Introduction by Croakey: What sort of health policy relies on “coercing the wealthy” and “conning the young” into supporting a product that arguably adds to health inequities and inefficiencies?
To find out more, dive into the #LongRead below by former senior public servant Charles Maskell-Knight, who has some suggestions about what policymakers can learn from the tortuous history of private health insurance.
His #LongRead, shaped by a lecture to public health students, provides a masterclass om the evolution of private health insurance in Australia, where he says the voluntary community rated system “is really a Ponzi scheme” that has long been unsustainable.
His analysis comes as the costs of private health insurance are set to rise, after insurers froze or deferred premium increases during the pandemic. Maskell-Knight expects that living pressures and rapidly rising mortgage repayments will also likely kick off another ‘death spiral’ for the hugely subsidised industry.
Charles Maskell-Knight writes:
When the Croakey team was discussing arrangements for Budget coverage, I said that I would be away in Canberra after the Budget giving a lecture on the private health policy problem to a Master of Public Health course at the ANU.
Never one to pass up a possible contribution, Croakey editor-in-chief Melissa Sweet asked if I could turn the lecture into an article. Here it is – revised to reflect good ideas that emerged during the lecture and the accompanying discussion with a very engaged group of students.
International context
Let’s begin with some international context. Across the OECD most countries have a publicly funded system of health provision. Private health insurance (PHI) – if it exists at all – operates at the margins of the public system. On average, across the OECD, PHI funds about 10 percent of health expenditure.
However, there is a small number of countries where the government contribution to the health system is provided as a subsidy to PHI intermediaries, and participation in PHI is effectively compulsory. In countries such as the Netherlands and Switzerland, PHI thus funds over half of health expenditure.
Elsewhere participation in PHI is voluntary, and PHI can fill one of three functions. It can be:
- supplementary: covering services not in the public offering
- complementary: covering co-payments for public services
- duplicate: providing faster access or choice of providers.
Facts and figures
In Australia PHI funds about 11 percent of health expenditure on a gross basis. But because the government provides a premium rebate of about 25 percent, the net contribution to health expenditure by PHI is 8.2 percent.
In Australia, PHI covers:
- the cost of treatment in a private hospital
- a share of the cost of treatment in a public hospital (for patients electing to be treated privately)
- a share of medical costs for hospital services
- on average, half of the fees for dentists, physiotherapists, glasses, and other allied health services – so called ‘general treatment’.
About 45 percent of people have hospital insurance, and 55 percent have general treatment insurance (in some cases covering only ambulances because of state government charging policies). Over 20 percent of people with hospital cover are aged over 65, compared with 17 percent in the general population. As normal migration patterns resume, the proportion of people aged over 65 in the general population is likely to decline for a number of years.
In terms of the OECD classification, PHI in respect of hospital services is essentially duplicate. People with PHI can use private hospitals and the doctor of their choice, and can be treated faster, but they essentially get the same range of services that are available in public hospitals.
In respect of general treatment, PHI is supplementary. People with PHI get benefits for services that are largely not available in the public sector, or only available for low-income groups.
The law forbids PHI from offering complementary cover for out-of-pocket costs for pharmaceuticals, GP attendances, or specialist consultations outside a hospital.
PHI is mainly a hospital insurance product, with spending on hospitals accounting for almost two-thirds of total PHI spending. On a gross basis, this makes up 15 percent of all hospital spending.
Just under a quarter of PHI spending is on general treatment. Even though there is little public spending on these services, PHI gross spending makes up only 27 percent of total dental spending, for example.
The remaining PHI expenditure – about 11 percent – goes on administration. PHI is both inefficient and profligate – PHI administration makes up almost half of total health administration costs.
The structure of the PHI sector is like many other sectors in the Australian economy – dominated by a small number of very large providers. While there are 32 insurers, the top five cover 82 percent of the market. The smallest five cover 0.3 percent of the market, or 31,000 people. Fun fact: the largest insurer employs more people than the smallest one covers.
Most small insurers avoid some of the diseconomies of small scale by belonging to the Australian Health Service Alliance, which acts as a buying group in negotiations with health providers.
Community rating and risk rating
The system of PHI premium setting in Australia makes overseas visitors shake their heads in bewilderment.
Consider the following four people:
- Ahmed: aged 27, fit, healthy, works in an office, volunteers as a surf lifesaver
- Belinda: aged 42, overweight, regular drug user, casual worker
- Chloe: aged 62, has diabetes, a smoker, living off savings
- Dimitri: aged 71, arthritic, bad knees, history of heart disease, retired.
In most countries their PHI premiums would differ greatly. But in Australia, the law requires insurers to charge them all the same premium, if they choose to purchase the same product.
Section 55-5 of the PHI Act 2007 prevents insurers from discriminating on the basis of:
- health history
- treatment history
- benefits paid
- age
- gender
- race
- sexual orientation
- religious belief
- residence
- occupation
- leisure pursuits
- any other factor likely to affect their need for health services.
This is quite unlike any other kind of insurance sold in Australia, which is risk-rated. Young inexperienced drivers pay higher car insurance premiums than older ones because they are more likely to be involved in accidents. People living in flood plains pay higher home insurance premiums than people at the top of hills, because they are (increasingly) more likely to be flooded. Actuaries make a lot of money figuring out how much to charge for different risks.
PHI is community rated, meaning that everybody in the insurance pool pays the same premium, regardless of risk.
All insurance involves a cross-subsidy from people who don’t make a claim to people who do. If insurance is risk rated the subsidy flows between people of equal risk – all policy holders paying the same premium are equally likely to make a claim in any given year.
Community rated insurance involves an ongoing subsidy flowing from people of low risk to people of high risk. Low risk people pay higher premiums than their risk justifies to support premiums for high risk people that are lower than required to meet their likely claims.
Across the OECD community rated PHI is usually associated with systems where government support is provided through PHI, and PHI participation is effectively compulsory.
Australia (and Ireland) are exceptions – and Ireland is only the same as Australia because it effectively copied the Australian system of PHI regulation late last century.
Community rating is required in a compulsory PHI system to address equity concerns. If people are to be compelled to purchase PHI, it would be grossly unfair to require high risk people to pay risk rated premiums.
Why is Australia one of only two countries to have community rated voluntary PHI?
But given Australian PHI is voluntary, why is it based on community rating?
The answer lies in the dead hand of history. Before Medicare, Australian government assistance for health costs for people other than pensioners was only provided to insured people. Hospital bed day subsidies and medical benefits were only available to the insured, so the equity argument for community rating was strong.
When Medicare was introduced, the government explicitly considered whether community rating for PHI should continue. It concluded that it should, because older, long-serving policy holders who had contributed over many years would be severely disadvantaged if risk rating was introduced and they were priced out of the market.
This approach demonstrates that community rated PHI is really a Ponzi scheme (or a Madoff scheme, or a Caddick scheme, to use more contemporary references).
It relies on younger people subsidising benefits for older people. And as those younger people age, insurers need to replace them with more younger people to meet the costs of the ageing cohort.
For some low-cost products sold to younger people, as much as 90 percent of the premium goes to cross-subsidising other products. Younger people can sense that PHI premiums are not actuarially fair, and they decline to pay them.
Over the first decade following the introduction of Medicare, private health insurance participation almost halved, and the ratio of people aged under 65 to those 65 and over fell from 9 to 6.
A so-called death spiral began. Fewer younger people cross-subsidising older people meant premiums increased faster. And as premiums went up faster, more younger people dropped out, meaning premiums increased even faster, meaning more younger people dropped out.
Government response
The Howard Government took office in 1996 with a commitment to solve the problem by subsidising premiums. Over the next few years it developed a tripartite strategy:
- subsidise PHI premiums
- coerce the wealthy into buying PHI
- con younger people into buying PHI.
This overall strategy is still in place, albeit different elements have been more or less successful over the years.
Subsidising premiums
The first subsidy scheme introduced in 1998 was the Private Health Insurance Incentive Scheme, which paid a maximum of $600 for a family policy. It had next to no impact on the participation rate, and was replaced in 2000 with a 30 percent premium rebate, which also had next to no impact.
Why was this so? All through the 1990s a market research firm called Tony Quint and Associates had been carrying out a survey on attitudes to PHI, and regularly asked people without PHI if they would purchase it if the premiums were 30 percent lower. About a quarter of people surveyed said they would.
In the event, of course, they didn’t. A post-rebate premium of $2,100 may have been 30 percent lower than the original price, but it still required households to find $2,100 by spending less in other areas. Very few households valued PHI enough to reallocate spending to that extent.
Coercing the wealthy
The Medicare levy surcharge on well-off people without PHI was also introduced in 1998. Single people on an income of $50,000 or more (at that time more than 40 percent higher than average weekly ordinary time earnings had to pay an additional one percent Medicare levy if they did not have PHI.
This measure also had a negligible impact on participation. Most well-off people already had PHI, and at least some of those who did not had an ideological commitment to a universal scheme.
However, since 1998 the threshold for the Medicare levy surcharge has only been increased occasionally, and the requirement to buy PHI or pay the Medicare levy surcharge has applied to more and more people. The threshold is now $93,000 – $1,000 lower than average weekly ordinary time earnings.
Market research carried out five years ago suggested that the Medicare levy surcharge was a factor affecting between a quarter and a third of people in their decision to buy PHI.
Conning the young
The most successful strategy by far was conning younger people into buying PHI through the Lifetime Health Cover campaign in 2001.
The basis of Lifetime Health Cover was the idea that people purchasing PHI in young adulthood would pay premiums for about 50 years until they died. They would on average contribute more than they cost for the first 35 years, and then draw out more than they paid for the last 15 years.
Someone buying PHI for the first time at age 40 had skipped on paying ten of their 50 years’ worth of premiums. They should be compelled to contribute the missing premiums by paying a premium loading once they did buy PHI.
Logically the loading should have increased exponentially – rather than linearly – as people’s age of purchase increased, but this was deemed far too complex to market. So the decision was made to impose a premium loading of two percent for every year after 30 that somebody deferred buying PHI for the first time (up to a maximum of 70 percent).
A massive award-winning advertising and public relations campaign preceded the measure coming into effect on 1 July 2001. When the dust had settled, the number of people covered had increased by about 50 percent, and the ratio of under 65s to 65s and over had almost returned to the immediate post-Medicare level.
Part of the success of the introduction of Lifetime Health Cover was probably due to the fact that most of the group it was targeting had experience of buying PHI in the pre-Medicare era. They still had some vestigial attachment to the idea of private health insurance.
A number of years after its introduction the government changed the rules to limit the premium loading to ten years, in response to ageing baby-boomers whinging to government MPs that they were being ‘locked out’ of PHI. This severely limited the efficacy of the measure, and in market research five years ago hardly anyone mentioned Lifetime Health Cover as a factor in their purchase decision.
While the government these days sends everybody turning 30 years old a letter about Lifetime Health Cover, there is only a very small jump in participation by 30- and 31-year-olds compared with 29-year-olds. People in this age range were born almost a decade after the introduction of Medicare, and have no memory of PHI as a desirable product. The ratio of under 65s to older people is now as low as it was before Lifetime Health Cover was introduced.
When the government in 2018 allowed insurers to introduce discounts of up to 10 percent for people joining PHI before age 30, we did a straw poll in the office and among our families of every eligible person we knew. Would they buy PHI now it might only cost $2,430 rather than $2,700 for a medium range product? The answer was a resounding “no”.
Insurer response
Insurers have adopted two responses to the difficulties facing PHI.
Reduce input costs
The first is to demand government action to reduce hospital costs.
They quite rightly point out that they are required to pay benefits for prostheses (such as artificial joints or lenses) at a level far higher than overseas prices for the same items.
They also (quite rightly) question the proliferation of day hospitals, often providing services such as surgery for skin “lumps and bumps” that could be done safely in doctors’ rooms.
They argue vehemently and irrationally that the existence of a default benefit for non-contracted hospitals at 85 percent of the contracted rate artificially pushes up the contracted rate.
But even if the government waved a magic wand and solved all these problems, there might be a once-off reduction in premiums of 10 or 15 percent. The evidence from the introduction of the 30 percent premium rebate suggests that this will have a minimal impact on participation.
Subvert community rating
Insurers’ second strategy has been to subvert community rating.
During the 1990s and early 2000s my colleagues and I spent a great deal of time addressing ingenious schemes by insurers to offer premium discounts to low risk groups, or provide what amounted to no-claims bonuses to policyholders fortunate enough to have enjoyed good health.
On the other side of the coin insurers were diligent in avoiding payments wherever possible. I once had to deal with an insurer who had sold a family policy to a couple who were expecting a baby which was diagnosed in utero with a major heart defect which would require prompt surgery after birth. The insurer wanted to know if they could apply the pre-existing ailment rule – precluding payment of benefits for a year for a condition known at the time the person began to be covered by the policy – to the baby.
However, insurers’ great success was in the early 1990s in persuading the then government that exclusionary policies were not inconsistent with community rating. An insurer decided that it could lower the cost of products by excluding some high-cost and relatively high-volume conditions from coverage and, following consideration within government, this was approved.
After a few missteps with obstetrics (over 65s gleefully bought cheaper products that excluded obstetrics but included all the conditions of old age), insurers began to exclude hip and knee replacements and cataract surgery that many older people require.
The sales of exclusionary products grew, as did the extent of exclusions. It is now possible to purchase a singles hospital policy for $1,200 a year which excludes everything except treatment for an accident (which, of course, will take place in a public hospital).
However, this policy still has to contribute to the risk equalisation arrangements, under which the marginal costs of all over-55s are pooled and allocated across all insurers on a per policy basis. A person buying a $1,200 policy with virtually zero coverage is making a significant contribution to the costs of policies which offer coverage for the full range of hospital services. Those policies cost about $4,000 a year for a single
Evidence suggests the market is now split in three ways. About 45 percent of policyholders have very basic cover, while another 45 percent have fully comprehensive cover. The other ten percent are in the middle somewhere.
Through the risk equalisation arrangements the bottom 45 percent subsidise the top 45 percent by about $4 billion a year.
Where is Australian PHI now?
After declining for a number of years through to 2020, people covered by PHI have now increased by 400,000 or 3.5 percent. This appears to be due to a combination of COVID-induced anxiety about access to public hospitals, and the Medicare levy surcharge affecting more and more people.
However, this pattern may soon be reversed.
Cost of living pressures and rapidly rising mortgage repayments will reduce the number of people who can afford thousands of dollars for a discretionary purchase. For many people it will be cheaper to pay the Medicare levy surcharge rather than pay a PHI premium.
This means that the decline in population coverage will continue. And as more younger people drop out (or don’t join) and the people covered grow older, premiums will continue to increase faster than CPI or wages. More people will then drop out: the death spiral will resume.
Voluntary community rated PHI is not sustainable.
Future policy options
The options are: either move to coerce more younger people to buy PHI, or move away from community rating.
Both are politically difficult.
People aged 40 and under are facing HECS debts increasing faster then they can repay them; housing that is unaffordable to buy, or even rent; and increasingly unstable employment. The Medicare levy surcharge income threshold would need to be reduced and the levy rate increased to coerce enough people into PHI to have a significant impact on overall participation and the age profile.
A move away from community rating would also have problems. The introduction of Lifetime Health Cover in 2001 carried with it an implicit commitment from the Federal Government that it would support the Ponzi scheme into the future. People aged 40 who bought PHI in 2001 to avoid the Lifetime Health Cover surcharge will be outraged if the rules are changed as they enter their 60s, and they face risk rated premiums several hundred percent higher than they are currently paying.
However, there may be a way forward.
The Federal Government could:
- Allow premium differentiation on the basis of age, but not any other risk factor. This would mean people with poor health status unrelated to age would not be priced out of the market or refused cover.
- Redirect the premium rebate to subsidising a proportion of the marginal costs of people aged over (say) 60 who have held PHI for at least 15 of the last 25 years. This would go a fair way to honouring the commitment the government made when Lifetime Health Cover was introduced.
- Require all PHI hospital policies to include cover for all hospital treatments. This would prevent insurers from offering really cheap but worthless ‘junk’ policies, in a post- community rating world.
In such a world 30-year-olds would be able buy hospital policies covering the full range of hospital treatments for about the same cost as a current policy that offers nothing except the opportunity to cross-subsidise older people. If the insurance sector cannot sell such products, it doesn’t deserve to survive.
We need only look a couple of thousand kilometres to the east to find an example of a country with risk rated PHI accompanying a universal public health system.
In Aotearoa/New Zealand over a third of the population hold PHI, which has much the same scope as the Australian version as far as access to private hospital services is concerned.
A sustainable PHI model operating without government coercion is possible – getting there will be the hard part.
Readers may be interested in Croakey’s archive of stories on private health insurance, which also investigate whether it should have a role at all in our health system and the inherent inequities in giving some people better access to care.
