KEY POINTS:
 
Most LTCI policies include a “guaranteed renewable” contractual provision requiring an insurance company to offer to renew these policies every year for a specific premium, however, the insurer may renew these policies at higher rates to reflect emerging experience.
   
The 2014 NAIC LTC Model Regulation stipulates that any excess of actual past claims over expected past claims cannot be reflected in the loss ratio compliance.
   
It would be inappropriate to use the “Phantom Premium” methodology alone to determine the amount of an allowable rate increase.
   
Restricting premium rate increases in such a way that future claims cannot be funded could have severe financial implications.



1850 M Street NW
Suite 300
Washington, DC 20036
202-223-8196 | www.actuary.org

Craig Hanna, Director of Public Policy
Cori E. Uccello, Senior Health Fellow

© 2018 American Academy of Actuaries. All rights reserved.

Members of the Long-Term Care Past Losses Considerations Work Group include: Dave Plumb, MAAA, FSA—Chairperson; Jamala M. Arland, MAAA, FSA; Jim Glickman MAAA, FSA, FCA; Perry Kupferman, MAAA, FSA; Al Schmitz MAAA, FSA; and Bruce Stahl, MAAA, ASA. The work group would like to extend special thanks to Jodi McAllister for her contributions.

 For a print-ready PDF of this page, click here.


Introduction

Private long-term care insurance (LTCI) is an option for financing future long-term care (LTC) needs. LTCI has received much attention due to the relative size and frequency of premium rate increases. As insurance regulators and LTCI companies have assessed the necessity and justification for these premium rate increase requests, they have discussed which types of past losses should not be recoverable.
 
This issue was considered in the drafting of the 2014 National Association of Insurance Commissioners (NAIC) LTC Model Regulation,i which stipulates that any excess of actual past claims over expected past claims cannot be reflected in the loss ratio compliance. This Model Regulation further stipulates that expected claims are based on original pricing assumptions until new assumptions are reflected as part of a rate increase where those new assumptions are to be used for all periods beyond each requested effective date of a rate increase. Although that 2014 language only applies prospectively to new LTCI policies issued after the effective date of a state’s adoption, it may be reasonable to apply the NAIC guidance on this issue retroactively to other policies. Recently, the NAIC Long-Term Care Pricing Subgroup has explored methods for the premium rate increase process to evolve and become more uniform among the various states and jurisdictions.
 
The American Academy of Actuaries’ Long-Term Care Past Losses Considerations Work Group has developed this issue brief to identify issues commonly raised in favor of and against recoupment of past losses in various situations.

Background

Most LTCI policies include a “guaranteed renewable” contractual provision requiring an insurance company to offer to renew these policies every year for a specific premium. However, the insurer may renew these policies at higher rates, raising premiums on a class basis after receiving approval from state regulators for the specific set of new premium rates. The new premium rates should be based on actual experience together with future anticipated experience. Thus, guaranteed renewable insurance policies only guarantee that the insurer will not cancel the policies; they do not guarantee that premium rates will remain the same after the policy is issued. Unlike the attained age rating methodology used in some other health insurance products, LTCI products are rated on an issue age basis. Issue age rating requires the insurer to establish an expected level premium structure from the date of issue while accounting for liabilities over a very long time horizon.
 
A premium rate increase could be justified when the overall lifetime loss ratio exceeds regulatory minimums. When a block of LTCI is initially priced, the projected profitability is measured using the entire life of the block. While actual experience will fluctuate from expected, the overall performance can still be measured using the lifetime loss ratio.
 
Under the statutory requirements of the 1980s and early 1990s, pricing actuaries were expected to certify that the premium rates were designed to produce a certain minimum lifetime loss ratio. The lifetime loss ratio is defined as the present value of projected claims over the present value of projected premiums. Most states required actuaries to certify that the minimum lifetime loss ratio be at least 60 percent, although some states required different minimums. The intent of this minimum lifetime loss ratio was to protect the consumer from high premium rates by requiring that most of the premiums were paid out in benefits.
 
In the early stages of the product life cycle, LTCI pricing assumptions were based on a combination of assumptions from other insurance products thought to be most comparable. Modifications to these assumptions were derived from population data, as there was no actual experience specific to LTCI products available. In the late 1990s and early 2000s, insurers and regulators began to observe that unfavorable experience was emerging, at least with respect to lapse experience.
 
The NAIC adopted a new pricing certification, requiring the insurer to build a margin into the pricing to cover moderately adverse experience. The regulation of LTCI shifted from facilitating the delivery of a minimum benefit in relation to the premium to a consumer protection focus of reducing the need for future premium rate increases. This new approach was coined the “Rate Stability” regulation. The intent of this new pricing methodology was to reduce the likelihood of future premium rate increases, due to the built-in margin for adverse experience. However, if that margin is exhausted, a premium rate increase can still be filed, justified, approved, and implemented, provided future experience is projected to be worse than originally projected and the other requirements of the regulation are met. This regulation first went into effect in Idaho on July 1, 2001 and was slowly adopted in all but nine states thereafter. However, it was several years before all insurers filed for and received approval for rate-stabilized policies.
 
Current discussions between insurers and regulators include whether premium rate increases can be used to recoup past losses, in addition to any increases needed to avoid expected future losses. There are many possible sources of past losses, including past persistency in excess of expected, past claims in excess of expected, delays in the request or approval of necessary premium rate increases, companies pricing different from industry standards, investment returns being lower than expected, and shortfalls in past premiums.
 

Past Persistency in Excess of Expected

Long-term care insurance is a lapse-supported product: To the extent that persistency is higher than expected, the lifetime loss ratio will be higher than expected. From an incurred claim and loss ratio perspective, higher persistency in the early policy years will generally not materially impact the historical loss ratio. The impact is primarily in the future, as more people are expected to persist to higher-cost claim years, which will result in more claims being paid out in the future than anticipated.
 
Assuming morbidity is as expected, higher persistency generally results in a higher historical loss ratio. However, this higher persistency can be adjusted for in the premium rate increase calculation.


Past Claims in Excess of Expected

There are arguments both for and against recouping past losses due to past claims being greater than expected. Those arguments weigh whether the company should bear all, some, or none of the burden of past losses of this sort, and how active premium-paying policyholders should share in that burden.
 
A key public policy issue surrounding allowing insurers to recoup past losses through premium rate increases centers around the “fairness” of the increased premium rate burden on policyholders who are still actively paying premiums. After a premium rate increase, those active policyholders will pay for future benefits at a higher premium rate. Those active policyholders could also be required to contribute to funding higher-than-expected claims that have already been incurred by other policyholders who are on claim and who are no longer paying premiums. Policyholders should fund the expected cost of future benefits, but it is not clear how much a policyholder should also be expected to fund higher than expected incurred claims by other policyholders who will not share in the burden. The 2014 NAIC LTC Model Regulation states that policyholders should not fund unexpectedly high benefits paid in the past (by only including the lesser of actual and expected past claims in the loss ratio calculation). Therefore, it is critically important for insurers to manage and monitor their inforce blocks proactively and seek any necessary premium rate increases in a timely manner when justified by experience or changes to future assumptions.


State Rate Approval Delays or Limitations

If a state delays or limits an approval of an insurer’s request for a premium rate increase, the company will lose both premium and investment income. Treating such losses as non-recoverable would appear to be unfair and would add a considerable amount of risk to the LTCI product line. Such treatment could also result in subsidization of premium rates across states, which in turn could create an unfair extra burden for policyholders in states that have approved justified premium rate increases in a timely manner.

Insurer Delays in the Filing Process

When premium rate increase filings are delayed, larger future increases will be necessary in order to achieve the same lifetime loss ratio, all else equal. In cases where an insurer should have filed for a premium rate increase but delayed doing so, some would contend the cost of such delay should be borne solely by the insurer. This would mean that policyholders would only be subject to the premium rate increase that would have been necessary had the filing been timely. Such delays could be due to companies not reviewing their data and projections regularly, or by making an error in their projections, thus, not being aware of the need for a premium rate increase. Delays could also be due to not responding in a timely manner to questions from states during the premium rate increase filing process.

However, there may be instances where companies waited for more credible experience—most likely morbidity—before filing for premium rate increases. There is a strong case to be made that this delay not be treated the same as neglecting to file. Even if a company knows that lapse or mortality experience is unfavorable, waiting to get more credible morbidity experience may be appropriate to avoid filing for premium rate increases that may turn out to be unnecessary, should morbidity emerge better than expected.


Pricing Different From Industry Standards

There is a view that some inforce blocks of LTCI may have been priced with assumptions that were different, and in hindsight unreasonable, relative to industry norms and standards at the time of pricing. As such, the financial shortfall resulting from any differences between the pricing assumptions and the industry standard might be the responsibility of the insurer. However, the reasonability of pricing assumptions, the industry standards, and even significant differences in the product itself are highly subjective in advance of emerging experience.
 
For example, an insurer’s pricing may be different from what might otherwise be considered industry standards for legitimate reasons, such as differences in benefits, policy provisions, administrative practices, underwriting, sales distribution channels, sales practices, geography, and target market. The pricing assumptions may also be subject to company-specific experience. It can be difficult to objectively assess how much each of these elements contributed to the pricing of a product, especially when experience turns out to be adverse.
 
Additionally, defining industry standards is challenging and subjective. The benchmarks chosen must be appropriate for pricing. Intercompany studies and industry tables only provide a hindsight view and can be distorted because of the difficulty in isolating differences across multiple insurers that participate.

However, if it is determined that an insurer purposefully used inappropriate assumptions—for example, in order to achieve a higher level of sales— then the insurer should not be allowed to recoup those losses.


Past and Future Investment Returns

Inforce blocks of LTCI are very sensitive to changes in investment income rates, as well as morbidity, mortality, and lapse experience. For determining the lifetime loss ratio at the point of re-rating an inforce block of LTCI, the Rate Stability regulation prescribes the use of “the maximum [statutory] valuation interest rate for contract reserves” (which suggests that the rate is based on the issue years of the block) with a disclosure that “the use of any averages” is permitted (which suggests a single weighted rate). As such, it would appear that investment returns better or worse than expected would be a gain or loss to the company and have no impact on changes to premium rates for Rate Stability products. The rest of this section is focused on pre-Rate Stability products.
 
Pre-Rate Stability regulation is silent on what discount rate to use, and different states vary in their approach to this issue. The 2013 NAIC Model Bulletinii does stipulate that the maximum statutory valuation discount rate be used in determining loss ratios (for pre and post-Rate Stability business). That same bulletin also introduced a dual loss ratio requirement for pre-Rate Stability business, which is 60 percent (or the originally expected lifetime loss ratio, if greater) of the premiums in effect as of the effective date of the new requirement plus 80 percent (75 percent for group) of any increases after that date. While not many states have adopted this Model Bulletin, it indicates the willingness of states to treat pre-Rate Stability products similar to post-Rate Stability products, where companies bear the investment return risk, and where investment returns would have no impact on changes to premium rates.

Across many insurance product lines, for products priced in the 1980s and early 1990s, it was common to use interest rates in excess of 8 percent, since investment rates on investment grade bonds were well into double digit levels. Long-Term Care Insurance: The SOA Pricing Projectiii noted that the average LTCI industry investment income assumption for new LTCI products priced in the year 2000 was 6.4 percent, while in 2014 it had dropped to 4.6 percent. Additionally, there usually is not a single discount rate that impacts the setting of premiums or the profitability of the business. While the expected investment return is important, the valuation discount rate is also important in setting premiums and measuring profitability because the level of reserves held has a material impact on profitability. The lifetime loss ratio calculation does not explicitly account for the impact of investment income—only implicitly in the discount rate used. Any difference between the discount rate used to determine the original pricing loss ratio and the discount rate used to determine the loss ratio at the point of re-rating will have an impact on the repricing of the block. As long as the insurer consistently uses the same discount rate between pricing and rate increase calculations, this concern can be alleviated.
 
For those states that have not subscribed to the Model Bulletin treatment for pre-Rate Stability products, the current sustained low-interest-rate environment is especially challenging. LTCI insurers should be allowed to consider differences in expectation of investment returns when calculating the actuarially justified premium rate increase. Any historical gain or loss from investment returns could be readily determined, although it may not be feasible to fairly attribute historical investment returns to a specific block of LTCI because most insurers manage their investment portfolios at a line of business or total company level. Projecting future gains or losses from investment returns is likely to be quite subjective, as it requires projections of future interest rates, which can be more volatile than morbidity, mortality, and lapse experience.

 

Past Premium Shortfalls

Although the 2014 NAIC LTC Model Regulation considered past losses, there have been continuing discussions about how to treat past losses in premium rate increase filings, most notably in the 2017–18 discussions among the NAIC Long-Term Care Pricing (B) Subgroup.
 
A few states have developed rules on how to adjust for past losses by assuming the new premium was charged since inception in demonstrating compliance with the minimum loss ratio. This is referred to as the “Phantom Premium” approach. This essentially means a policyholder should pay no more in the future than what he or she would have paid had the insurance company known exactly how experience would develop when the product was originally priced. Using this approach raises some serious concerns, which are outlined below.
 
If all the adverse claims experience is expected to be in the future, it follows that there are no past claim losses to recoup, and the higher future premiums are needed to offset the higher future claims. In this case, assuming those higher premiums were paid from the original issue date could expose the insurers to much higher risks retroactively, compared with what they may have believed to be the case when they decided to enter this product line. If this situation arises after, say, two-thirds of the premiums have been paid on a particular policy form, the company could only increase premiums to address one-third of the now- expected additional claims. This approach can cause serious solvency concerns, especially when companies have older blocks of business. Therefore, it would be inappropriate to use the “Phantom Premium” methodology alone to determine the amount of an allowable rate increase.

There are many reasons why an insurer could find itself in the scenario of expecting adverse experience in the future, including:
  1. While early-duration claims may have been better than expected, claims experience at the older ages and later policy durations (after underwriting has worn off) may be worse than originally expected. When the insurer obtains enough experience to recognize the problem at the older ages and later durations, it may still have observed gains in the past or just modest losses, while the significant losses are to be expected to unfold in future years.
  2. A new type of LTC service provider may become available that customers find more attractive and which is more costly than other service providers. A shift to the new service provider can increase the expected future costs above what was originally expected. A typical example of this phenomenon occurred with the emergence of assisted living facilities in the mid-1990s and the eventual mandating of their inclusion by some jurisdictions, in policies previously issued and priced without anticipation of that benefit.
  3. Medical advances or lifestyle changes can extend peoples’ lives so that more people survive to the ages where LTC services are typically needed.
Medical advances can extend the lives of people who are already using LTC services, thus extending the duration of claims.

One of the revisions in the 2014 NAIC LTC Model Regulation prevents insurers from recouping past claim losses by requiring them to use the lesser of actual and expected past claims when demonstrating compliance with the minimum loss ratio. Thus, if companies are experiencing greater claims than expected, and fail to act in a timely manner, those past claim losses cannot be passed on to policyholders through premium rate increases. But premium rate increases are allowed in order to fund increases in future claims expectations, subject to the minimum loss ratio and Rate Stability requirements.

Conclusion

The LTCI premium rates developed by insurers from at least 10 years ago have generally turned out to be underpriced, and thus many premium rate increases have been filed on this business. It is common for LTCI carriers today to anticipate higher-than- expected future claims, and higher-than-expected overall lifetime claims, on their inforce LTCI policies. The guaranteed renewable feature of LTCI is meant to encourage insurers to enter this line of business, and if necessary to revise their premiums to adjust for the greater-than-expected future claims.

The current LTC Model Regulation has been revised to avoid the recoupment of past claim losses. Restricting premium rate increases in such a way that future claims cannot be funded could have severe financial implications and does not follow the 2014 NAIC LTC Model Regulation.
 

i. Long-Term Care Insurance Model Regulation; NAIC; 1st Quarter 2017. Accessed at naic.org/store/free/MDL-641.pdf.
 
ii. Model Bulletin; NAIC; Aug. 9, 2013. Accessed at naic.org/documents/committees_b_senior_issues_130809_ltc_model_bulletin.pdf.
 
iii. Long-Term Care Insurance: The SOA Pricing Project; Society of Actuaries; November 2016. Accessed at soa.org/Files/Sections/ltc-pricing-project.pdf.