement, a problem that can be exacerbated if the claim function is transferred to a new entity lacking the historical knowledge. Because the number of IRs is so low, even a small percentage of such errors could cause the IR overturn rate to spike.
• Third Party Notification (TPN) provisions provide another consumer protection. When someone incurs dementia, bills often accumulate unpaid. Under contract law, insurers could then terminate policies for non-payment of premium, despite the cause of non-payment being unknown claim qualification. In response to this issue, insurers and regulators agreed long ago to encourage applicants to name a third-party who must be notified before a policy can be terminated for non-payment of premiums. Insurers do not contact such a third-party until the grace period is over, and then provide at least another 30 days for the third-party to stave off cancellation. Every two years, insurers give policyholders an opportunity to update the contact information or change the third party. The benefit of this provision to policyholders often receives little attention, perhaps partly because no one seems to have ever studied the impact.
  
  This year, we asked some one-time questions related to TPNs. Only two participants were able to provide data. One had TPNs identified for 63 percent of its inforce policies and the other had TPNs identified for 77 percent of its policies. One-fifth of the policyholders who had identified a TPN increased their protection by identifying more than one TPN. Only about 10 percent of single purchasers have named a TPN, despite seemingly having a greater need for a TPN. Financial advisors could serve an important role by stressing the importance of a TPN for single people. Not surprisingly, fewer than half of the policies with third-party billing have TPNs. Thus direct-billed policies and couples have distinctly higher percentages of TPNs than indicated above.
• Current prices are more stable than past prices because today’s prices reflect much more conservative assumptions based on far more credible data1 and lower assumed investment yields. Many financial advisors presume that new policies will face steep price increases, hence can be reluctant to encourage their clients to consider LTCI. Although three participants have never increased premiums on policies issued under “rate stabilization” laws, it may take a long time before the market is comfortable that prices are stable.
• Linked benefit products have increased market share, due to influences such as: i) regardless of whether the policyholder has a long-term care claim, their family will receive benefits, ii) they often have guaranteed premiums and benefits, iii) they no longer need to be funded with single premiums; and iv) they are issued to younger ages than before.
  
  According to LIMRA, in the first half of 2019 (the most recent data available when this article went to print), combination life sales dropped slightly, even in the recurring-premium market which continued to gain market share compared to single premiums. The biggest drop (six percent both in premium and policies) was in the Chronic Illness market.
• As noted earlier, roughly 59 percent of applications have resulted in placed policies in the past three years, with the low placement rates contributing to financial advisors being hesitant to recommend that clients consider LTCI. The industry may be able to improve placement rates in a variety of ways.
  • Utilize E-applications for faster submission and reduced processing time by assuring that all apps are “in good order.” The speed and accuracy increase placement. It would be helpful if brokerage general agents could populate an eApp with quote information, then forward that partial eApp to the broker for completion.
  • Pre-qualify an applicant’s health. It is important to stress the importance of such pre-qualification to advisors and prospects and to continue to find effective ways to accomplish quality pre-qualification.
  • More effective education of their distribution system by insurers, such as with drill-down questions in on-line underwriting guides and eApps.
  • Require cash with the application (CWA), which led to about five percent more of the apps getting placed according to our 2019 survey.
  • Continue to improve messaging regarding the value of LTCI and of buying now (rather than in the future). Such messaging would increase the number of applications and improve the placement rate by attracting younger and healthier applicants.
• Over the past two years, 79 percent of our participants’ policyholders exercised their FPO (future purchase option, a guarantee that, under specified conditions, a policyholder can purchase additional coverage without having to demonstrate good health). As both the additional coverage and price per unit increase over time, such options become increasingly expensive, even more so with the price increases that the industry has experienced. The high election rate demonstrates the importance of the coverage to the policyholder and the effectiveness of annual (as opposed to triennial) negative-election FPOs (negative-election FPOs activate automatically unless the client rejects them, as opposed to positive-election FPOs which activate only if the client makes a request). At least in some markets and with some designs, policyholders reliably exercise FPOs when they must do so to continue to receive future offers. Considering such FPOs and other provisions, we project a maximum benefit at age 80 of $306/day for an average 58-year-old purchaser in 2019, which is equivalent to an average 2.9 percent compounded benefit increase between 2019 and 2041. Purchasers may be disappointed if the purchasing power of their LTCI policies deteriorates over time.
• Only four participants offer coverage in all U.S. jurisdictions and no worksite insurer does so. Insurers continue to be reluctant to sell in jurisdictions which are slow to approve new products, restrict rate increases, or have unfavorable legislation or regulations.
• Eight of our nine participants use third party administrators (TPAs) and seven use reinsurers. We thank American United, LifeCare Assurance, Manufacturers, Munich American, RGA, Wilton Re, Long-Term Care Group, Life Plans, and CHCS for their contributions to the LTCI industry. Other reinsurers and TPAs support insurers not in our survey. In some cases, affiliated companies provide reinsurance or guarantees.
• As we reported last year, the NAIC adopted a significantly improved 2019 version of the Shopper’s Guide, covering combination products as well as stand-alone LTCI and adding other information, while reducing the number of pages from 80 to 76. We asked last year’s readers to email [email protected] to comment on the new Guide and how (much) it is used to educate consumers. We received no comments. This year, we polled insurers; their answers suggest they view the Shopper’s Guide primarily as a regulatory requirement. Their suggestions to improve the usefulness for consumers included shortening it, linking to it from state department of insurance websites, and regulators using the guide in a campaign to encourage long term care planning. We visited 31 jurisdiction websites to see if they linked to the NAIC Shopper’s Guide. We found that 13 jurisdictions use their own guides. Five (plus the Federal government) link to the 2019 NAIC Guide, while four link to the 2013 NAIC Guide. One state links to an AHIP Guide, and we found no guide or minimal wording in the remaining eight states.

Claims

• Eight participants reported 2019 claims. As some companies are not able to provide detailed data, some statistics are more robust than others.
• The eight insurers’ combined claim payments on individual policies rose 8.1 percent in 2019 over 2018, despite only a 0.2 percent increase in inforce policies and a 3.7 percent increase in their inforce premium.
• The LTCI industry has had a much bigger impact than indicated above, because a lot of claims are paid by insurers that do not currently sell LTCI or did not submit claims data to us.

LTCI claims paid by insurers no longer selling LTCI may differ significantly from the following statistics as their claimants are more likely to have facility-only coverage, be older, etc.

Table 1 shows the total dollar and number of reported individual LTCI claims. The large drop in claims is primarily the result of having Genworth data in 2018 but not in 2019.

Table 2 shows the distribution of those claims by venue, which have shifted away from nursing homes over the years (except in 2019) due to consumer preferences and more claims coming from comprehensive policies. Regarding the change in 2019, we note that, for insurers that contributed data in both 2018 and 2019, the number of claims in nursing homes dipped 0.6 percent following the prior years’ patterns.

The inception-to-date data shows 53.3 percent of the number of claims being in Nursing Homes (vs. 45.0 percent last year). Here again, if we limit data to insurers reporting each year, we see a drop of 0.8 percent instead of an increase of 8.3 percent. Surprisingly, the inception-to-date number of claims is more weighted to Nursing Home than the dollar of claims. The nursing home claims seem to come from older policies with lower minimums.

In the distribution based on number of claims, a person who received care in more than one venue is counted once for each venue, but not double-counted in the total line.

Seven carriers reported their number of open individual claims at year-end, ranging between 59 and 107 percent of the number of claims paid during the year, averaging 75 percent overall. The insurer reporting 107 percent noted that they include open claims that have had no payments made yet. With that carrier excluded, the overall ratio was 76 percent.

Table 3 shows average size individual claims since inception. Because 41 percent of claimants since inception have submitted claims from more than one type of venue, the average total claim generally exceeds the average claim paid for any particular venue. Nonetheless, individual Assisted Living Facility (ALF) claims stand out as high each year, probably because:

a) ALF claims appear to have a longer duration compared with other venues.

b) Nursing home costs are most likely to exceed the policy daily/monthly maximum. Hence the maximum daily benefit negates part of the additional daily cost of nursing homes.

c) People who maximize the use of their maximum monthly benefits can generally spend as much in an ALF as in a nursing home.

d) Although some surveys report that ALFs cost about half as much as nursing homes on average, ALFs often charge more for a memory unit or for levels of assistance that align more closely with nursing home care. Upscale ALFs seem to cost a higher percentage of upscale nursing home costs than the average ALF/nursing home ratio.

Several insurers extended ALF coverage to policies which originally did not include ALF coverage. Providing these additional benefits provides the policyholder with significant flexibility at time of claim, but has contributed to the insurers’ need for rate increases.

The following factors contribute to a large range of average claim results by insurer (see Table 3):

• Different markets (by affluence; worksite vs. individual; geography; etc.).
• Demographic differences (distribution by gender and age).
• Distribution by benefit period, benefit increase feature, shared care and elimination period.
• Distribution by facility-only policies vs. 50 percent home care vs. 100 percent home care vs. home care only, etc.
• Different lengths of time in the business.

The following factors cause our average claim sizes to be understated.

• For the carriers reporting claims this year, 11 percent of the inception-to-date individual claims are still open. Our data does not include reserve estimates for future payments on open claims.
• People who recover, then claim again, are counted as multiple insureds, rather than adding their various claims together.

Besides being understated, average claim data does not reflect the value of LTCI from some purchasers’ perspectives, because the many small claims drive down the average claim. LTCI can provide significant financial return for people who need care one year or longer. The primary purpose of insurance is to protect against adverse results, so the amount of protection, as well as average claim, is important.

Six insurers provided their current individual (excludes group) monthly LTCI claim exposure, which exceeds $4.8 billion (note: reflects only initial monthly maximum for one insurer). As shown in Table 4, this figure is thirty times their corresponding monthly LTCI premium income and more than 47 times their 2019 LTCI monthly paid claims.

Eight insurers contributed data regarding their inforce distribution by benefit period. Treating endless (lifetime benefit periods) as a 15-year benefit period, we found that their average inforce benefit period is 6.8 years. Changing the assigned value of the endless benefit period by one year has an impact of approximately .25 years on the average inforce benefit period. With annual exposure thirty times annual premium and assuming an average benefit period of about 6.8 years, we estimate that total exposure is 202 times annual premium.

Four insurers reported their current average individual maximum monthly maximum benefit for claimants, with results ranging from $4,696 to $6,552.

Nursing home (NH) claims are more likely to use the policy’s maximum daily/monthly benefit than ALF claims, because ALF costs are generally lower and because policies sometimes have lower maximums for ALFs. ALF claims correspondingly are more likely to use the policy maximum than are adult day care and home care claims.

STATISTICAL ANALYSIS
Nine insurers contributed significant background data, but some were unable to contribute data in some areas. Six other insurers (Auto-Owners, CalPERS, Genworth, LifeSecure, Transamerica and United Security) contributed their number of policies sold and new annualized premium, distinguishing worksite from other sales, but not clarifying whether FPOs were included in the premium.

Sales characteristics vary significantly among insurers based on market differences (individual vs. worksite, affluence, gender distribution, etc.). Year-to-year variations in policy feature distributions may reflect changes in participants, participant practices and designs, participant or worksite market shares and industry trends.

Market Share
We include purchased increases on existing policies in new premium (we call them FPOs and include board-approved increase offers) because new coverage is being issued. However, looking at new premiums ignoring FPOs spotlights insurers with more new policy sales. Table 5 lists the top 10 participants in 2019 new premium including FPOs and shows their sales (if submitted) without FPOs. Northwestern ranks #1 including FPOs, with Northwestern and Mutual of Omaha accounting for 59 percent of the market. Ignoring FPOs, Mutual of Omaha is #1, 63 percent ahead of #2 Northwestern. The premium includes 100 percent of recurring premiums plus 10 percent of single premiums.

Worksite Market Share
The three stand-alone LTCI carriers specializing in worksite each saw sales jump by 62 to 76 percent despite increased competition from a major new carrier and from linked-benefit policies. Thus, worksite business produced 21.0 percent of new insureds (see Table 6) and 13.4 percent of new annual premium (including FPOs but excluding single premium). Worksite sales consist of three different markets, the first two of which produce a higher percentage of new insureds than of new premiums:

• Voluntary group coverage generally is less robust than individual coverage.
• Core/Buy-Up programs have particularly young age distributions and modest coverage because a lot of people do not buy-up and are less likely to insure spouses.
• Executive carve-out programs generally provide the most robust coverage. One- or two-couple executive carve-out sales may not qualify for a multi-life discount with some insurers, hence may not be labeled as worksite sales in submissions to our survey.

The amount of worksite sales reported and its distribution among the three sub-markets significantly impact product feature sales distributions. Table 6 is indicative of the full market, but this year’s policy feature distributions significantly underweight the large group market, as most of that market is not reflected in our statistical data. More information about worksite sales will appear in the August issue of Broker World.

Affinity Market Share
Affinity groups (non-employers such as associations) produced 8.5 percent of new insureds (see Table 7), but only 5.9 percent of new business premium. Less than 20 percent of the lower affinity average premium is attributable to the affinity discount. The balance may be due to younger issue age or less robust coverage.

Characteristics of Policies Sold
Average Premium Per Sale
Table 8 shows the average new business premium per insured ($2,551), subtracting FPOs for the insurers that reported statistics. For insurers that reported distribution by couples vs. individuals, the number of insureds was 41.5 percent higher than the number of buying units (a couple comprise a single buying unit), boosting the $2,551 to $3,608.

The average premium for new policies for the 15 insurers rose 1.7 percent compared to 2018 (we recalculated the 2018 results to include more carriers). Our statistics show a higher average initial maximum monthly benefit, a higher issue age and a slightly higher benefit period, which generate a larger increase in average premium for the insurers which provided statistical data.

Data for 2017 and earlier years included FPOs in these calculations, overstating the average premium per new insured and buying unit.

Average premium per new policy ranged from $1,443 to $3,701 among the 15 insurers.
The lowest average new premium (including FPOs) was in Puerto Rico ($2,036), followed by Oklahoma ($2,586) and Kansas ($2,646), while the highest was in the District of Columbia ($4,575), followed by Connecticut ($4,531), West Virginia ($4,349), and New York ($4,270).
Due to rate increases, FPO elections and termination of older policies, the average inforce premium jumped to $2,426, 9.8 percent more than at the end of 2018.

Issue Age
Table 9 summarizes the distribution of sales by issue age band based on insured count. The average issue age increased to 57.7 mostly due to change in participants. The age distributions for 2016 and earlier had more worksite participants than recent years. Note: Two survey participants have a minimum issue age of 40, one will not issue below 30, and two will not issue below 25.

Benefit Period
Table 10 summarizes the distribution of sales by benefit period. The average notional benefit period increased slightly from 3.74 to 3.79 because of change in participants. Because of Shared Care benefits, total coverage was higher than the 3.79 average suggests. Three-year to five-year benefit periods accounted for 75.2 percent of the sales, quite a concentration compared to the past.

Monthly Benefit
Table 11 shows that, because of the change in participants, monthly determination applied to a record 83.2 percent of 2019 policies. With monthly determination, low-expense days leave more benefits to cover high-expense days. Among companies which make monthly determination optional, the percentage electing monthly determination ranged from 16 to 43 percent.

Table 12 shows that the average maximum monthly benefit increased from $4,763 in 2018 to $4,882 in 2019, two-thirds of which increase resulted from the change in participating insurers.

Benefit Increase Features
Table 13 summarizes the distribution of sales by benefit increase feature. The big increase in three percent compounding is less meaningful considering that there was a big decrease in “Other compound” most of which was a variation of three percent compounding.

“Indexed Level Premium” policies are priced to have a level premium, but the benefit increase is tied to an index such as the consumer price index (CPI).

As shown in Table 14, we project the age 80 maximum daily benefit by increasing the average initial daily benefit from the average issue age to age 80, according to the distribution of benefit increase features, using current future purchase option (FPO) election rates and a five percent/year offer for fixed FPOs. The maximum benefit at age 80 (in 2041) for our 2019 average 58-year-old purchaser projects to $306/day (equivalent to 2.9 percent compounding). Had our average buyer bought an average 2018 policy a year ago at age 57, her/his age 80 benefit would be $313/day (equivalent to three percent compounding). The two percent drop is attributable to the change in participants this year. The increase in the FPO election rate added $4/day to the 2041 benefit. Most policyholders seem likely to experience eroding purchasing power over time if cost of care trends exceed three percent.

FPOs are important to insureds in order to maintain purchasing power, and as demonstrated by the fact that 82 percent of our participants’ 2018-2019 FPOs were exercised. The high election rate is very impressive, considering that the cost increases each year due to larger coverage increases each year, increasing unit prices due to age and additional price increases due to rate increases.

The high election rate, coupled with a range of responses from 20 to 90 percent election, demonstrates the effectiveness of annual (as opposed to triennial) options and negative-election (negative-election FPOs activate automatically unless the client rejects them) as opposed to positive-election FPOs (which activate only if the client makes a request). At least some blocks of business demonstrate that policyholders will exercise FPOs if they must do so to continue to receive future offers.

FPOs are also important to insurers, accounting for at least 18 percent of new premium in 2018 and at least 22 percent in 2019. Two insurers had nearly half their new premium come from FPOs.

Elimination Period
Table 16 summarizes the distribution of sales by facility elimination period (EP). Ninety-one percent (91 percent) of buyers opt for 90-day elimination periods. However, two carriers report 23 to 30 percent of their sales in the 20-44-day EP range. Three carriers reported six to 11 percent of their sales having more than a 200-day EP.

Table 17 shows the percentage of policies with zero-day home care elimination period (but a longer facility elimination period). With most insurers offering an additional-cost zero-day home care EP option, 15 to 30 percent of buyers purchased a zero-day home care elimination period, but one insurer had nearly a 50 percent election rate.

Table 17 also shows the percentage of policies with a calendar-day EP. It is important to understand that most calendar-day EP provisions do not start counting until a paid-service day has occurred.

Sales to Couples and Gender Distribution
Table 18 summarizes the distribution of sales by gender and single/couple status.

The percentage of buyers who are female dropped to 54.2 percent and the percentage of females among single insureds dropped to 64.8 percent. Changes in participants accounted for only 33 and 17 percent of those respective drops. The percentage of females varies significantly and consistently among companies each year based on their markets.

The 81.8 percent of accepted applicants who purchased coverage when their partners were declined is the highest we have seen. However, some insurers which reported approximately 50 percent success in the past did not contribute data this year.

Shared Care and Other Couples’ Features
Table 19 summarizes sales of Shared Care and other couples’ features.

• Shared care—allows one spouse/partner to use the other’s available benefits if their own coverage has been depleted or offers a third independent pool that the couple can share.
• Survivorship—waives a survivor’s premium after the first death if specified conditions are met.
• Joint waiver of premium (WP)—both insureds’ premiums are waived if either qualifies for benefits.

Changes in distribution by carrier can greatly impact year-to-year comparisons in Table 19, because some insurers embed survivorship or joint waiver automatically (sometimes only in some circumstances) while others offer it for an extra premium or do not offer the feature. However, the 2019 increased percentages of Shared Care and Joint WP are not attributable to changes in participants.

In the top half of Table 19, percentages are based on the number of policies sold to couples who both buy (only limited benefit, for Shared Care). The bottom half of Table 19 shows the (higher) percentage that results from dividing the number of buyers by sales of insurers that offer the feature. Four participants sold Shared Care to more than half their limited benefit period couples; four others sold Shared Care to fewer than half of such couples. Three insurers sold Joint WP to 50 percent or more of their couples; three others sold joint WP to fewer couples. Three carriers sold Survivorship ranging to nine to 16 percent of their couples, while two insurers sold Survivorship to 2.5 to three percent of their couples.

Table 20 provides additional breakdown on the characteristics of Shared Care sales. As shown on the right-hand side of Table 20, two- to four-year benefit period policies are most likely (26 to 32 percent) to add Shared Care. Partly because three-year benefit periods comprise 52 percent of sales, most policies with Shared Care (62 percent) have three-year benefit periods, as shown on the left side of Table 20.

Above, we stated that Shared Care is selected by 35.9 percent of couples who both buy limited benefit period policies. However, Table 20 shows Shared Care comprised no more than 31 percent of any benefit period. Table 20 has lower percentages because Table 19 denominators are limited to people who buy with their spouse/partner whereas Table 20 denominators include all buyers.

Shared Care is more concentrated in two- to four-year benefits periods (88.3 percent of shared sales) than are all sales (71.5 percent). Couples are more likely to buy short benefit periods because couples plan to help provide care to each other and Shared Care makes shorter benefit periods more acceptable. Single buyers are more likely to be female, hence opt for a longer benefit period. Shared Care percentages rebound at the longer benefit periods reflecting buyers who are trying to cover catastrophic risk and might prefer an endless benefit period.

Existence and Type of Home Care Coverage
For the first time, no participant reported home-care-only sales. Five participants reported sales of facility-only policies, which accounted for 0.7 percent of total sales. Ninety-six percent (96.4 percent) of the comprehensive policies included home care benefits at least equal to the facility benefit.

Other Characteristics
As shown in Table 21, partial cash alternative features (which allow claimants, in lieu of any other benefit that month, to use between 30 and 40 percent of their benefits for whatever purpose they wish) were included in 49.1 percent of sales, including some non-participant sales.

Return of premium (ROP) features were included in 10.4 percent of policies. ROP returns some or all premiums (usually reduced by paid LTCI benefits) when a policyholder dies. Approximately 80 percent of ROP features were embedded automatically in the product. Embedded features are designed to raise premiums minimally, typically decreasing the ROP benefit to $0 by age 75.

Nearly fourteen percent (13.6 percent) of policies with limited benefit periods included a restoration of benefits (ROB) provision, which typically restores used benefits when the insured does not need services for at least six months. Approximately 89 percent of ROB features were automatically embedded.

Shortened benefit period (SBP) makes limited future LTCI benefits available to people who stop paying premiums after three or more years. Although every insurer is obligated to offer SBP, some carriers did not report any SBP sales. Removing their sales from the denominator, we found that 3.1 percent of buyers selected SBP.

Only one insurer issued non-tax-qualified (NTQ) policies, which accounted for 0.1 percent of industry sales.

“Captive” (dedicated to one insurer) agents produced 57.0 percent of the policies. The balance was produced by brokers. At one time, “captive” agents who sold LTCI tended to specialize in LTCI. Now many are agents of mutual companies.

Sales distribution by jurisdiction is posted on the Broker World website.

Limited Pay and Paid-Up Policies
In 2019, only two insurers sold policies that become paid-up in 20 years or less, accounting for 0.8 percent of sales.

Because today’s prices are more stable, premium increases are less likely. One of the key reasons for buying 10-year-pay (avoidance of rate increases after the 10th year) is greatly reduced, while the cost of 10-year-pay has increased, making it less attractive than in the past. Nonetheless, limited-pay and single-pay policies are attractive to minimize post-retirement outflow and to accommodate §1035 exchanges.

Participants reported that 3.8 percent of their inforce policies are paid-up. Less than one percent (0.8 percent) are paid-up because they have completed their premium period. Another 0.8 percent are paid-up due to shortened benefit period and 0.4 percent are due to survivorship features. Almost half of the paid-up policies are paid up for unidentified reasons. For example, shortened benefit period percentages could be as much as three times as high as suggested herein.

PARTNERSHIP PROGRAM BACKGROUND
When someone applies to Medicaid for long term care services, states with Partnership programs disregard assets up to the amount of benefits received from a Partnership-qualified policy. Partnership sales were reported in 43 jurisdictions in 2019, all but California (no participants offer California Partnership policies) and Alaska, District of Columbia, Hawaii, Massachusetts, Mississippi, Utah, and Vermont, where Partnership programs do not exist. Massachusetts has a somewhat similar program (MassHealth).

The Partnership rules in California, Connecticut, Indiana and New York (“original” Partnership states) are significantly different than in other Partnership jurisdictions (“Deficit Reduction Act (DRA)” jurisdictions). The “original” states legislated variations of the Robert Woods Johnson Partnership (RWJ) proposal, whereas the “DRA” jurisdictions use more consistent rules based on the Deficit Reduction Act of 2005. For example, the “original” states require a separate Partnership policy form, generally still have more stringent benefit increase requirements and assess a fee for insurers to participate (none of which applies in DRA states). As a result, only two to five insurers sell Partnership policies in CA (two), CT (four), IN (five) and NY (two). At the time that this article is written, insurance brokers do not have access to Partnership policies in CA and NY.

The National Reciprocity Compact (NRC) requires member states to recognize Medicaid Asset Disregard earned in any other member state. States creating Partnerships under the Deficit Reduction Act of 2005 were automatically enrolled in the NRC but had the right to secede. The four original Partnership states (California, Connecticut, Indiana and New York) had the right to opt in. California is now the only jurisdiction with a Partnership program that is not a member of the NRC. Last year, we reported that New Hampshire had created a unique regulation in 2018, limiting asset disregard for policies sold in other jurisdictions. New Hampshire has reversed that limitation.

PARTNERSHIP PROGRAM SALES
Insurers sometimes delay certifying policy forms as “Partnership” because of other priorities (e.g., needing time to comply with state-specific requirements to notify existing policyholders or offer an exchange). Such delay is not harmful, as certification is retroactive to policies already issued on that policy form if the policies have the required characteristics. For this reason, the “original” Partnership issues mentioned above and because some insurers are not licensed in all jurisdictions, none of our participants sold Partnership policies in more than 40 jurisdictions in 2019. Three had Partnership sales in 38-40 jurisdictions, four in 29-34 jurisdictions, one in six jurisdictions, and the other has chosen not to certify Partnership conformance.

In the DRA states, 55 percent of policies qualified for Partnership status. Once again, Minnesota led the way; 82.5 percent of the policies sold qualified as Partnership. Maine (75 percent) and Wisconsin (74 percent) followed. North Dakota and Georgia were also above 70 percent.
In the original RWJ states, only 0.4 percent of the issued policies qualified as Partnership: Indiana (3.5 percent), Connecticut (1.0 percent), New York (0.4 percent) and California (zero percent).

Partnership policies are expected to have higher average premiums because of the requirement that Partnership policies issued under age 76 have benefit increase features. In our survey, most states did not have such a relationship because we asked insurers to include FPOs in their sales and the carriers with the most FPOs had high average premiums while selling few Partnership policies.

Approximately 60 percent of Partnership states now allow one percent compounding to qualify for Partnership, which can help low-budget buyers qualify for Partnership and also enables worksite core programs to be Partnership-qualified. A higher percentage of policies would qualify for Partnership in the future if insurers and advisors leverage these opportunities. However, currently only three insurers offer one percent compounding.

Partnership programs could be more successful if:

1. Advisors offer small maximum monthly benefits more frequently to the middle class. For example, a $1,500 initial maximum monthly benefit covers about four hours of home care every two days and, with compound benefit increases, may maintain buying power. Many middle-class individuals would like LTCI to help them stay at home while not “burning out” family caregivers and could be motivated further by Partnership asset disregard. (This approach does not work where a Partnership minimum daily benefit is required such as CA, $230/day; CT, $291; or NY, $337.) When policies reflecting CA SB 1248 become available in California, California’s minimum size policy will drop to $100/day. South Dakota requires a $100 minimum size for all LTCI and Indiana’s Partnership requires $115.)
2. Middle-class prospects were better educated about the importance of benefit increases to maintain LTCI purchasing power and to qualify for Partnership asset disregard.
3. The four original Partnership states migrate to DRA rules.
4. More jurisdictions adopt Partnership programs.
5. Programs that privately finance direct mail educational LTCI content from public agencies were adopted more broadly.
6. Financial advisors were to press reluctant insurers to certify their products and create one percent compounding.
7. More financial advisors were certified. Some people argue that certification requirements should be loosened. At a minimum, the renewal certification process could be improved.
8. More insurers offer one percent compounding.
9. Linked benefit products became Partnership-qualified.

UNDERWRITING DATA
Case Disposition
Eight insurers contributed application case disposition data to Table 22. In 2019, 59.2 percent of applications were placed, including those that were modified, a little higher than in 2017 and 2018. Half of the improvement was related to change in participants.

One insurer reported a 76.7 percent placement rate, the second highest being 58.3 percent. Two insurers placed fewer than 50 percent of their apps, the lowest placement rate being 32.3 percent. Low placement rates increase insurers’ cost per placed policy. More importantly, low placement rates can discourage advisors from discussing LTCI with clients. In addition to not wanting to waste time and effort, advisors fear that declined clients will be dissatisfied.
Decline rates slipped slightly for the second straight year. The decline rate by carrier varied from 11.2 to 41.8 percent, affected by factors such as age distribution, market, underwriting requirements, and underwriting standards. Our placed percentages reflect the insurers’ perspective. A higher percentage of applicants secures coverage because applicants denied by one carrier may be issued either stand-alone or combo coverage by another carrier or may receive coverage with the same insurer after a deferral period.

Underwriting Tools
Eight insurers contributed data to Table 23, which divides the number of uses of each underwriting tool by the number of applications processed. For example, the number of medical records was 86 percent of the number of applications. That does not mean that 86 percent of the applications involved medical records, because some applications resulted in more than one set of medical records being requested.

Insurers are trying to speed underwriting to increase placement rates. In the worksite market, insurers are less likely to use some of these tools.

Year-to-year changes in distribution of sales among insurers significantly impact results. Lower maximum ages result in fewer face-to-face exams. Insurers might underreport the use of an underwriting tool because they may lack a good source for that statistic. For example, the cognitive phone calls are understated below because one insurer could not split them apart from their other phone interviews (PHI) and reported them all as PHI with no or minimal cognitive testing.

Underwriting Time
Table 24 shows the average processing from receipt of application to mailing the policy (37.0 days) was 2.4 days faster than in 2018, which in turn had been 3.4 days faster than in 2017. About 38 percent of the reduction was due to the change in survey participants.

The reduction came at both ends of the spectrum, a much higher percentage processed within 29 days and many fewer taking 45 or more days. All but one insurer reported an increase.

Rating Classification
Table 25 shows that a lower percentage of policies was issued in the most favorable rating classification, but for participants which contributed data both years, the percentage increased 0.4 percent despite a higher average issue age. Two insurers placed more than 80 percent of their applicants in the best underwriting class. Only 8.1 percent of the applicants were placed beyond the second-best classification. In 2016 and prior years, the “best” percentage was lower partly because we received more data from insurers writing worksite business, which does not offer preferred health discounts and because one insurer eliminated its preferred health discount (hence “standard” ratings became its “best”).

Underwriting placements have become more favorable. To gauge whether insurers are declining applicants rather than placing them in a less-attractive rating, we added a line in Table 25 that shows the percentage of decisions which were either declined or placed in the third or less-attractive classification. This approach confirms the more favorable decisions in 2019.

Tables 26 and 27 show by issue age range the percentage of policies issued in the most favorable category in 2019 and the percentage of decisions that were declines. One participant was not able to provide the by-issue-age data, so Tables 26 and 27 do not exactly match Table 25.

Please click on the links below to find the following additional information:

• Product Exhibit shows, for eight insurers: Financial ratings, LTCI sales and inforce, and product details. Please note that, during the COVID-19 pandemic, some insurers have temporarily discontinued sales that would require face-to-face interviews. Our Exhibit ignores such temporary restrictions.
• Product Details, a row-by-row definition of the product exhibit entries, with some commentary.
• Premium Exhibit, which shows lifetime annual premiums for each insurer’s most common underwriting class, for issue ages 40, 50, 60, and 70 for single females, single males, and heterosexual couples (assuming both buy at the same age), based on $100 per day (or closest equivalent weekly or monthly) benefit, 90-day facility and most common home care elimination period (other aspects vary), three-year and five-year benefit periods or $100,000 and $200,000 maximum lifetime buckets, with and without Shared Care and with flat benefits or automatic three or five percent annual compound benefit increases for life. The exhibit includes facility-only policies, as well as comprehensive policies. Worksite products do not reflect any worksite-specific discount, though some carriers offer this.
• Premium Adjustments (from our Premium Exhibit prices) by underwriting class for each participant.
• Distribution by underwriting class for each participant.
• State-by-state results: percentage of sales by state, average premium by state and percentage of policies qualifying for Partnership by state.

CLOSING
We thank insurance company staff for submitting the data and responding to questions promptly. We also thank Nicole Gaspar, Alex Geanous, and Anders Hendrickson of Milliman for managing the data expertly.

We reviewed data for reasonableness and insurers reviewed their product exhibit displays. Nonetheless, we cannot assure that all data is accurate.

If you have suggestions for improving this survey (including new entrants in the market), please contact one of the authors.

Reference:

1. Society of Actuaries (November 2016). Long-Term Care Insurance: The SOA Pricing Project. Retrieved May 16, 2019, from https://www.soa.org/globalassets/assets/files/static-pages/sections/long-term-care/ltc-pricing-project.pdf.

The 2019 Milliman Long Term Care Insurance Survey is the 21st consecutive annual review of stand-alone long-term care insurance (LTCI) published by Broker World magazine. It analyzes the marketplace, reports sales distributions, and describes available products including group insurance.

More analysis of worksite sales will appear in the August issue of Broker World magazine.

Unless otherwise indicated, references are solely to U.S. stand-alone LTCI sales, excluding exercised future purchase options (FPOs) or other changes to existing coverage. “Stand-alone” refers to LTCI policies that do not include death benefits (other than returning premiums upon death or waiving a surviving spouse’s premiums) or annuity or disability income benefits. Where referenced, “combo” policies provide LTCI combined with life insurance or annuity coverage.

Highlights from This Year’s Survey

Participants
Eleven carriers participated broadly in this survey. Four others provided sales information so we could report more accurate aggregate industry individual and multi-life sales. From these submissions, we estimated total industry production.

We estimate our statistical distributions reflect up to 90 percent of total industry sales and about 65 percent of worksite sales.

State Farm discontinued stand-alone LTCI sales in May 2018, hence is no longer included in the Product Exhibit.

Although not displaying products, Northwestern LTC and State Farm provided background statistical information. Auto-Owners, LifeSecure, Transamerica and United Security Life contributed total and worksite sales (new premium and lives insured) but did not provide broad statistical information.

Sales Summary

• The 15 carriers reported sales of 56,288 policies and certificates (“policies” henceforth) with new annualized premium of $171,537,644 (including exercised FPOs) in 2018, compared to 2017 restated sales of 64,800 policies ($181,506,770 of new annualized premium), a 13.1 percent drop in the number of policies and a 5.5 percent drop in the amount of new annualized premium. As noted in the Market Perspective section, sales of policies combining LTCI with other risks continue to increase.
• Five of the top 10 insurers sold more new premium than in 2017, but only three sold more new policies. This difference emerges because elected FPOs add on-going premium but not new policies.
• With FPO elections included in new premium, Northwestern garnered the number one spot in new sales. Mutual of Omaha was a strong second and had a large lead in annualized premium from new policies sold. Together, they combined for 57 percent of new premium including FPOs and 52 percent of new premium excluding FPOs.
• For the fourth straight year (and fourth time ever), our participants’ number of inforce policies dropped, this time by 1.1 percent, after 5.1 percent (2017), 0.3 percent (2016) and 0.2 percent (2015) drops previously.
• Nonetheless, year-end inforce premium per policy continues to increase (3.0 percent in 2018) to $2,169. Inforce premium increases from sales, price increases, and benefit increases (including FPOs), and reduces from lapses, reductions in coverage, deaths, and shifts to paid-up status for various reasons.
• Participants’ individual claims rose 5.9 percent. Overall, the stand-alone LTCI industry incurred $11.0 billion in claims in 2017 based on companies’ statutory annual filings, raising total incurred claims from 1991 through 2017 to $129.9 billion. (Note: 2017 was the most recent year available when this article was written.) Most of these claims were incurred by insurers that no longer sell LTCI. The reported 2017 incurred claims is similar to the $11.1 billion of incurred claims reported in 2016. Combo LTC claims are in their infancy and amounted to $5.9 million. The claim figures are even more startling considering that only a small percentage of the 7 million covered individuals were on claim at the end of 2017.
• The average processing time in the industry was eight percent faster in 2018 than in 2017. Nonetheless, active policies resulted from only 58.8 percent of applications, even lower than 2017’s record low of 59.0 percent. Financial advisors often are reluctant to risk a bad experience by recommending that clients apply for LTCI. As noted in the Market Perspective section, the industry may be able to improve placement rates in a variety of ways.

About the Survey
This article is arranged in the following sections:

• Highlights provides a high-level view of results.
• Market Perspective provides insights into the LTCI market.
• Claims presents industry-level claims data.
• Sales Statistical Analysis presents industry-level sales distributions reflecting data from 11 insurers
• Partnership Programs discusses the impact of the state partnerships for LTCI.
• Product Exhibit (click here for PDF) shows, for nine insurers: financial ratings, LTCI sales and inforce, and product details.

Click here to view the following additional information available only online.

• Product Details, a row-by-row definition of the product exhibit entries, with a little commentary.
• Premium Exhibit shows lifetime annual premiums for each insurer’s most common underwriting class, for issue ages 40, 50, 60, and 70 for single females, single males, and heterosexual couples (assuming both buy at the same age), based on $100 per day (or closest equivalent weekly or monthly) benefit, 90-day facility and most common home care elimination period, three-year and five-year benefit periods or $100,000 and $200,000 maximum lifetime buckets, with and without Shared Care and with flat benefits or automatic three percent or five percent annual compound benefit increases for life. Worksite premiums do not reflect any worksite-specific discount, though some carriers offer this.
• Premium Adjustments (from our published prices) by underwriting class for each participant.
• Distribution by underwriting class for each participant.
• State-by-state results: percentage of sales by state, average premium by state and percentage of policies qualifying for Partnership by state.

Market Perspective (more detail in subsequent parts of the article)

• The stability of current prices bears no resemblance to the past instability because today’s prices reflect much more conservative assumptions based on far more credible data and low investment yields. Unfortunately, many financial advisors presume that new policies will face steep price increases. It is likely to take a long time before the market becomes comfortable that prices are stable.
• Combo products have increased market share because: i) their other benefits mean that regardless of whether the policyholder has a long term care claim, they will receive benefits, ii) they often have guaranteed premiums and benefits, and iii) they now offer alternatives besides single premium.

According to LIMRA, combo life policies (LTCI combined with life insurance) represented 16 percent of new 2017 annualized life insurance premium (25 percent and $4.1 billion, if you include 100 percent of single premium sales). Fourteen percent of the 260,000 combo policies included LTCI benefits after the death benefit had been fully advanced (“extension of benefits; EOB”). Of the 86 percent in which LTCI benefits could not exceed the death benefit, nearly six in 10 (57 percent) used “chronically ill” provision that is not allowed to be called “long term care insurance”, rather than a §7702 LTCI provision.

Looking at the total LTCI market, stand-alone policies accounted for 20.0 percent of the 2017 policies sold, policies with extensions of benefits (EOB) accounted for 11.2 percent and policies with accelerated death benefits but no EOB accounted for 68.8 percent.

• As anticipated in last year’s report, a new entrant offers unisex LTCI pricing for the worksite market, another carrier re-entered the worksite market and a third carrier is test-marketing a worksite product for possible release. In addition, more combo products are being proposed in the worksite market. For businesses with employees residing in multiple jurisdictions, consistent product availability can be a challenge. Worksite LTCI is more attractive to employers when packaged with solutions for employees who are caregivers for their elders.
• As noted earlier, fewer than 59 percent of applications have resulted in issued policies in the past two years. The low placement rate makes financial advisors hesitant to recommend that clients consider LTCI. The industry may be able to improve placement rates in a variety of ways.
  • E-applications make the process faster, secure better health information (some applications may then not be submitted) and avoid long delays on apps that are “not in good order” when submitted. In our 2018 survey article, we reported that eApps were 71 percent more likely to be in good order and were less likely to be declined. It would be helpful if brokerage general agents could populate an eApp with quote information, then forward that partial eApp to the broker for completion.
  • Better pre-qualification of prospects’ health will guide applications to an insurer most likely to accept the applicant. As noted above, eApps help. If advisors are reluctant to discuss health issues with their clients, some general agencies interview the client on behalf of the advisor or provide a link to a website where the client can answer health questions. Data quantifying the positive impact of thorough pre-qualification would be helpful.
  • Higher placement rates result when cash is required with the application (CWA). This year, we asked insurers for data regarding placement (excluding worksite business) based on whether cash was required with the application. Carriers who accept applications with or without cash reported a combined placement rate of 58.1 percent with cash and 54.0 percent without cash. Insurers which reported only CWA data placed 60.4 percent of their cases. Insurers which reported only data without CWA placed 55.0 percent.
  • Faster processing may help. However, as noted earlier, the average processing time was 10 percent faster in 2018 than in 2017, yet the placement rate still dropped 0.2 percent. Maybe it would have dropped more without the faster processing.
  • Better messaging regarding the value of LTCI and about the value of buying now (rather than in the future) would improve the placement rate (as well as increase the number of applications).
  • Insurers may be able to educate their distribution system more effectively, such as with drill-down questions in on-line underwriting guides.

• Three participants have never increased premiums on policies issued under “rate stabilization” laws. Three insurers reported that their highest cumulative increase on such policies has been about 40 percent and two had increased prices 100 percent or more. Two carriers did not indicate their maximum increase, one of which we believe was in the lower end and one in the higher end.

All the insurers allow clients to reduce the maximum daily/monthly benefit (typically not below the original minimum) with a proportionate reduction in future premium. All allow clients to move to a shorter originally-available benefit period with premiums the same as they would currently be had they purchased that benefit period originally. Eighty percent of the insurers allow clients to select a longer elimination period, with premiums the same as they would currently be had they purchased that longer elimination period initially. Some insurers may not offer this option because they do not have longer elimination periods available.

If a client drops a compound benefit increase rider, most insurers freeze the current benefit and charge the premium that the client would be paying today had the client purchased the current amount originally with no compound benefit increases. One insurer takes a more consumerist approach, freezing the benefit and reducing the current premium by the cost of the benefit increase rider. Some policyholders may face reversion to the original maximum daily or monthly benefit.

One insurer explained that the above approaches for dropping a compound benefit increase rider result in higher future premiums as a result of sacrificing future benefit increases if the maximum daily/monthly benefit has increased beyond the additional percentage cost for the compound benefit increase rider. Therefore, that insurer suggests that clients reduce the daily/monthly maximum benefit and retain compounding.

• Our projected amount of issued protection on new policies increased significantly in 2018 because an insurer with negative-election FPOs (negative-election FPOs activate automatically unless the client rejects them, as opposed to positive-election FPOs which activate only if the client makes a request) submitted FPO election rate data this year but not last year. For the average 57-year-old purchaser in 2018, we project a maximum benefit in 2041 of $313/day, equivalent to an average 3.0 percent compounded benefit increase. Had he/she purchased last year’s average policy at age 56, he/she would have had $279/day by age 80, equivalent to 2.3 percent compounding. Without the change in insurers providing FPO data, the age 80 maximum daily benefit would have dropped from $279 to $276. Purchasers may be disappointed if the purchasing power of their LTCI policies deteriorates over time.
• Claimants rarely challenge insurer claim adjudications. Since 2009 (varies by jurisdiction), if an insurer concludes that a claimant is not chronically ill, the insurer must inform the claimant of his/her right to appeal the decision to independent third-party review (IR). The IR determination is binding on insurers. As shown in our Product Exhibit, most participants have extended IR beyond statutory requirements, most commonly to policies issued prior to the effective date of IR. At least four participating insurers report never having a request for IR. Four other insurers have reported a total of 72 IR requests resulting in the insurers’ denials being upheld more than 90 percent of the time. This data is consistent with the experience of LTCI Independent Eligibility Review Specialists, LLC. Steve LaPierre, president of that firm, indicated that they have performed approximately 125 IRs and have upheld the insurer more than 90 percent of the time. The existence of IR, the insurers’ voluntary expansion of IR and the insurer success rate when appeals occur help justify confidence in claim decisions.
• Only four participants offer coverage in all U.S. jurisdictions and no worksite insurer does so. Insurers are reluctant to sell in jurisdictions which are slow to approve new products, restrict rate increases, or have unfavorable legislation or regulations.
• Seven of our 11 participants use reinsurers and seven use third party administrators (TPAs). The reinsurers are Reinsurance Group of America, LifeCare, Manufacturers, Swiss Re and Union Fidelity. The TPAs are Long-Term Care Group, Life Plans, LifeCare Assurance, and CHCS. Other reinsurers and TPAs support insurers not in our survey. In some cases, affiliated companies provide reinsurance or guarantees.
• A significantly-changed LTCI Shopper’s Guide was recently adopted by the NAIC. We’re interested in readers’ reviews of the new Guide and comments as to how much it is used to educate consumers. (Please email comments to [email protected].)

Claims

• Ten participants reported 2018 claims. The consistency of the data is improving, but some companies were not able to respond to some questions or could not respond in a way that justified including their data for some questions.
• For the ten insurers which reported individual claims for both 2018 and 2017, claim dollars rose 5.9 percent, despite a 1.1 percent decrease in inforce policies.
• The LTCI industry has had a much bigger impact than indicated above, because a lot of claims are paid by insurers that no longer sell LTCI.

LTCI claims paid by insurers no longer selling LTCI likely differ significantly from data reported below as their claimants are more likely to have facility-only coverage, be older, etc.

Table 1 shows the total dollar and number of individual LTCI claims paid by those carriers which provided information to use.

Table 2 shows the distribution of those claims by venue. In the distribution based on number of claims, a person who received care in more than one venue is counted once for each venue, but not double-counted in the total line. Individual claims in general continue to shift away from nursing homes. We expect on-going shift away from nursing homes due to consumer preferences and more claims coming from comprehensive policies.

Six of eight carriers that submitted their number of open claims at year-end reported a pending number of claims between 57 percent and 84 percent of the number of claims paid during the year.

Table 3 shows average size individual claims since inception, all of which rose compared to 2017. Because claimants can submit claims from more than one type of venue, the average total claim generally exceeds the average claim paid for any particular venue. Nonetheless, individual ALF claims are consistently high, probably because:

a) ALF claims appear to last long compared with other venues.

b) Nursing home costs are most likely to exceed the policy daily/monthly maximum. Hence the maximum daily benefit negates part of the additional daily cost of nursing homes.

c) People who maximize the use of their maximum monthly benefits can spend as much on an ALF as on a nursing home.

d) Although some surveys report that ALFs cost about half as much as nursing homes on average, ALFs often charge more for a memory unit or more substantial care for levels of assistance that align more closely with nursing home care. And upscale ALFs seem to cost a higher percentage of upscale nursing home costs than the average ALF/nursing home ratio.

Some people may have expected ALF claims to be less expensive than nursing home claims because ALFs cost less per month. But that has not been the case. Some observers applaud insurers which have extended ALF coverage to policies which originally did not include ALF coverage, even though such action has contributed to rate increases on in force policies.

The following factors cause our average claim sizes to be understated.

• Roughly 1/8 of the inception-to-date individual claims are still open. Our data does not include reserve estimates for future payments on open claims.
• People who recover, then claim again, are counted as multiple insureds rather than adding their various claims together.

The range of average claim results by insurer (see Table 3) is startling. What can contribute to such differences?

• Different markets (by affluence; worksite vs. individual; by geography; etc.).
• Different lengths of time in the business. A newer issuer may have a higher percentage of open claims bringing down the average.
• Percentage of policies sold to women or with compound benefit increases, monthly home care determination, or 50 percent home care benefits or reduced benefits for ALFs.
• Distribution by elimination period. Zero-day EP policies might result in a lot of small claims. Ironically 180-day or longer EPs could also result in relatively small claims.
• Perhaps erroneous reporting. In some of our calculations we have eliminated some submitted data because it seemed very unlikely to be accurate.

Average claim data understates the value of buying LTCI because the many small claims drive down the average claim. LTCI can provide significant financial return for people who need care one year or longer. The primary purpose of insurance is to protect against adverse results, so the amount of protection, as well as average claim, is important.

Six insurers were able to provide data regarding their current monthly exposure. The average current monthly maximum nursing home benefit per inforce claimant ranged from $4,575 to $8,182.

The six insurers’ total monthly exposure (including non-claimants) exceeds $5.5 billion, more than thirty times their corresponding monthly premium income. Recognizing that claims can continue up to the full benefit period, these insurers’ potential claim exposure is about 120 times their monthly premium income.

Nursing home (NH) claims are more likely to use the policy’s maximum daily/monthly benefit than ALF claims, because ALF costs are generally lower and because policies sometimes have lower maximums for ALFs. ALF claims correspondingly are more likely to use the policy maximum than are adult day care and home care claims.

Statistical Analysis
Eleven insurers contributed significant background data, but some were unable to contribute data in some areas. Four other insurers (Auto-Owners, LifeSecure, Transamerica and United Security) contributed their number of policies sold and new annualized premium, distinguishing worksite from other sales.
Sales characteristics vary significantly among insurers. Year-to-year variations in policy feature distributions may reflect changes in participants, participant practices and designs, participant or worksite market shares and industry trends.

Market Share
Table 4 lists the top 10 participants in 2018 new premium, among those still offering LTCI. Because we include FPO elections, Northwestern surpassed Mutual of Omaha. Without FPOs, Mutual of Omaha would have been number one. Together, they produced 57 percent of annualized new premium in 2018 and 52 percent of annualized premium on new policies. They are followed by five insurers with four percent to 10 percent market share each.

Worksite Market Share
Worksite business produced 15.0 percent of new insureds (see Table 5), but only 9.1 percent of new premium (including FPOs) because of its younger issue age distribution and less robust coverage. Worksite sales consist of three different markets:

• Voluntary group coverage generally is less robust than individual coverage.
• Core/Buy-Up programs have particularly young age distributions and modest coverage because a lot of people do not buy-up and are least likely to insure spouses.
• Executive carve-out programs generally provide the most robust coverage. One- or two-couple executive carve-out sales may not qualify for a multi-life discount with some insurers, hence may not be labeled as worksite sales in submissions to our survey.

The amount of worksite sales reported and its distribution among the three sub-markets significantly impact product feature sales distributions. Although Table 5 reflects the full market, this year’s policy feature distributions underweight the voluntary and core/buy-up markets because carriers in those markets shared less policy feature distribution data. More information about worksite sales will appear in the August issue of Broker World magazine.

Affinity Market Share
The affinity sales percentage is based on the participants who provide significant statistical data. Their reported affinity sales produced 7.8 percent of their new insureds (see Table 6), but only 6.2 percent of premium. Only about 20 percent of the lower affinity average premium is attributable to the affinity discount. The balance may be due to younger issue age or less robust coverage.

Characteristics of Policies Sold
Average Premium
As noted last year, our inclusion of FPOs as new business premium overstated the average premium per new insured and buying unit (Table 7) increasingly over time. More precise queries this year allow us to quantify the impact. The average premium per new life ($2,544) is 18 percent less than we would have quoted including FPOs in the numerator. Three insurers reported average premiums for new insureds below $1,700, while five insurers were over $2,800. The average premium per new buying unit (counts a couple only once) was $3,598.

The lowest average new premium (including FPOs) was in Puerto Rico ($1,960), followed by Kansas ($2,448), while the highest was in New York ($4,243), followed by Connecticut ($3,886).

Due to rate increases, FPO elections and termination of older policies, the average inforce premium jumped to $2,168, 3.0 percent more than our restated 2017 figure.

Issue Age
Table 8 summarizes the distribution of sales by issue age band based on insured count. The average issue age was 56.6. The higher average issue ages of the past two years are partly caused by not having issue age distribution for some worksite business. Furthermore, two participants have a minimum issue age of 40, one won’t issue below 30, and two won’t issue below 25.

Benefit Period
Table 9 summarizes the distribution of sales by benefit period. The average notional benefit period slightly increased from 3.73 to 3.74. Because of Shared Care benefits, total coverage was higher than the 3.74 average suggests. For the first time, a single benefit period (3-year) accounted for half the sales.

Monthly Benefit
Table 10 shows that monthly determination applied to 77.8 percent of 2018 policies. With monthly determination, low-expense days leave more benefits to cover high-expense days. When it is offered as an option, we conclude that more than 50 percent of buyers opt for monthly determination. Measuring election rates for optional monthly determination can be challenging due to influences such as insurers switching products during the course of a year and state variations.

Table 11 summarizes the distribution of sales by maximum monthly benefit at issue. Sales were more clustered between $4,500 and $5,999/month than in any year since 2011. The average initial maximum monthly benefit was $4,763, almost exactly the average of 2015-2017.

Benefit Increase Features
Table 12 summarizes the distribution of sales by benefit increase feature. “Other compound” has grown a lot in the past two years; most involve three percent compounding.

The increase in “FPO: Indexed” and the decrease in policies having “no benefit increases” were caused by a reclassification of an insurer’s sales. The insurer guarantees to offer FPOs of indeterminate amount, but the insurer bases its decision on interest rates.

Five percent compounded for life, which represented 56 percent of sales in 2003 and more than 47.5 percent of sales each year from 2006 to 2008, now accounts for only two percent of sales. Simple five percent increases for life were 19 percent of 2003 sales but are now only 0.4 percent of sales.

“Indexed Level Premium” policies are priced to have a level premium, but the benefit increase is tied to an index such as the consumer price index (CPI).

We project the age 80 maximum daily benefit by increasing the average daily benefit purchased from the average issue age to age 80, according to the distribution of benefit increase features, using current future purchase option (FPO) election rates, a five percent/year offer for fixed FPOs and assuming a long-term three percent CPI. The maximum benefit at age 80 (in 2041) for our 2018 average 57-year-old purchaser projects to $313/day (equivalent to 3.0 percent compounding). Had our average buyer bought an average 2017 policy a year ago at age 56, her/his age 80 benefit would be $279/day (equivalent to 2.3 percent compounding). This healthy increase in age 80 maximum benefit (from $279/day to $313/day) is attributable to the inclusion of an additional carrier in the data (see Table 14 discussion). Without that insurer, the projected age 80 maximum daily benefit would have dropped from $279 to $276. Most policyholders seem likely to experience eroding purchasing power over time if cost of care trends exceed three percent.

Eight insurers reported new premium from FPO elections; the percentage of their new premium that came from FPOs ranged from two percent to 43 percent.

Seven insurers provided the number of available FPOs in 2018 and the number exercised. Six of those insurers provided similar data last year; overall, they had a small increase in election rate from 34.7 percent to 36.8 percent. By insurer, election rates varied from 13 percent to 91 percent. The insurers with the three lowest FPO election rates averaged a 25.8 percent election rate; they probably use a “positive election” approach. The other four insurers averaged 90.0 percent election; they probably use a “negative election” approach; i.e., the increase applies unless specifically rejected. The carrier which did not contribute such data last year caused the huge increase in election rate (Table 14); it probably uses a negative election rate.

Elimination Period
Table 15 summarizes the distribution of sales by facility elimination period. Ninety percent of buyers opt for 90-day elimination periods, however, two carriers report fewer than 60 percent of their policies have 90-day EPs (one has many shorter EPs and the other has many longer EPs). The percentage of purchasers buying 180-day or longer EPs hit a record (6.7 percent vs. 6.1 percent in 2012).

Table 16 shows that the percentage of policies with zero-day home care elimination period (but a longer facility elimination period), with adjustments for two carriers which reported only sales, but we know has either type of provision. With most insurers, fewer than 25 percent of buyers purchased a zero-day home care elimination period, but one insurer had nearly a 50 percent election rate.

The percentage of policies with a calendar-day elimination period (EP) definition (33.7 percent) also dropped back toward 2017 levels. For insurers which offer calendar-day EP, 79.4 percent of policies had the feature; in some cases, it was automatic. It is important to understand that most calendar-day EP provisions do not start counting until a paid-service day has occurred.

Sales to Couples and Gender Distribution
Table 17 summarizes the distribution of sales by gender and single/couple status.

The percentage of purchasers who consisted of couples who both buy (55.5 percent) and the percentage of all buyers who were female (67.7 percent) were the lowest percentages since at least 2010.

The 77.8 percent of accepted applicants who purchased coverage when their partners were declined was the highest over that time period. Three factors contributed approximately equally to the increase: General improvement; change of practice to allow a lower marital discount instead of no discount when one spouse is declined; and a change in participants reporting this data.

Fifty-five percent (55.1 percent) of all buyers were female, the lowest percentage since 2012. Insurers’ sales ranged from 47.2 percent female to 61.4 percent female, varying based on market (older age; fraternal, worksite vs. individual) and whether unisex rates were offered. Likewise, insurers ranged from 54.4 percent of single buyers being female to 79.8 percent.

Among couples, 51.0 percent are females, with insurers ranging from 47.9 percent (no other insurer showed less than 49.9 percent) to 52.1 percent. When only one of a couple buys, 56.4 percent are females, probably because the male partner is likely to be older and less likely to be insurable.

Shared Care and Other Couples’ Features
Table 18 summarizes sales of Shared Care and other couples’ features.

• Shared care—allows one spouse/partner to use the other’s available benefits if their own coverage has been depleted or offers a third independent pool that the couple can share.
• Survivorship—waives a survivor’s premium after the first death if specified conditions are met.
• Joint waiver of premium (WP)—both insureds’ premiums are waived if either qualifies for benefits.

Changes in distribution by carrier can greatly impact year-to-year comparisons in Table 18, because some insurers embed survivorship or joint waiver automatically (sometimes only in some circumstances) while others offer it for an extra premium or do not offer the feature. These percentages are lower than in the past, because one carrier reported the number of sales to couples last year, but not this year.

In the top half of Table 18, the percentages are based on the number of policies sold to couples who both buy. The bottom half of Table 18 shows the (higher) percentage that results from dividing the number of buyers by sales of insurers that offer the feature. Two insurers sold Joint WP to 56 to 60 percent of their couples and Survivorship to 10 to 12 percent of their couples.

Table 19 provides additional breakdown on the characteristics of Shared Care sales. As shown on the right-hand side of Table 19, three-year and four-year benefit period policies are most likely (nearly 30 percent) to add Shared Care. Partly because three-year benefit periods comprise 50 percent of sales, most policies with Shared Care are three-year benefit period policies (61.1 percent, as shown on the left side of Table 20).

Above, we stated that Shared Care is selected by 33.4 percent of couples who both buy limited benefit period policies. However, Table 19 shows Shared Care comprised no more than 29.8 percent of any benefit period (it was elected by 25.4 percent of purchasers overall). Table 19 has lower percentages because Table 18 denominators are limited to people who buy with their spouse/partner whereas Table 19 denominators include all buyers. Shared care is more concentrated in two- to four-year benefits periods (89.4 percent of shared sales) than are all sales (72.6 percent). Couples seem more likely to buy short benefit periods, perhaps because couples plan to help provide care to each other, because Shared Care makes shorter benefit periods more acceptable and because single buyers are more likely to be female, hence opt for a longer benefit period. On the other hand, we also see a relatively high percentage of Shared Care on longer benefit periods; these people are probably trying to cover catastrophic risk and might prefer an endless benefit period. Some insurers were more likely to sell Shared Care on their short benefit-period policies, while others were more likely to sell Shared Care on their long benefit-period policies.

Existence and Type of Home Care Coverage
One participant reported home-care-only policies, which accounted for 0.7 percent of industry sales. Four participants reported sales of facility-only policies, which accounted for 0.9 percent of total sales. Ninety-five percent (95.1 percent) of the comprehensive policies included home care benefits at least equal to the facility benefit. These percentages are all lower than in 2017.

Partial cash alternative (and similar) features (which allow claimants, in lieu of any other benefit that month, to use between 30 percent and 40 percent of their benefits for whatever purpose they wish) were included in 41.3 percent of sales (lower than last year’s 44.1 percent).

Other Characteristics
As shown in Table 20, return of premium (ROP) features were included in 12.7 percent of all policies. ROP returns some or all premiums (usually reduced by paid LTCI benefits) when a policyholder dies. Approximately 79 percent of policies with ROP arise from ROP features embedded automatically in the product, compared to 93 percent in 2016. Embedded features are designed to raise premiums minimally, typically decreasing the ROP benefit to $0 by age 75.

Twelve percent (12.4 percent) of policies with limited benefit periods included a restoration of benefits (ROB) provision, which typically restores used benefits when the insured does not need services for at least six months. Approximately 87 percent of policies with ROB arose from ROB features automatically embedded, compared with 79 percent in 2016.

Shortened benefit period (SBP) nonforfeiture option was included in 1.8 percent of policies. Although every insurer is obligated to offer SBP, some carriers did not report any SBP sales. It seems appropriate to remove their sales from the denominator when determining the percentage of purchasers who selected SBP. On that basis, the percentage was 2.9 percent. Only one insurer reported a percentage above six percent (13.3 percent). SBP makes limited future LTCI benefits available to people who stop paying premiums after three or more years.

Only one insurer issued non-tax-qualified (NTQ) policies, which accounted for 0.1 percent of industry sales.

“Captive” (dedicated to one insurer) agents produced 55.8 percent of the policies. At one time, “captive” agents who sold LTCI tended to specialize in LTCI. Now many are agents of mutual companies. The 1.5 percent “other” sales did not involve a broker or agent.

Sales distribution by jurisdiction is posted on the Broker World website.

Limited Pay and Paid-Up Policies
In 2018, only two insurers sold policies that become paid-up, accounting for fewer than 0.5 percent of sales.

Because today’s prices are more stable, premium increases are less likely. One of the key reasons for buying 10-year-pay (avoidance of rate increases after the tenth year) is greatly reduced, while the cost of 10-year-pay has increased, making it less attractive than in the past. Nonetheless, limited-pay and single-pay policies are attractive to minimize post-retirement outflow and to accommodate §1035 exchanges.

Partnership Program Explanations
When someone applies to Medicaid for long term care services, states with Partnership programs disregard assets up to the amount of benefits received from a Partnership-qualified policy. Partnership sales were reported in 44 jurisdictions in 2018, all but Alaska, District of Columbia, Hawaii, Massachusetts, Mississippi, Utah, and Vermont, where Partnership programs do not exist. Massachusetts has a somewhat similar program (MassHealth).

The Partnership rules in California, Connecticut, Indiana and New York (“original” Partnership states) are significantly different than in other Partnership jurisdictions (“DRA” jurisdictions). The “original” states legislated variations of the Robert Woods Johnson Partnership proposal, whereas the “DRA” jurisdictions use more consistent rules based on the Deficit Reduction Act of 2005. For example, the “original” states require a separate Partnership policy form, generally still have more stringent benefit increase requirements and assess a fee for insurers to participate (none of which applies in DRA states). As a result, only two to four insurers sell Partnership policies in CA (2), CT (3), IN (4) and NY (2).

National Reciprocity Compact (NRC) requires member states to recognize Medicaid Asset Disregard earned in any other member state. States creating Partnerships under the Deficit Reduction Act of 2005 were automatically enrolled in the NRC but had the right to secede. The four original Partnership states (California, Connecticut, Indiana and New York) had the right to opt in. California is now the only jurisdiction with a Partnership program that is not a member of the NRC. However, New Hampshire (NH) created a unique regulation in 2018, limiting asset disregard related to policies sold in other jurisdictions to coverage issued on or after April 1, 2007, and only if the policy form was approved by the NH Insurance Department. We’ve asked NH about this regulation in 2018 and 2019, but have not received an explanation yet, so we’ll venture some thoughts below.

NH natives did not have an opportunity to purchase a Partnership policy until April 1, 2007. NH’s April 1, 2007, deviation from NRC standards seems intended to avoid extending asset disregard to an insured who moved in from another state if a NH resident who purchased on the same date can’t benefit from asset disregard.

The requirement that the policy form need have been approved by the NH Insurance Department is puzzling. Why should a NH citizen be harmed because they bought a policy as a resident of another jurisdiction? The impact may be limited because NH has been a member of the Interstate Insurance Product Regulation Commission since June 1, 2007. The IIPRC (“Compact”) approves LTCI policy forms for about 40 states. Policyholders with IIPRC-approved Partnership coverage issued since April 1, 2007, might still qualify for asset disregard if they move to NH.

It seems unlikely that individuals considering relocating to NH and their financial advisors will understand the potential Partnership ramifications. In the unlikely event that they are aware of NH’s unique law, it seems unlikely that they would know whether their policy form was approved by NH.

Partnership Program Sales
Insurers sometimes delay certifying policy forms as “Partnership” because of other priorities (e.g., needing time to comply with state-specific requirements to notify existing policyholders or offer an exchange). Such delay is not harmful, as certification is retroactive to policies already issued on that policy form if the policies have the required characteristics. For this reason and the “original” Partnership issues mentioned above, none of our participants sold Partnership policies in more than 40 jurisdictions in 2018. Five sold Partnership in 35-40 jurisdictions, three sold Partnership in 26-32 jurisdictions, one sold Partnership in eight jurisdictions, one in one jurisdiction (it sells in only one jurisdiction) and the other has never certified Partnership conformance and apparently does not intend to do so.

In the DRA states, 52.5 percent of policies qualified for Partnership status, whereas in the original states only 1.0 percent qualified. In Minnesota, more than 82 percent of the policies sold qualified as Partnership and Wisconsin and Wyoming both had 75 percent of their policies qualify.

Generally speaking, experts have expected Partnership policies to have higher average premiums because of the benefit increase requirements for Partnership policies. Only 26 of the 44 Partnership states demonstrated that pattern, with the other 18 having higher average premiums on non-Partnership policies.

In past survey articles, we noted that Partnership programs could be more successful if states broadened the eligibility requirements. Now, approximately 60 percent of Partnership states allow one percent compounding to qualify for Partnership, which can help low-budget buyers qualify for Partnership and also enables worksite core programs to be Partnership-qualified. A higher percentage of policies should qualify for Partnership in the future if insurers and advisors leverage these opportunities. Currently only three insurers offer one percent compounding.

Partnership programs could be more successful if:

• Advisors offer small maximum monthly benefits more frequently to the middle class. For example, a $1,500 initial maximum monthly benefit covers about four hours of home care every two days and, with compound benefit increases, may maintain buying power. Many middle-class citizens would like LTCI to help them stay at home while not “burning out” family caregivers and could be motivated further by Partnership asset disregard. (This approach does not work where a Partnership initial maximum monthly benefit must be at least $8608 (CT) and $9114 (NY). When policies reflecting CA SB 1248 become available in California, California’s minimum size policy will drop to $3000/month.)
• Middle-class clients were better educated about the importance of benefit increases to maintain LTCI purchasing power and to qualify for Partnership asset disregard.
• The four original Partnership steps migrate to DRA rules.
• More jurisdictions adopt Partnership programs.
• Programs that privately finance educational LTCI direct mail from public agencies were adopted more broadly.
• Financial advisors were to press reluctant insurers to certify their products.
• More financial advisors were certified. Some people argue that certification requirements should be loosened. At a minimum, the renewal certification process could be improved.
• More insurers offer one percent compounding.
• Combo products became Partnership-qualified.

Underwriting Data
Case Disposition
Ten insurers contributed application case disposition data to Table 21. In 2018, 58.8 percent of applications were placed, including those that were modified, a new low slightly below 2017’s previous record low of 59.0 percent.

One insurer reported a 77.1 percent placement rate; the second highest being 56.0 percent. The lowest placement rate was 34.0 percent. Low placement rates increase insurers’ cost per placed policy. More importantly, low placement rates can discourage advisors from discussing LTCI with clients. In addition to wasting time and effort encouraging clients to apply for LTCI, advisors fear disappointment for clients who are rejected.

Although the placement rate dipped slightly, the decline rate also dipped slightly to 25.1 percent of applications (28.2 percent of insurer decisions). Unfortunately, both the suspended/withdrawn and the Free Look refusals/not takens increased. The decline rate by carrier varied from 12.0 percent to 39.3 percent, affected by factors such as age distribution, market, underwriting requirements, underwriting standards.

Perhaps the best way to improve placement rates is to do a better job of pre-qualifying clients’ health profiles prior to submitting applications. General agencies and insurers are promoting approaches which make it easier for advisors to get detailed health information or have a third party ask health questions. For example, eApps should develop better health information and result in speedier processing which may help placement rates.

Our placed percentages reflect the insurers’ perspective. A higher percentage of applicants secure coverage because applicants denied by one carrier may be issued either stand-alone or combination coverage by another carrier or may receive coverage with the same insurer after a deferral period.

Underwriting Tools
Nine insurers contributed data to Table 22, which divides the number of uses of each underwriting tool by the number of applications processed. For example, the number of medical records was 86 percent of the number of applications. That does not mean that 86 percent of the applications involved medical records, because some applications resulted in more than one set of medical records being requested.

Insurers are trying to speed underwriting to increase placement rates. Thus, phone interviews without cognitive screening, prescription profiles and MIB have increased significantly compared to levels of several years ago.

Year-to-year changes in distribution of sales among insurers significantly impact results. Lower maximum ages result in fewer face-to-face exams. Insurers might underreport the use of an underwriting tool because they may lack a good source for that statistic.

*MIB Underwriting Services alert underwriters to errors, omissions, misrepresentations and fraud on applications for life, health, disability income, long term care and critical illness coverage. MIB, Inc. provides its Underwriting Services exclusively to authorized individuals in MIB Group, Inc. member companies.

Underwriting Time
Table 23 shows that the average time from receipt of application to mailing the policy was the lowest since 2012. The improvement came at both ends of the spectrum, a much higher percentage processed within 29 days and many fewer taking 60 or more days. One insurer averaged only 29 days and four others average about 5 weeks or less. Three insurers averaged 51 to 59 days.

Rating Classification
In 2017, a higher percentage of policies was issued in the most favorable rating classification (and in the top two most favorable rating classifications) than in any year since 2012. Table 24 shows 2018 showed similar results, with 48.3 percent in the most favorable classification and 89.5 percent of the policies in the two most favorable classifications. The “most-favorable” results in the past two years benefited from lack of reporting of rating classification distribution from insurers which do not offer preferred health discounts in the worksite and because one insurer eliminated its preferred health discount (hence “standard” ratings were its “best”). However, the reduction in the third-best and less attractive categories were not influenced by fewer worksite business participants being included in this part of the survey.

Table 25 shows that the percentage of policies issued in the most favorable category increased for each age range in 2018 and Table 26 shows that the decline rate went up in each age range except for ages 70+. Perhaps insurers are declining cases that previously might have been placed in the third-best or less attractive rating classifications. Tables 25 and 26 exclude most (perhaps all) policies underwritten with health concessions.

The by-age decline rates are a little high compared to the overall decline rate reported above. That’s because a significant carrier with a low decline rate was unable to provide their data by age.

Click here to view the following additional information available only online.

• Product Details, a row-by-row definition of the product exhibit entries, with a little commentary.
• Premium Exhibit, which shows lifetime annual premiums for each insurer’s most common underwriting class, for issue ages 40, 50, 60, and 70 for single females, single males, and heterosexual couples (assuming both buy at the same age), based on $100 per day (or closest equivalent monthly) benefit, 90-day facility and default home care elimination period (other aspects vary), three-year and five-year benefit periods or $100,000 and $200,000 maximum lifetime buckets, with and without Shared Care and with flat benefits or automatic three percent and five percent annual compound benefit increases for life. The exhibit includes facility-only policies, as well as comprehensive policies. Worksite products do not reflect any worksite-specific discount.
• Premium Adjustments (from our published prices) by underwriting class for each participant.
• Distribution by underwriting class for each participant.
• State-by-state results: percentage of sales by state, average premium by state and percentage of policies qualifying for Partnership by state.

Closing
We thank insurance company staff for submitting the data and responding to questions promptly. We also thank Nicole Gaspar and Alex Geanous of Milliman for managing the data expertly.

We reviewed data for reasonableness and insurers reviewed their product exhibit displays. Nonetheless, we cannot assure that all data is accurate.

If you have suggestions for improving this survey (including new entrants in the market), please contact one of the authors.

Footnote:
Society of Actuaries (November 2017). Long-Term Care Insurance: The SOA Pricing Project. Retrieved May 16, 2018, from https://www.soa.org/Files/Sections/ltc-pricing-project.pdf.