• Are fair premiums key to private LTC insurance?

    I give John Goodman credit for writing about actual policy options for improving long term care (LTC) policy in the wake of the demise of CLASS (Avik Roy also has a post; I blogged last week about Stuart Butler’s suggestions). I am glad conservatives are entering the fray on what they are for in LTC.

    However, I think the premise of one of John’s 5 suggestions is wrong, and wanted to point it out since he linked to a post of mine as support for it:

    Allow Insurers to Price Risk Accurately. The principal reason for the failure of CLASS is community-rated pricing. Everyone was going to be charged the same premium, regardless of expected long-term care expenses.

    I think this is wrong. The primary problem with CLASS was not community rating, but adverse selection (only bad risks would sign up). It is true that community rating would have made CLASS even more attractive for poor risks, but if you had a massive risk pool, you wouldn’t only have poor risks and community rating would work just fine. Consider two possibilities:

    • Adverse selection and pricing due to true risk. For example, if you changed CLASS only by allowing for the assignment of an actuarially fair individual premium. In this case there is no community rating, but you also wouldn’t expect a functioning insurance market to develop because no one could afford the premiums due to adverse selection; you would be assigning premiums to only a sliver of those at (high) risk. All insurance markets need good risks to balance bad ones.
    • Forced risk pooling. This could be done either through a mandate, or we could borrow some of the auto-enroll procedures from Rep. Paul Ryan’s Patients’ Choice Act (essentially soft mandates that we agree to call something else). If a large number of persons were forced/nudged into the risk pool, then a market for LTC insurance could probably develop. If you got enough people into the risk pool, then a second step of assigning an actuarially fair premium could make the market more efficient, but of course there is a tradeoff between the transaction costs of assigning a more correct premium and the benefits of the fairness achieved. Once the pool got large enough, it is probably not worth it.

    Assigning an actuarially fair premium makes sense within the (poorly) functioning LTC insurance market, especially since it is voluntary. However, the only way to get such a market started from nothing is forced risk pooling of some sort. Whether such a step is worth it or not depends on how bad you understand the default system to be.


    • Mr. Taylor,

      I’m having a hard time understanding your point of view on this. It seems to me that the problem is really community-rating+adverse selection. Adverse selection occurs because everyone is subject to the same price. So the higher-risk insurees benefit more from the policy and lower-risk lose.

      You then go on to say that actuarially fair premiums would prevent an insurance market from forming, but I can’t follow your reasoning. I would think low-risk and medium-risk people could afford the premiums, but high-risk people could not. You say, however, no one could.

      Maybe I’m reading this wrong.

    • @Jeremy N
      If starting from nothing and trying to create a market for self sustaining insurance (like CLASS), then I don’t think it can work without forced risk pooling. IF low risk people wanted to buy into CLASS then what you say is true, they could afford premiums if actuarially fair ones set. I just don’t think the good risks will be interested without a mandate or nudge of some sort. If there is a way for this to happen sans intervention, it should have already happened.

      If looking at the current private LTC insurance market (that is quite different than CLASS, covers very large benefits for short period v. CLASS-like unlimited period with small benefits) then it is possible that assigning actuarially fair premiums could help make the market bigger (about 7 million covered lives now). There are big problems with this market though, incl aspects of the insurance products that may make it very hard for the market to sustain or expand (benes in $/day, but limits in inflation riders coupled with very long time horizon).

      Bottom line is that I think it depends on whether you are trying to start a new market, or to boost an existing one.

    • Adverse selection is often the result of community rating, so I think Goodman is correct to point to the latter as the problem

    • Re the current LTC market being “poorly” functioning – my impression is that the spate of rate hikes were caused by insurers underestimating morbidity (especially in the lifetime policies), and underestimating earnings from investments. In addition, the premium hikes have been on policies not covered by the NAIC’s new rate stabilization regulations.

      In terms of risk pooling, there doesn’t appear to be adverse selection into the LTCI market that is causing any premium spirals. I once asked an actuary about this, and she said she thought the LTCI market is functional. Where LTCI functions poorly is that the market is small, and it’s too expensive for most middle-income Americans. LTC insurers are also generally able to meet the care needs of their claimants.

      In summary, I think I would say that LTCI is a functioning market, but doesn’t suffice to meet a social goal. In that, it’s even less successful than voluntary health insurance for working adults (face it, this system did cover about 85% of Americans more or less successfully, whereas only about 10% of Americans over age 50 have an LTCi policy).

    • It seems to me that everyone is avoiding the real problem. Insurance works best when everyone is insured. This spreads risk, avoids adverse selection and allows low real prices.
      The CLASS system was designed to fail since it did not include everyone. If it had been made compulsory as part of Medicare there would be no problem. However, the politicians insisted on a free market and voluntary participation model and this cannot be successful.
      What makes Medicare successful is (near) universal participation.
      Medicare is a social program, it is socialist. Everyone is taxed and everyone is included. Get over it. We can’t go back to the “every man for himself” model. The rest of the world has moved on to socialist social services. We can’t repeal the social progress of the last 70 years as much as some would like to try. The CLASS failure is indicative of the effort to go back and it failed. If we want LTC insurance, it needs to be a universal program.

      • @Mark Spohr
        Yes, if you get everyone in the pool community rating, or individual rating will work fine. Whether the policy necessary to achieve this is worth it depends on what you understand the reality of the default to be.

    • The problem I see with John’s ideas is that no one has tried them anywhere. I think the markets are telling us it wont work. His plan is a vote for the status quo.


      • I agree – Goodman’s ideas are, imo, the talking points of someone who hasn’t really studied the issue. Medicaid is already making efforts to encourage community over institutional care.

        Most states already have LTC Partnership programs, and the only other way to encourage LTC insurance is to give tax breaks, which will probably be of dubious effectiveness (benefit < cost).

        Insurers are already allowed to price risk accurately – has he never heard of this thing called underwriting, and substandard (and preferred) risk classes?

        And the bit about mandatory insurance = ponzi scheme? As we've all been over before, voluntary insurance – underwriting = ponzi scheme. voluntary insurance + underwriting = the status quo, a small pool of insured people in LTC.

        I've never been that impressed by the quality of NCPA's analyses.

    • Sorry, Folks, but community rating is never a good idea. People who are high risk are willing to pay more because they know they will need the coverage. People who are at low risk are just the opposite — they may want to be covered “just in case,” but they won’t pay much for it.

      Mandatory coverage does not solve that problem. There is no mandatory program that I can think of that actually works as promised — auto insurance, child support, helmet laws, even paying taxes. Remarkably, about 15% of the affected population is always in non-compliance — about the same number that doesn’t buy health insurance.

      My preference for long term care insurance is to combine it with something else, such as a death benefit. That way, the insured person knows that premiums paid in will be collected one way or the other.