The Second Failed Attempt At Public Insurance For Long-Term Services And Supports

Long-term services and supports (LTSS) are a broad range of health and health-related services and assistance that individuals of any age may need and use for an extended period of time. This care can be provided in an institutional setting, such as a nursing home, or at the individual’s residence, through home care provided by personal aides. It is generally triggered by limitations in an individual’s ability to perform daily personal care activities such as bathing and dressing, or the need for personal supervision owing to severe cognitive problems.

When it comes to LTSS, one perennial issue is the respective scope of public versus private financing for these services. Medicare currently provides some time-limited benefits for LTSS for the elderly and disabled populations. Medicaid, which covers extended LTSS needs, is intended for poor people. Owing to the exclusion of housing and retirement accounts from countable assets in eligibility determination and the use of various schemes to hide income and assets, Medicaid can be and sometimes is used by middle-class and even well-to-do older individuals for LTSS. Private long-term care insurance is available, but the crowd-out from Medicaid limits its market, and the insurance industry has pulled back from the product in recent years. As the population ages, and the birth rate falls, more and more people will need paid LTSS care in the future, and the budgets of the federal and state governments to pay for the services through existing programs will increase much faster than gross domestic product.  

The CLASS Act

For some time, advocates have pushed for the creation of new social insurance programs to finance LTSS care for the working age and, especially, the older (above ages 55 or 65) disabled populations. A version of a public insurance approach to LTSS was part of the 2010 Affordable Care Act (ACA), called Community Living Assistance Services and Supports (CLASS). Its inclusion in the ACA was motivated largely by the forecasted front-end collection of significant insurance premiums for voluntary federal CLASS insurance coverage for LTSS, to help establish the political claim that the ACA was fully funded. In fact, as I had already shown in January 2010 (see appendix page 19), CLASS was a poorly designed program that would have been subject to a severe adverse selection of risks. That’s because the program was designed without underwriting of insurance policy purchases, and it included various internal subsidies to expensive-to-cover groups and non-working spouses. As a consequence of this approach, those most in immediate or imminent need of the insurance benefits would have purchased the insurance offered by the federal government, while others, of average and especially low-risk exposure, would have viewed the premium charged as too expensive and avoided purchase. Moreover, the policies offered gave inadequate insurance coverage for possible costs of care. Hence, there would have been very low overall take-up for the voluntary public insurance, especially given consideration of an overall lack of awareness and demand among the general public.

With this likely weak initial condition, adverse selection generally results in a dynamic death spiral—large immediate claims and losses, higher and higher premiums charged to cover losses, and fewer and fewer purchases of insurance—and the complete and sometimes quick collapse of the program, unless there are outside interventions and external subsidies. The Congressional Budget Office was mistaken to forecast so much revenue ($70 billion) in the ten10-year budget window from CLASS on the presumption that it was a viable program. CLASS was never implemented. It was abandoned completely by the Obama administration in 2011 and repealed by Congress in early 2013.

WA Cares  

More recently, Washington State decided in 2019 to set up a semi-mandatory state LTSS social insurance program. The program, called WA Cares, was enacted with the strong endorsement of the local Democratic Party over the strong opposition of the local Republican Party and is financed by a payroll tax set at 0.58 percent of workers’ wages (with no cap) of all Washington State workers with at least 500 hours of work in a year. Taxes were to be paid by employees—not employers—starting in January 2022, and the first insurance claims could have been made by those individual workers eligible in 2025 (non-working spouses are excluded). Those employees who attested before December 31, 2022, with no state verification, that they had private long-term care insurance comparable to the public benefit before November 1, 2021, would be exempted from paying the tax but would be permanently barred from the program. Responsibility for tracking exemption status falls on employees and employers, not the state. Self-employed individuals could opt into the program on a voluntary basis within the first three years of the program or on becoming self-employed for the first time. This election, to pay taxes and become eligible for benefits, is permanent until the individual retires or is no longer self-employed. Tribes can opt in or out for any reason and at any time for their members.

Benefits would be paid if the individual was a Washington State resident, at least 18 years old, have either temporarily or permanently vested, and was determined by the state to need assistance with at least three activities of daily living, such as bathing and eating, or severe cognitive impairment. An employee is temporarily vested if they have worked at least 500 hours per year for three years within the past six years from the date of application of benefits. An employee is permanently vested if they have worked at least 500 hours per year for at least 10 years, with at least five of those years being consecutive. Upon becoming eligible, a person would receive coverage of services, paid to providers, of up to $36,500 over the course of person’s lifetime. Services could include in-home personal care, assisted living and nursing home care, respite for family caregivers, transportation, meals, home modifications, adaptive equipment, and so on. 

Every two years, the state actuary must value the program and make recommendations to a state council on any reduction or redesign in benefits to maintain trust fund solvency. The statutory tax rate, however, cannot be increased. 

The council is composed mainly of state department heads, legislators, and interested parties appointed by the governor. It is supposed to recommend service payment rates and annual benefit adjustments, set an investment strategy for the trust fund, and work with the private insurance industry on design of supplemental long-term care insurance policies. This social insurance program is intended to be financed mainly on a pay-as-you-go basis. Although, it does include some pre-funding—hence the need for investment guidelines and oversight. At least initially, however, the state had chosen a low-risk strategy of investing in US Treasury securities.

In 2020, the state commissioned the Milliman actuarial consulting firm for an external actuarial study of the legislated program, projected over a 75-year horizon. The actuarial team found that achieving 75-year program solvency would require a payroll assessment between 0.61 percent and 0.71 percent of payroll (centered on 0.66 percent). With the statutory tax rate of 0.58 percent, the program’s projected cash flow in the first decades was positive and large (initially $1.1 billion) but turned negative in 2052, and the program’s account would be insufficient to pay benefits beginning in 2076. At that point, only 71 percent of benefits could be paid. In 2021, the state actuary calculated that, based on the Milliman analysis, that the program had an initial $15 billion actuarial deficit.

Many assumptions are made to produce this projection; among the more important are the assumed increase in benefits arising from indexing, the extent of adverse selection, and that, based on statute, benefits are not portable for those moving away from the state. Also, it was assumed that there is a 45-day elimination period—that is, benefits do not start immediately with disability, although this feature is unclear in the law. The actuaries assumed that benefits would increase with the consumer price index; if benefits increased with average wages in the state, the premium would need to be 1.16 percent of payroll. If adverse selection was severe in terms of initial exemptions coming from the low-risk population, then the premium would need to be as high as 0.81 percent. If the program design that only Washington State residents are eligible for benefits were changed so that someone who leaves the state is entitled to 50 percent of the value of the benefit, then the premium would need to be 0.88 percent.

The actuaries show that assuming higher opt-out than the low rates used in the base case, and assuming lower opt-in rates (from the self-employed) will not have much of an impact on the premium needed for program solvency. This is surprising because, with only self-attestation required for workers to indicate that they have private long-term insurance coverage, one might have assumed that nearly all higher-wage and young workers would declare themselves exempt and avoid the uncapped payroll tax and small LTSS benefits. Because it is also likely that these workers are generally more mobile, healthier, and work longer than the average covered worker, significant revenues would be lost from these exempt workers who in the long run would have cost the program little in benefits as a present value, on average. This rational dynamic would surely necessitate a much higher premium rate needed for program solvency than the base case. It is concerning that the assumed extent and identity of those exempting themselves from the program were set by state officials in the base case, as stated explicitly in the actuarial report, and perhaps did not represent the actuary’s best judgement; hence, the assessment of the actuarial report used in evaluating the program was not arrived at independently. 

Another actuarial analysis, issued in November 2021 by Milliman, showed that the LTSS program would save money for the state and, mostly, the federal government in the form of Medicaid resources. In 2019, Washington State spent $2.1 billion on Medicaid LTSS. In 2025, the actuaries project that Medicaid would be spared $70 million in costs covered instead by WA Care. Those savings would increase (in 2025 dollars) to $140 million in 2040, to $290 million in 2060, and to $360 million in 2080. The increase in savings arises as more people become vested in the program benefit. One might ask whether there are any net savings to the government, assuming a solvent LTSS program, viewing Medicaid and the state LTSS program together. The answer depends mainly on whether the purchase of private long-term care insurance or increased savings dedicated to LTSS are incentivized or expensive LTSS services avoided.

Program Failure  

After several weeks of discussion and mounting pressures, discussed below, on December 17, 2021, Governor Jay Inslee and Washington State Democratic leaders of the legislature announced that collection of the payroll tax of the LTSS program would be delayed until at least 2023. Their justification statement focused on the aspects of the program whereby near retirees, non-residents, military families, and those leaving the state to work elsewhere or to retire could not claim benefits even though they paid at least some taxes for the program. Of course, extension of the benefits to these groups will raise program costs and the necessary significant increase in premium rates. Also, they expressed concern that there needed to be assurance that those who opted out would maintain their private insurance policies and not allow them to lapse. Indeed, insurers had stopped selling long-term care insurance in Washington because of this concern about churning.

As of early December 2021, 443,649 exemption requests were filed with the state; the average hourly wage of people seeking exemptions was $63.60—clearly higher-wage workers, confirming the adverse selection scenario I put forward above. The pace of exemption requests was increasing in November, and workers had until December 2022 to file. (There are about 3.5 million wage workers in Washington State, including part-time workers.) It is reasonable to surmise, given the moratorium by insurers on product sales and that penetration by private long-term care insurance is less than 5 percent nationwide, that most of those attesting that they had insurance coverage did not. It is also worth noting that the $36,500 lifetime program benefit would be quickly used up at the 2020 average costs of services in Seattle—according to Genworth, home care there costs $219 a day, and nursing home semi-private room costs $349 a day. Also, compared to a 90- or 120-day elimination period, a 45-day elimination period adds costs; the shorter period is probably unnecessary given that Medicare covers most short-term LTSS. 

In fall 2021, a group of employers and workers filed a class-action lawsuit against the program requesting a declaratory judgment that the act establishing the program was unenforceable as it violates the Employee Retirement Income Security Act and federal and state laws governing employee benefit plans. Also a referendum was filed to allow workers to opt-out of the program at any time and for any reason. The attorney general of Idaho also filed a cease-and-desist order that Idaho residents working in Washington State be exempted from the program.  

In short, we have another public LTSS insurance program that is deeply unpopular, poorly designed, unstable, insolvent ab initio, perhaps illegal, and, without major changes, failed.

Does anyone want to try a third time, or should we instead focus on improving the mixed public-private system we currently have to be sustainable for federal and state government budgets and to encourage private resourcing of LTSS care needs?