At the beginning of the development process, we endeavored to generate a list of all possible designs we could think of. We ranked them into 3 groups:
- those that do not address adverse selection and have no market appeal,
- those that address adverse selection but have no market appeal, and
- those that have both23.
From the last group, we refined the concepts and came up with the following plan designs. They are presented from an actuarial, not legal, perspective.
Limited Initial Benefits
This plan pays a low benefit (for example, $5 daily) if claimed during the first 20 years and a regular benefit (for example, $50 daily benefit, indexed) otherwise. Because the benefit in the early years is quite low relative to the premiums, the Program may be less attractive to the high risk workers. Thus this approach can moderate adverse selection. The disadvantage of this design is that it may also be unattractive to the healthy workers, especially older workers who, as a whole, are closer to claim.
Before discussing Phased Enrollment, it is helpful to review the types of risk mitigation practices typically used in the group long-term care insurance market. Insurers offer group coverage in the workplace with endorsement from the employers. Premiums are usually collected through payroll deduction. For large employee groups of over 500, insurers would generally offer guaranteed issue to full-time active workers during a limited enrollment period (1-2 months). Some form of underwriting is required for those enrolled outside of this period and for spouses and other immediate family members. New hires can enroll without underwriting during a short period following hire. Presumably, insurers are protected by the very limited window of opportunity for enrollment as opposed to a situation where the workers can join whenever it is to their advantage24. Enrollees tend to be white-collared workers who have available discretionary income for insurance and who are generally in better health status than the average worker. To date, premium rate increase filings for group long-term care insurance are uncommon. The overall claim experience appears to be close to the anticipated claim costs in the development of the premiums.
The statute specifies two methods for workers to voluntarily enroll. Workers can enroll either through their participating employers or through an alternative individual method. Individual enrollment is for the self-employed, individuals with more than one employer, and for those workers whose employers are not participating in the Program. The statute is silent regarding the manner and timing for implementing either method.
The Phased Enrollment design adopts certain risk mitigating practices from group long-term care insurance. Under this design, enrollments with be phased according to a pre-determined risk tolerance. Initial enrollments will first be limited to the group setting through employers with a specified minimum number of employees. The initial enrollment period will be 1 to 2 months. Subsequent enrollment periods will be indeterminate. This is similar to the current practice for guaranteed issue group long-term care insurance.
Such practice is workable in mitigating adverse selection because:
High risk workers cannot seek insurance. Only those whose employers participate can enroll.
Enrollment period is short. There is little opportunity for a working individual becoming unhealthy to enroll at will.
The relatively large size of the groups provides a spread of risk.
The alternative individual enrollment method allows any workers not included in group enrollments to sign up. Without a way to temper high risk workers to enroll as in the case for group enrollment, we expect most of adverse selection to come through this method. Accordingly, individual enrollment will commence in a controlled manner to ensure solvency of the Program.
Premiums will be set with a specific margin that provides for a reasonable allowance for individual enrollment in the future. The annually required actuarial analysis25 will specify the number of future individual enrollments.
The following is an example of how Phased Enrollment works. A 20% load is added to the premiums for CLASS specifically for expected higher claims through individual enrollments. Suppose we determined that enrollees from individual enrollment can be expected to be 5 times more likely to claim than those through group enrollment26. When the number of enrollees reaches 200,000 and provided that the claim assumption remains unchanged, we would allow up to 10,000 (= 200,000 x 20%/400%) new enrollees through individual enrollment.
Thus, under the Phased Enrollment design, the degree of adverse selection is rationed based on what is available to accommodate the expected higher claims. Individual enrollment starts when the group enrollment meets a pre-set threshold. We will allow a pre-determined number of workers to enroll during a limited individual enrollment period. In order to validate the adverse selection assumption, we plan to evaluate individual enrollees’ medical records but naturally would not use them to deny enrollment. Lessons learned from the evaluation will help us to devise subsequent individual enrollments, including enrollments of employees from smaller sized employers.
The advantage of this design is that the impact of adverse selection is controlled. Note that this design does not materially alter our expectation of the ultimate number of enrollees in the CLASS Program since we expect that the majority of enrollees would come from group enrollment anyway. Phased Enrollment therefore does not mean the CLASS Program will necessarily be small. The disadvantage is that the margins designated for individual enrollments could be depleted due to overall unfavorable experience of the Program. The timing and the allotted number of individual enrollments are not guaranteed. There may be public pressure to expand individual enrollments thus causing harm to the Program. Finally, additional legal analysis will be required to determine whether the Phased Enrollment plan must be available to all employers.
This design aims to control the impact of adverse selection through the claim process rather than the enrollment process. Under this design, no benefits will be paid during the first 15 years27 of an individual’s enrollment if the qualifying ADL (Activities of Daily Living) or cognitive deficiencies can be determined to be the result of a prevailing serious medical condition that existed at the time of enrollment. At the time of enrollment, the enrollees will have access to a list of such conditions and acknowledge that they understand this restriction, if applicable. This list would be similar to the list of uninsurable conditions in private insurance’s field underwriting manual. We will not collect medical information at time of enrollment. At time of claim, medical records will be reviewed to determine whether a serious medical condition existed at enrollment caused the deficiencies. If no determination can be made, or medical records are not available, we will pay the claim. From experience of the private group market, we expect that the majority of the enrollees will be relatively young (in their 40s). Because of this expectation, we believe the vast majority of beneficiaries with functional limitations during the first 15 years of their enrollment will have a medical origin, rather than due to frailty.
Qualifying deficiencies due to conditions not on the list or conditions developed after enrollment are eligible for benefits after the 5 year vesting period, assuming other benefit eligibility requirements have been met. This design puts the high risk enrollees on equal footing with the other enrollees, enabling the law of large numbers to work.
Temporary exclusion is a risk mitigating technique used in life and disability insurance. We expect claims under this design to be estimable even though they would still be higher than claims from guaranteed issue group plans.
The advantage of this design is that all workers are treated fairly under the principle that enrollees insure only their unknown risk. The degree of subsidization among the workers is expected to be acceptable. The disadvantage is that this design could be perceived as a pre-existing condition exclusion where the insured is denied coverage or permanently excluded from claiming under the condition. Covering pre-existing conditions is possibly workable in an insurance program where there is substantial participation (for example, over 80%) to spread the extra risk. It violates the insurance principle of homogeneity of the insured group in programs such as CLASS where participation is expected to be low (typically less than 10%)28.
Scheduled Increasing Benefits
This design is similar to the Limited Initial Benefits. If the enrollee qualifies for benefits, it pays a $20 daily benefit if claimed during the 6th years of enrollment. The $20 daily benefit increases by $6.50 each year of enrollment to an ultimate of $150 at the 26thyear and thereafter. The benefit will pay for a maximum of 36 months only. The intent is to discourage high risk individuals to enroll given the relatively low benefits in the early years. In order to comply with the statute, we believe the potential enrollees must choose between this design and the pre-paid plan as described in the Adverse Selection section.
In order to attract workers who desire a richer benefit, this design can work in conjunction with a private insurance offering. Together with coverage from the private plan, the total benefits of the package will pay a level $150 from the first year of enrollment. Thus the private plan will provide coverage during the vesting period of the CLASS plan and the coverage will decrease each year to no coverage after 25 years of enrollment. To provide inflation protection, all daily benefits described for this design will increase by 3% each enrollment year.
The advantage of this design is that the effect of adverse selection is controlled to a certain degree but not eliminated. To be actuarially sound, this design will probably be more expensive than private insurance. It will be a challenge to compete with the more straightforward $150 daily benefit private insurance plan. Another disadvantage is the complexity of the benefit schedule and the offering the pre-paid plan as well as the private insurance supplemental plan.
Actuarially Sound Plans
The above plan designs are not mutually exclusive. A plan combining these concepts can address the shortcomings of the individual design. According to the statute, the Secretary will present at least 3 actuarially sound plans to the CLASS Independence Advisory Council for review. The Council will recommend one of them to the Secretary for designation as the CLASS Independence Benefit Plan. For consideration, we propose the following 6 candidates for actuarially sound plans29, some of which are combinations of the plan designs presented above:
Phased Enrollment as described above.
Phased Enrollment with Temporary Exclusion for Individual enrollment.
Phased Enrollment with Limited Initial Benefit for Individual enrollment.
Temporary Exclusion as described above.
Scheduled Increasing Benefits as described above.
Pre-paid plan as described in the Adverse Selection section.
This proposal is tentative since we have not yet completed our plan development process which will include final premium determination and further legal clearance.
In order to ensure success of the CLASS Program, all designs will also need to pass the marketability criterion. For example, the pre-paid plan assumes that there is no barrier to prevent or discourage who might be attracted to the program. Such a plan can be made actuarially sound if we assume that virtually everyone enrolled will claim. Thus the premiums (estimated to be $3,000 per month) become that of a pre-paid arrangement. There will obviously be few enrollees.
The first 5 proposed plans are designed to mitigate the risk for adverse selection to a certain extent. Premiums for some of these plans can be made competitive with private insurance because the expense and profit savings could partially offset the anticipated extra claims.
Using comparable plan features, the following preliminary comparison illustrates the pricing position specifically of the Phased Enrollment plan relative to the offerings from three current group insurers30:
The CLASS premium estimates for the Phased Enrollment design above are derived from Genworth’s net premium rates (net of anticipated expenses and profits) adjusted for:
- CLASS-specific expenses,
- a loading for adverse selection from individual enrollment,
- a margin for unfavorable experience, and
- an increasing premium methodology (see Increasing Premium section below).
Genworth’srates are chosen because we believe that the associated pricing morbidity, mortality, lapse and investment assumptions are the most current among the three insurers. It is important to recognize that these are preliminary illustrations of premiums, not final pricing. As noted in the Next Steps section, premium estimation needs to be further refined with experience claim data, among other steps.
Since the Phased Enrollment process is similar to the enrollment process for guaranteed issue group plans, we expect that claim assumptions for the CLASS plans to closely reflect the claim experience of that segment of the private group market. As discussed earlier, residual effect of adverse selection from private insurance’s guaranteed issued group enrollment is already embedded in group long-term care insurance’s claim experience and their premiums. There are approximately 1.5 million certificates of group long-term care insurance currently in force. A survey is underway to collect recent claim assumptions for guaranteed issue group plans from six insurers that have group in-force business. This data source can provide considerable confidence in setting claim assumptions for a number of the proposed CLASS plans.
Summary on Plan Design
Having a competitive CLASS plan also creates an incentive for private insurance to work side-by-side, rather than directly compete, with the CLASS Program. Co-marketing with insurers (discussed below in Marketing Considerations section) should reach more workers and provide a better spread of risk.
The proposed plans provide a potential array of choices to achieve the goals set forth in the statute: “balance price and benefits while optimizing the probability of long-term sustainability of the Program”31. As will be covered in the following section, besides adverse section, other issues need to be addressed such that the proposed plans are competitive and relatively free from moral hazard.