Professor Marc Cohen has spent many years following the market for long-term care insurance. He’s watched the market from its days of broad-based popularity to its current appeal among primarily affluent policyholders. Cohen recently talked about that market and ideas about how insurance coverage could change in the future. The following is an edited version of his comments.

Why should people care about long-term care insurance, especially if the potential risk may be many years away?

That’s a real challenge. A lot of people misunderstand LTSS and most believe incorrectly the government pays for care, or they don’t understand government pays for care only after people impoverish themselves. For most, the risk is 30 to 40 years in the future and about half will not develop significant long-term care needs. But the risk is highly unpredictable for an individual. You don’t know if you’re going to be struck with Parkinson’s or dementia. Most people also underestimate the cost of home care or nursing homes.

Twenty years ago, private long-term care insurance seemed like a popular idea. Now it’s seen as very expensive for most middle class people and many carriers are no longer interested in offering the product.

Like any insurance, long-term care coverage involved financial assumptions and projections about future claims. Many of them turned out to be off target. As a result, prices went way up and demand for the coverage went way down. Insurance companies worried about their risk and didn’t see much profit potential. In 2000, over 100 carriers wrote a meaningful number of policies. By 2015, there were fewer than a dozen.

What assumptions were wrong?

These policies were designed to be level-premium products. A 50-year-old who bought this coverage could expect premiums would not increase as long as underlying pricing assumptions were met. In the early years, the insurance company collected much more in premiums than it needed to cover immediate risks and set aside most of the money in a reserve to fund future claims.  In pricing the product, the company assumed the reserve would earn a certain level of interest. Many assumed they would make 4, 5 or even 6 percent a year. But we’ve been operating in basically a zero percent environment for the past decade.

So how did that play out?

A one percentage point swing in an interest assumption at age 50 can lead to a big change in the premium. So companies started to lose a tremendous amount of money. Insurers assumed that they could go to regulators, explain the problem and get permission to increase premiums. Consumers got very upset because they thought they had been guaranteed premiums would never increase. They complained to regulators that they could not approve these big rate increase requests. The result was carriers didn’t get the kind of rate relief they wanted, and for many, actually needed.

What other projections turned out to be wrong?

Insurers also missed on the voluntary lapse assumption. Every year, a certain number of people will pay premiums but then drop their policies. If you assume 3 percent of policyholders are going to lapse every year and in fact it’s 0.5 percent, that means more people are going to be part of an insurer’s risk pool and live to become claimants. Turns out this coverage has one of the lowest lapse rates of almost any insurance product.

So what does the long-term care market look like today?

The stand-alone long-term care policy has become a niche product, primarily purchased by upper-middle and upper-class people who want to protect their assets, ensure they can pay for services, and avoid being a burden on family caregivers should they become disabled or require long-term care. Sales of other products that combine long-term care coverage with life insurance, annuities and so on are growing but they may also turn out to be niche products that only reach a small percentage of the market.

So what’s a better approach?

A combination of private coverage and public support. I started out believing we should let the private market address this problem on its own. We’ve now had close to 30 years of experience with private long-term care insurance and the evidence shows there is a role for government intervention. But not exactly the way government support works now. Medicaid is the largest public payer of long-term services and supports in this country. If you don’t have any resources, that’s the social safety net. The main issue is that it’s a welfare paradigm for paying for LTSS, it’s not an insurance paradigm. But insurance is a good mechanism for addressing the long-term care risk because it is very unpredictable, highly skewed and potentially catastrophic for the few who will incur significant expenses.

How would that work?

Here’s an example: If you’re disabled and need long-term services and supports for more than two years, the public sector might pay you $100 a day, say, for as long as you need it – after your first two years of need. But you’ve got to cover that initial two-year period. Are you going to try to save money for that risk or buy insurance? With a private insurance approach for the front-end and a public approach for the back-end, I think you’d see much more interest among carriers to get back into the market.  It would be much easier for consumers to understand where personal responsibility lies and social obligations begin.

What are the financial implications?

For individuals, it’s a two-year risk that won’t break the bank. A pre-funded public program covering everyone not currently in need would not be that expensive because 80 percent of people will never need more than two years of care.

Who else supports an approach like that?

I was a member of something call the Long Term Care Financing Collaborative, a strange bedfellows group of people from the right and left who met quietly for two years in Washington trying to address this issue. We had people from the Heritage Foundation, AARP, Brookings, AHIP, Urban Institute and others. We settled on this general approach in the end. There was this strong sense that people need to take some personal responsibility, and that the government needed to step in where the private sector would not– to provide coverage for catastrophic risks.

Can you see government taking action like that in the current environment?

Politically, it’s hard to believe anyone in Washington would take this up right now. But you would have great interest in the states, where Medicaid is their fastest-growing budget item. And the problem will only become more serious over time. When policymakers are foced to address that, it is important that researchers and practitioners will have done the intellectual “pre-work” on potential solutions.