Published On: Fri, Apr 29th, 2011

NHIS’ One-Time Premium — An Actuarial Estimate

The concept of a one-time premium payment promised by the ruling government in its manifesto in the run-up to the 2008 elections has garnered a lot of press during the last two years.

There are two schools of thought: those who say it is impossible to derive and those who contend it is a matter of time before it is rolled out to the awaiting public.

Just recently, the MD of the NHIS, Mr. Mensah, came out that very soon his outfit will come up with an annual premium and a one-time premium payment options. And a few weeks ago the deputy minister of information, Mr. Samuel O. Ablakwa, went on the offensive against the former Minister for Health, Dr Kumbour on the latter’s stance that the one-time premium payment option is impossible.

This piece is mostly an excerpt from a presentation the writer gave on this topic to the Actuarial Society of Ghana in July of 2009. You may note that the writer went on record opposing the government-run NHIS in an article in September 2001 on Ghanaweb.

It is disheartening to observe that most of the reservations in the old article are being borne out when one hears of the Board of Directors (rent-seekers?) of the various district schemes, who have no stake in the schemes, are flexing their autonomous powers over the central government largesse to drive up the costs.
The concept of a one-time premium payment is usually presented with the following advantages:

a) Cheaper collection or premium payment administrative costs over the lifetime of a policyholder, and helps reduce the cost to the policyholders

b) Better persistency or avoidance of non-payment of premiums that helps the NHIS’ operation

c) Earned Interest spreads on the invested one-time premiums to help defray the cost of running the NHIS

d) A lifetime peace of mind for the policyholder in that regardless of what happens to medical costs, personal finances and health status in the future you are covered

The above factors are always considered in the actuarial pricing of one-time premium payment products, mainly life insurance and immediate or payout annuities. The other factors considered as disadvantages in the pricing and marketing of such one-time premium products are:

a) Only a few people have the money to make a one-time premium payment, thus the seller ends up spending a lot of resources, and in this case political capital also, to market to a small segment of the population.

b) Even if one has the money to make a one-time payment, one will most likely not do it, since the money will be lost in case of an early death, because there are no refunds of projected unused funds at death.

c) Also there is no surrender value in the policy once you purchase a one-time premium. That is, if after purchasing the one-time premium and later your financial circumstances change and you need the unused portion of the one-time premium to satisfy a more pressing financial need, sorry there are no refunds.

d) How to correctly price an ever changing covered sickness/diseases or Diagnostic Related Groups (DRG) and drugs (formulary). Growing up in the 1960s and 1970s, the formulary was predominantly: APC/Paracetamol, M & B, Chloroquin, Atwood and Resochin (spellings!!!). I bet the pharmacist today will have a hard time knowing what these are. So, how quickly will the NHIS change to accommodate the changing needs of those who have paid the one-time premium?

As an illustration, let us assume that this scheme was in existence in 1980 and someone had made a one-time payment in 1980. Then in 1988 the policyholder finds out that he or she is HIV positive; would the NHIS have covered him or her for the expensive HIV drugs in 1988? The afflicted policyholders would have demanded coverage. Note that HIV was unknown in 1980.

The issue would have more than likely ended up in the court system; and more than likely the plaintiffs would have won. This assumption is based on the fact that should some existing and covered (1980) diseases such as diabetes become curable and malaria is eradicated with Bill Gate’s money and tenacity just as he wiped out Lotus 123, WordPerfect and Netscape in short order, the NHIS would not give the policyholders refunds for charges in their premiums for the now curable diseases.

So, why should the NHIS be allowed to avoid paying for new diseases it did not know of, and price for?
e) Moral hazard comes in when a simple headache that would normally call for self-medication with Paracetamol/APC will now find the policyholder at the doctor’s office, thus increasing the cost of medical care, in the absence of adequate co-pays or deductibles for both annual and one-time premium.

The Estimate
“The work of science is to substitute facts for appearances, and demonstrations for impressions. “ ___ John Ruskin , the Stones of Venice 1853

To derive the premium for any health insurance product, using the principles of actuarial science, one needs data for the following:

a) Demographics of the covered group: age, gender and family size

b) The list of sicknesses and diseases(DRGs) and medicinal drugs(formulary) covered

c) The rates of utilization of the above services and severity of the diagnosis

d) Medical inflation rates which usually exceeds general inflation rates

e) Future changes in the mix of diseases and services and the impact of medical technology, and

f) In this particular case, since this a government-run scheme, the impact of the one-time premium policyholders becoming a vocal voting bloc with a lot of demands beyond what they originally paid for has to be priced in.

You may note that items (c) through (f) are not easy to predict accurately into the future. They even confound the best actuarial minds when the predictions are for just one year, much more predicting it for the lifetime of a person in this case.

For those interested, you may read the following article by the late Robert J. Myers; the actuarial godfather of the US social security system on how “bad” his initial costs estimates of the US Medicare system was in 1965, after 25 years in operation.
Since the NHIS data for pricing is not public information, let us use a fairly good proxy to estimate the one-time premium. This is along the lines of what a Chief Actuary will do to get a ballpark figure or a frame of reference before the detailed work is done.

Here are the underlying assumptions for a simplified high level illustration:

a) The likely policyholder is a 45 year-old married male with four dependents, a wife (40 years old) with three school age children.

b) The average combined life expectancy for him and his wife is assumed to be 30 years (years of coverage). The pricing assumes that the left over premiums from those who die before age 75 will be used to support those will live beyond age 75.

c) Most countries with good economic data spend between 6% (Mexico) and 11% (Switzerland) of their GDPs on healthcare costs. The average spent on healthcare by these countries is about 9% of their GDPs. The US is an outlier with 15+%.

d) Ghana can do it with 4.5% of its GDP; more than likely at the lower end of the estimates since the costs of imported modern drugs and medical devices far exceed the per capita income in Ghana relative to Mexico and Switzerland.

e) Salaries, wages and private business income in the country equal 90% of Ghana’s GDP. Therefore, the average 45 year old will spend approximately 5% (4.5% / 0.90) of his annual salary, income or wages on healthcare costs.

f) Salaries and healthcare costs will both rise at the same rate of 12% a year in the future.

g) The NHIS can invest the proceeds from the one-time premium at annual earned interest rate of 15% in the future.

h) The one-time premium will equal 105% of annual salary, income or wages. That is almost 13 months of income.

See the footnote for the applicable formula. ****

i) Please feel free to make your own assumptions and plug them into the formula to come up with your own one-time premium estimate.
Based on the above high-level illustration with a point-estimate, and assuming a national average monthly income of GHC 250 or an annual income of GHC 3,000, the average one-time premium will be GHC 3,150. Would the masses be able to afford it even if healthcare was a top priority for them? For the few relatively rich people who can afford it, is this something the NHIS should waste any marketing and political resources on?

This estimate is only a guide to start the debate along the lines that should the actual costs even be a half, GHC 1,575, or two times, GHC 6,300, how many people are able and willing to part with such amounts for healthcare? So, is the one-time premium possible? In the spirit of the political quote “Dzi wo fie asem” another quote from a different political era is apt in this circumstance, albeit this is a more conducive political atmosphere:

“Asem yi dzi ka, na hwana beka”, P. K. K. Quaidoo, 1960s
which roughly translates into “who is to tell the Emperor that he has no clothes on”?

Then again, did the ruling government’s manifesto specify that the one-time premium will be on a fair economic value basis? An educated guess will put the one-time premium on a fair economic value basis closer to GHC 6,300, with 5 million income-earning and premium paying ( both annual and one-time premium paying modal options) policyholders out of the roughly 24 million people. We are all aware that the NHIS scheme has been and continues to be heavily subsidized by the government.

We await the actual set of one-time premiums that will be marketed by the NHIS and also to read the signing actuary’s memorandum that will accompany the pricing. Hopefully, it will include a dynamic validation (back casting) of the premiums. This is when the model which is being used to project the future healthcare costs to derive the premiums is used to project backward to show the premiums that would have been generated if the model had been in use say 30 years ago.

The results are then compared to the actual costs over the last 30 years. This is to demonstrate the robustness of the model being used for the current pricing. May REASON guide and comfort the actuaries through the POLITICAL REALM of this one-time premium project.

There are other pertinent issues that affect the overall cost of the current NHIS scheme that need discussing besides the one-time premium.

Some Cost Issues
Anti-selection through Open Enrollment: The concept of insurance rests on the assumption that many will pay into the pool when they are healthy or not afflicted; and those unlucky ones who later on get afflicted will benefit from the pool of money. A financial danger to a healthcare system such as the one being operated by the NHIS is the open enrollment system.

That is, people can enroll in the system whenever they choose to. It may have been setup that way in the beginning as a marketing strategy or a loss-leader “Ye de fre edwa”. But after six years of operation it is tantamount to being able to buy car insurance to cover an accident that has already happened.

The system is currently allowing people to wait it out until they really need the health insurance before they purchase one. No insurance system can sustain itself under this mode of operation. The 25 to 40 year-old income earners who have fewer health issues now may not purchase this insurance until they begin to face health issues later in life.

The NHIS has to set a date and let people know that if you do not get in by then, you would have to go through underwriting when you later enroll and your premiums will reflect your health risks at the time of enrollment. Otherwise, the system will be hard pressed financially to go on like this for another five years. Or a smart entrepreneur will setup a private scheme and insure 25 to 40 year olds at far lower premiums than the NIHS can offer.

The policies will expire at age 41 and they can go ahead and purchase new policies at the NHIS when they are at the beginning of the chronic disease zone.

Fraud: Back in 2001, the article above implored the government to look for a claim/administrative system in the US. It looks like the NHIS is on the verge of purchasing a cheaper administrative and claims systems from a European country.

One reason for advocating for a US system is the ability to detect medical fraud. I seriously doubt that a country like Belgium or the Netherlands faces the kind of insurance fraud that is common in the boroughs of New York City, Chicago, Miami and the parishes of Louisiana. They keep the US insurance companies on their toes. The insurance companies in the US have written logic into their claims and administrative systems to quickly flush these out. And knowing how our governmental-run institutions perform with regard to fraud, the NHIS is highly encouraged to look into purchasing one from the US.

It may cost more upfront and take longer to train on because of in-built sophistication, but it will eventually pay for itself by stopping majority of the potential medical fraud which can easily amount to over 10% of the healthcare costs.

Finally, what is surprising is that for such a major structural change (one-time premium) to the NHIS scheme, it does not look like the executive branch needs parliamentary approval. A similar thing happened under the previous government when 2.5% of SSNIT proceeds were redirected to the NHIS. This will sooner or later present SSNIT with its own problems. You cannot take away 15% of the proceeds from any financial scheme and assume that everything will be fine with the scheme in the future. This method of implementing major changes to schemes without serious public debate at the parliament and APPROVAL needs to change.

In the midst of all of this, the hope is that the NHIS will pay more attention to the concerns of my uncle Kwasi Krappah of Kumawu. He had asked me the most probing questions two years ago about the NHIS scheme. His concerns were mainly: the publication of the covered DRGs and formularies (may be have all providers prominently display these lists); the ease of obtaining the scheme cards, improved services and facilities, and ability to use the scheme cards at all the providers around the country. He did not utter a word about the one-time premium in the 30 minute conversation.

Are there more political brownie points to be gained from addressing Uncle Kwasi Krappah’s concerns or from one-time premium sales by a social democrat government to the relatively rich?

“When we build, let us think that we build forever. Let it not be for present delight or for present use alone. Let it be such work as our descendants will thank us for; ——, and that men will say, as they look upon the labor and wrought substance of them, “See! This our fathers did for us.” __John Ruskin the Seven Lamps of Architecture 1849 Footnote: **** Formula: The mathematically savvy only: the required one-time premium will equal = 5% *1.15* 1 – (1.12 / 1.15) ^30 / (0.15 -0.12) = 1.049401 or 105% of one’s annual pay or 13 months of pay.

Read the rest here:
NHIS’ One-Time Premium — An Actuarial Estimate



Leave a comment

XHTML: You can use these html tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>