This Part describes the organizational, analytical, policy, and implementation steps taken by HHS to develop the CLASS plan alternatives and prepare for implementation. It consists of seven sections. Section One outlines the offices and divisions within HHS and the roles played by them, and the functions and status of two federal advisory committees created by the CLASS Act. Section Two briefly outlines the HHS process used for identifying policy issues and enumerates the issues identified. Significant documents (both internally and externally developed) that informed policy and implementation discussions are noted. Section Three lists public presentations, along with links to the relevant Congressional hearing record. Section Four discusses the work undertaken to draft proposed regulations. Section Five presents the activities conducted to support marketing the program to employers and individuals, as well as consumer research. Sections Six and Seven describe the development of two actuarial models for conducting estimates for CLASS premiums and the plan options that were developed and modeled, respectively.
Section Three: Public Presentations
The Department presented and discussed its work on CLASS in numerous public meetings and Congressional hearings following enactment. Public presentations included forums and meetings sponsored by AcademyHealth, Alliance for Health Reform, AARP, the Long-Term Care Discussion Group, and the Kaiser Family Foundation (where Secretary Sebelius spoke about CLASS in February 2011; the speech can be accessed at http://www.hhs.gov/secretary/about/speeches/sp20110207.html).
CLASS leaders and staff spoke at national meetings (e.g., 17th Annual Policy Briefing of the National Association of Area Agencies on Aging, Intercompany LTC Insurance Conference) in March, April and May 2011. Administrator Greenlee spoke about CLASS to the American Health Lawyers Association in February 2011 and to the Society of Professional Benefits Administrators in March 2011.
In addition, CLASS was the focus of a hearing held by the House Energy and Commerce Committee, Subcommittee on Health on March 17, 2011. (Written testimony can be accessed at http://republicans.energycommerce.house.gov/Media/file/Hearings/Health/031711/Greenlee.pdf) Secretary Sebelius also discussed CLASS at a March 30, 2011 hearing before the Senate Finance Committee.
Section Four: Regulation Development
The CLASS Office began developing CLASS Act implementing regulations in January 2011, building on the policy option papers prepared by the LTC Work Group and legal advice from the HHS Office of General Counsel (OGC).
The CLASS Office prepared the CLASS Regulations Development Plan (CLASS RDP) document in February 2011. The CLASS RDP established a framework for rulemaking and compiled documents describing: the roles and responsibilities of the various entities participating in the regulation development process; the Secretarys rulemaking authorities and requirements; rulemaking steps; development activities; and other considerations.
Also in February 2011, the CLASS Office began forming the CLASS Regulations Project Team, an interdepartmental group of subject matter experts that included representatives from ASPE, the Centers for Medicare and Medicaid Services (CMS), the Office on Disability (OD), the Office for Civil Rights (OCR), and the Office of the General Counsel (OGC). The team was tasked with providing initial informal review of draft regulations. Ultimately the CLASS Office determined that additional actuarial and legal work was required prior to drafting the appropriate regulatory language needed for the Notice of Proposed Rulemaking (NPRM). After the CLASS Chief Actuary was hired, the CLASS Office recognized that several critical issues needed to be more fully developed internally before regulations could be developed.
Many of the regulations related solely to operational aspects of the CLASS program have been drafted. For example, the CLASS staff has made significant progress in drafting regulations in the following areas: enrollment; waiver of automatic enrollment; lapse in enrollment and disenrollment; reenrollment; payment of premiums; and benefit eligibility. The draft regulations did not address key benefit design issues because policy and legal analysis were still underway.
Section Five: Marketing Research and Ltc Planning
Secretary Sebelius and Administrator Greenlee have clearly stated on multiple occasions that the CLASS program will not go forward unless it is solvent, sustainable, and consistent with the law. Program solvency depends on premiums, benefit payouts, and take-up rates -- enough people buying CLASS policies. Attracting enrollees with lower health risks, people who pay premiums over a long period of time before needing long-term services and supports, is also critical. Achieving sufficient take-up rates and attracting an average mix of enrollees with respect to their health status both depend heavily on marketing.
To prepare for implementing the CLASS program, the Department made a targeted set of investments in consumer awareness and marketing of possible CLASS benefit options that are described below. Not all of the findings from the marketing research are available yet. The research will provide an understanding of: how potential buyers think about long-term care planning; how they make decisions about what and when to buy and how much they are willing to spend; and, how employers think about whether to offer LTC coverage and how they would respond to the opportunity to offer CLASS to their employees. HHS commissioned this research in order to understand whether potential CLASS plan designs would be attractive to a large enough group of buyers. The observations about marketing that are made later in this report rest on preliminary analyses of the marketing research conducted thus far or consultations that HHS has had with experts in long-term care insurance.
In addition, HHS has conducted research for the past fifteen years to understand consumers knowledge about long-term care, their experiences arranging or providing care, their attitudes about planning ahead, and their assessment of their own risk for needing long-term care. HHS has considered findings from this research in formulating and modeling premiums and take-up rates for the proposed plan options.
Initial CLASS Marketing Strategy. Initial planning for a CLASS marketing strategy identified two primary sets of customers -- employers and consumers. For CLASS to obtain a sufficient level of enrollment, marketing campaigns would have to target both groups. To determine how best to market to each group, HHS sought to learn more about their respective attitudes and preferences, then identify those within each group who were most likely to participate in the CLASS program. To prepare for developing marketing strategies for both groups, several research and message development procurements were conducted by ASPE and the CLASS Office in the three following areas: Consumer Research, Employer Research and Long-Term Care Awareness Activities.
Long-Term Care Awareness Survey. In 2010, ASPE awarded a contract to RTI International to design a large, nationally-representative survey to study the attitudes, experiences, opinions and actions of Americans related to planning for long-term care services. The data collection contract was awarded to Knowledge Networks. At the time of contract award, ACA had not yet passed and the purpose of the contract was to gain knowledge for future phases of existing long-term care policy (such as the Own Your Future campaigns). Upon passage of ACA, ASPE expanded the scope of the project to include background research for CLASS. The survey, which is not yet completed, will also employ a discrete choice experiment that will measure individuals preferences for various attributes of plans at specific price points.
Qualitative Research. ASPE contracted with Thomson Reuters to conduct a number of in-person focus groups and interactive discussions as part of the background research for both the CLASS program and the larger survey effort. Participants in the focus groups, which took place in three cities (Baltimore, MD, St. Louis, MO and Edison, NJ) considered the value to consumers of various CLASS program proposals, consumers cost/benefit analyses and their reaction to federal government sponsorship. The research sought to help identify factors that facilitate or inhibit planning for long-term care. Knowledge Networks convened the interactive discussions using members of their KnowledgePanel®. The data from the interactive discussions informed hypothetical questions for the design and administration of the survey mentioned above. Each interactive and in-person discussion solicited from participants reactions, opinions and ideas related to various aspects of long-term care planning and awareness.
Highlights of the findings from the focus groups include: (1) women are more likely to believe that they will need care in their older years compared to men; (2) the belief that one will need long-term care does not necessarily translate into purchasing LTC insurance; (3) many people believe that postponing the purchase of insurance will save money; and (4) many people think that an insurance policy should cover all costs of care; anything less is inadequate. Respondents reacted negatively to vesting periods and complicated benefit plans. Respondents reacted positively to the absence of underwriting, and the option of a cash benefit. Reasons for not purchasing insurance involve cost, first and foremost, but also involve other expenses (such as college tuition and weddings), a perception that insurance is akin to gambling, and the lack of a perceived need for it (particularly among men). Respondents also believed that there should be incentives to purchase long-term care insurance, such as tax deductions.
Development of a Strategic Brand for CLASS. The CLASS Office released a solicitation to develop a strategic brand for CLASS, but no procurement was awarded.
Employer Research. The CLASS Office released a solicitation to assess the potential for employers of all types and sizes to sponsor the CLASS program as a voluntary employee benefit, but no procurement was awarded.
Section Six: Actuarial Model Development
By April 2010, it became clear that existing actuarial models that had been used before enactment of the CLASS Act (both those already relied on by HHS and those being developed by outside groups such as Boston College) would be insufficient to provide CLASS estimates and new models would have to be developed. Actuarial modeling of the CLASS program was undertaken by staff in ASPE, and reviewed by the CLASS Office. The model development and modeling were largely supported through a long-standing contract between Actuarial Research Corporation (ARC) and ASPE, and a new contract with AvalereHealth that began in September 2010. The rationale for developing two models was to compare premiums and other program dynamics using different methodological approaches and data and to assess the sensitivity of results to varying model assumptions. This is standard practice in the insurance industry when developing new products. Further, the ARC model does not include Medicaid offset estimates, while the Avalere model does. The key economic and demographic/actuarial assumptions are largely the same. This section briefly describes the two actuarial models, and model development and estimation across three phases: early model development and estimation; model refinement and development of preliminary benefit options; and final model development and estimation.
Both models adopted conservative assumptions that would tend to produce higher premiums and lower take-up rates than the best existing empirical evidence might suggest. For example, in modeling adverse selection it was assumed that potential enrollees would sort themselves perfectly by health status and join CLASS in reverse order (the most disabled first). The models used conservative assumptions because the existing empirical evidence is relatively sparse and there is great uncertainty around the existing estimates.
Section Seven: Plan Options
Since the passage of the ACA, numerous CLASS plan options have been considered (see the Actuarial Report on the Development of CLASS Benefit Plans, Appendix O, for the CLASS Chief Actuarys description of several of the benefit options). Those plan options whose parameters could be well-specified were modeled using the actuarial models described above, or by the Chief Actuary of the CLASS Office, under various assumptions about adverse selection, and different economic and demographic/actuarial parameters. Although a large number of plans have been modeled, the options can be grouped into roughly three categories: (1) those that are closest to the natural reading of the CLASS statute (benefit plan option one below); (2) benefit options that vary in limited, but important ways from the baseline (benefit plan option two below); and (3) benefit designs that vary much more from the baseline, either because of the sheer number of changes or because of modifications to key features of the program (benefit plan options three through eight).
The models described above estimate premiums for plans under a set of specific assumptions. The most critical of these assumptions are the assumptions around participation rates and adverse selection. Given these assumptions, the estimated premiums are, by definition, actuarially sound. However, the question of long-term solvency of the program depends on whether the assumptions around take-up and adverse selection, as well as other model assumptions, are plausible. As neither the CLASS program nor any other program like it has existed before, there is much greater uncertainty around these assumptions than is the case around the corresponding assumptions for either private long-term care insurance or existing programs, such as Social Security and Medicare. As a consequence, less confidence can be placed in actuarial judgments about the long run solvency of the CLASS program than about corresponding assessments of private insurance or existing government programs.
Existing data sources provide an uncertain picture of what the CLASS claims experience would be. Survey data, such as those used in the ARC and Avalere models, provide information on the entire population but do not provide information on the future claims experience of the CLASS program. Private insurers claims data provide information for those who qualify for private insurance (either underwritten or large group) but do not provide information for the CLASS benefit, which is very different from the typical private market product and targets a more diverse population.
Table 1 presents a summary of the actuarial model estimates for four representative plan options (Options 1-4) that were either modeled for the TEP meeting in June 2011 and the federal actuaries meeting convened by the CLASS Chief Actuary or estimated over the last few months. Because of the uncertainty around parameter assumptions, a range of average premiums is presented rather than a point estimate. Below, we describe each of these benefit plans, provide estimates of premiums, discuss actuarial soundness, and summarize points made in the discussion of these plan options.
BASIC CLASS PLAN
This plan option is based on the most natural reading of the statute and incorporates the key features of the plan described in law (e.g., eligible enrollees must be at least 18 years old and actively employed; there is no underwriting required for enrollment; the primary benefit is a lifetime $50/day [on average] cash payment; before being eligible to receive a benefit, enrollees must wait five years and meet certain work and earnings requirements; etc.). Estimates for this option were produced by ARC and Avalere Health, and are described in Column 1 of Table 1 (baseline). Though the plans cash benefit would increase by the annual percentage change in the consumer price index for all urban consumers (CPI-U), the plan modeled by the actuaries assumes that the cash benefit would increase annually by a fixed percentage, 2.8 percent, which is equal to the long-range inflation forecast published in the 2011 OASDI Trustees Report. The actuaries did this because actuarial models cannot easily estimate future costs when benefits increase by an unknown and variable amount. It is important to emphasize that the 2.8 percent inflation adjuster is for actuarial modeling purposes only; for this option it is contemplated that CPI-U would be used for ongoing program operations.
Under the set of assumptions designated as Scenario II (Expected) (see Appendix Q for Table 2) discussed at the June 2011 TEP meeting, the average premium for a $50/day lifetime benefit with a 2+ ADL trigger (or similar level of cognitive impairment) with full waiver of premium while in claim range from $235/month to $391/month. These estimates are based on a take-up assumption of 2 percent.
In the current private long-term care insurance market, most buyers choose products that provide a substantial daily benefit (e.g., $150/day to $200/day) for three to five years of coverage--daily benefit amounts that are significantly higher than the $50/day lifetime benefit. This could be an issue for marketing CLASS to a broad population as participants in focus groups specifically mentioned that they preferred a benefit that covered more of the total cost of long-term care. Moreover, premiums for products similar to the CLASS benefit, when they are sold to an underwritten population in the private market, would cost much less than the estimated premiums above. Thus, most discussion of this Basic CLASS Plan suggested that the assumed take-up rates used to compute premiums could not be achieved and were not plausible.
MODIFIED CLASS PLAN OPTION
The benefit plan shown in Column 2 modifies three key aspects (highlighted in yellow [ROWS are "Eligibility for Benefits"/"Work Req. Over Vesting Period" and "Earnings Req. Over Vesting Period", COLS are 2 through 4b]) of the baseline CLASS benefit: first, the work requirement during the vesting period is increased from at least three of five years to five of five years; second, the earnings requirement during the vesting period is increased from $1,120 per year to $12,000 per year (the amount of earnings that SSA uses to determine whether a nonblind person is engaged in substantial gainful activity); and finally, the monthly premium is increased annually by a fixed percentage (modeled at 2.8 percent in this example). The latter feature is sometimes referred to as an increasing premium schedule or indexed premium.
Increasing the work and earnings requirement over the vesting period significantly mitigates adverse selection, thus reducing the average premium. In addition, moving to an indexed premium instead of a constant (level) premium lowers the initial premium required to balance expected costs and expected income.
Under the set of assumptions designated as Scenario II (Expected) discussed at the June 2011 TEP meeting, the average premium for a $50/day lifetime benefit with a 2+ ADL trigger (or similar level of cognitive impairment) with full waiver of premium while in claim declines significantly; premium estimates range from $114/month to $160/month. These estimates assume a take-up rate of 2 percent.
The reduction in premiums achieved under this option make the take-up assumption more plausible for the Modified CLASS Plan than for the Basic CLASS Plan. However, the ultimate take-up level is still unknown. The daily benefit amount remains lower than what is prevalent in the private market, which likely increases the risk of low participation rates, especially by those who are able to purchase private policies. In addition, as the federal actuaries noted, the statutory 3 percent limit on administrative costs could make it very challenging to market the product and achieve the expected level of participation. Thus, while the assumed take-up rate used to compute premiums under this model is plausible, there is a high degree of uncertainty about the long-run solvency of this option.
ENHANCED CLASS PLAN WITH PHASED ENROLLMENT
Column 3 of Table 1 shows the key features of a benefit option described in detail in the Actuarial Report on the Development of CLASS Benefit Plans. In various documents it is referred to as the Enhanced CLASS Plan with Phased Enrollment or simply Phased Enrollment. This benefit plan builds off the Modified CLASS Plan, but differs in two important respects (highlighted in blue [ROW "Coverage/Benefits"/"Daily Benefit Amount", COLUMNS "Phased Enrollment"]). First, it uses an explicit two-tiered benefit structure for the first five years that a person is on claim:
- an initial $50/day cash benefit for persons with 2-3 limitations in ADLs
- an initial $60/day cash benefit for persons with 4+ limitations in ADLs or cognitive impairment.
After the fifth year, the daily benefit amount declines by 80 percent. Beneficiaries would therefore receive $10/day and $12/day for the above two tiers, respectively. For modeling purposes, it is assumed that the amount of the cash benefit is equivalent to a lifetime $57.50 daily benefit.
The second difference between the Modified Class Plan and the Enhanced CLASS Plan is that initial enrollment in the program would be limited to certain group settings first, such as large employers; individual enrollment would begin after group enrollment meets a pre-set threshold, explained in more detail by the CLASS Chief Actuary on page 10, Actuarial Report on the Development of CLASS Benefit Plans, Appendix O).
Early modeling of the Enhanced CLASS Plan with Phased Enrollment using the ARC Long-Term Care Premium Model produced an average indexed premium that ranges from $99/month to $106/month for a $57.50/day lifetime benefit with full waiver of premium. A preliminary comparison of age-specific premiums is also shown on p. 14 of the Actuarial Report on the Development of CLASS Benefit Plans.
As observed by the CLASS Chief Actuary, this plan achieves a greater reduction in premiums than does the Modified CLASS Benefit. The range of estimated premiums is also more similar to what is observed in the private LTC insurance market, although the daily benefit is lower in CLASS. Successfully marketing the program remains a serious challenge due to the changing benefit amounts for beneficiaries. The phased enrollment approach could substantially reduce the degree of uncertainty around the rates of enrollment by healthier individuals. By opening the program to individual subscribers only when take-up has reached a threshold level, this approach could manage the risk of adverse selection and potential insolvency.
FAMILY OF OPTIONS: MODIFIED CLASS PLAN & SCHEDULED INCREASING BENEFITS
Columns 4a and 4b of Table 1 describe a set of benefit plans referred to as the Family of Options. One of the options would be consistent with the CLASS statute (e.g., the Modified CLASS Plan in the case of Variation 1). The structure of the other options would vary more extensively, but would continue to incorporate similar requirements for enrollment; a primary benefit that is cash; a five year vesting period; and no underwriting except for age. The Family of Options would be structured to offer either one or two tiers of eligibility for benefits. The Family of Options would be actuarially sound, either at the individual option level or, through cross-subsidization in their entirety. Finally, one of the options within the family would be designed so that purchasers could buy a private (underwritten) insurance product to wrap around this option and provide a higher level of benefit.
Column 4a shows one variation of the Family of Options that includes the Modified CLASS Plan and the Scheduled Increasing Benefits Plan discussed above. (Column 4b shows the corresponding Family of Options with the Enhanced Class Plan with Phased Enrollment paired with the increasing benefit option.) Several features of this plan (highlighted in orange [ROWS "Coverage/Benefits"/"Daily Benefit Amount" and "Minimum Duration in Years", COLUMNS "Family of Options: Variation 1: Increasing Benefits" and "Family of Options: Variation 2: Increasing Benefits"]) differ from aspects of the plans presented in Column 1 and Column 2. Specifically, the daily benefit amount increases the longer the CLASS policy is held without going into claim, rising from approximately $20/day after the vesting period to $150/day after 25 years. Also, the duration of coverage is limited to three years, although the expected payout for this benefit option could be designed in such a way as to be actuarially equivalent to that of the Modified CLASS Plan.
Figure 1 illustrates how the basic daily benefit amount (dark blue area) increases over a 25-year period to $150/day (see Appendix R for Figure 1). This plan is sometimes referred to as the CLASS Partnership because the structure of the benefit provides an opportunity for private insurers to develop products that would naturally wrap around and supplement the underlying basic benefit (light blue area in Figure 1).
If there is no subsidization across benefits options, then the individual plans that make up any set of Family of Options can be priced independently (although specific assumptions related to participation and adverse selection could be adjusted to take into account expected interactions). The range of estimates for an average premium at 2 percent participation assuming a 2+ ADL trigger (or similar level of cognitive impairment) with full waiver of premium is $112 per month to $148 per month. These estimates do not include the cost of a supplemental policy. The total cost of an initial combined policy, for example, for a 50 year old enrollee who could pass underwriting, is currently estimated to be $154 per month ($118 per month for the basic policy and $36 per month for the supplement).
This model achieves a somewhat greater reduction in premiums than does the Modified CLASS Plan. Because of the choice of benefit structure, this option offers benefits more similar to those available in the private market. With private supplementation, purchasers could achieve coverage comparable to that in the private market at similar prices. The design significantly mitigates adverse selection, and premiums do not vary much even under alternative assumptions about take-up rates.3 There were varying opinions about the marketability of the Family of Options design. Some believed that offering choice would be attractive; others thought that it would be burdensome and confusing, especially since the low administrative load for marketing permitted under CLASS would limit the ability to explain the plan. The great uncertainty about the marketability of this option means that uncertainty about the long run solvency of this option is very high.
TEMPORARY EXCLUSION PLAN
This benefit option addresses adverse selection through the claims process rather than the enrollment process. Specifically, any person who meets the enrollment requirements could join CLASS, but no benefits would be paid for the first fifteen years in the program if a limitation in ADLs or cognitive impairment during this period resulted from a serious medical condition that existed at the time of enrollment. The CLASS program would provide enrollees with a list of possibly exclusionary medical conditions, but no health information would be collected at enrollment. Only when a person sought benefits would a review of medical records occur to ensure that the limitation was not the result of an underlying condition at enrollment. Existing data available to the modeling team did not provide sufficient longitudinal information about underlying conditions and subsequent disability to model this option.
This plan would likely reduce premiums substantially because potential buyers with existing health conditions would recognize that they would not be able to claim for pre-existing conditions for fifteen years. There was concern that uncertainty about future benefit receipt would make it challenging to market this option (as purchasers could not be certain that a subsequent disability would not be tied to an underlying condition). Those who could meet an underwriting standard would likely prefer to buy a policy where there was no subsequent uncertainty. See Appendix O (Actuarial Report on the Development of CLASS Benefit Plan) for additional information on this plan option.
TEMPORARY EXCLUSION PLAN WITH PHASED ENROLLMENT
This benefit option combines the features of Temporary Exclusion with phased enrollment as described above. Because the Temporary Exclusion Plan was not modeled, this option was not modeled either. Clearly, the combination of temporary exclusion and phased enrollment would provide substantial protection for the program against actuarial risk. It might, however, be challenging to market this package. See Appendix O (Actuarial Report on the Development of CLASS Benefit Plan) for additional information on this plan option.
LIMITED INITIAL BENEFIT PLAN WITH PHASED ENROLLMENT
This benefit option is analogous to the Enhanced CLASS Plan with Phased Enrollment but has a different benefit structure. While the Enhanced CLASS Plan has a two-tiered benefit that is reduced after five years on claim, this benefit option starts with a low daily benefit amount (e.g., $5 per day or $10 per day) for a fixed period of time (e.g., 20 years) before increasing to its ultimate $50 per day value.
This plan was not formally modeled. While the approach would certainly mitigate adverse selection to a great extent, the initial low benefit and extended period before the benefit increases are unlikely to be very attractive, especially to healthy older workers. See Appendix O (Actuarial Report on the Development of CLASS Benefit Plan) for additional information on this plan option.
PRE-PAID BENEFIT PLAN
Under extreme levels of adverse selection when 100 percent of the enrolled population is eligible for benefits, the monthly premium is essentially the amount that is required for enrollees to pre-pay their future benefit. Because the cost of a pre-paid plan is too high to make it marketable, it is not a viable benefit design. However, the exercise of determining the cost of a pre-paid plan can be instructive, since it provides us with the high end of the range of costs for a plan. The Chief Actuary of the CLASS Office estimated that a pre-paid plan would cost approximately $3,000 per month in premiums. Because enrollees are essentially pre-paying their future long-term care costs, this plan does not include a nominal premium for low income persons and full-time students. See Appendix O (Actuarial Report on the Development of CLASS Benefit Plan) for additional information on this plan option.
In addition to evaluating the formal benefit options discussed above, HHS staff also considered several features, either individually or together, to determine their impact on premiums and program dynamics. The goal was to add specific aspects that would mitigate adverse selection, lower premiums, and increase the marketability of the CLASS program. These features included adding incentive payments for delaying claim, combining CLASS with disability insurance, using variable inflation protection for the benefit instead of a fixed percentage, and possibly returning all or a portion of an enrollees accumulated premiums if he or she died at an early age before going on to claim. Most of the features were eventually discarded because they either did not significantly lower premiums or were deemed to be too complicated to implement.