Philadelphia Reflections

Basic Coverage: Three Big Problems, No Little Ones

Columbus and the Egg

Let's summarize. We started with the classical Health Saving Account (C-HSA), which may need a little updating, but appeals to millions of frugal people as a simple way to avoid the tangles of present-day healthcare financing. The law hasn't changed much, but used by millions of subscribers has turned up many surprising features, all good ones. Deliberately overfunding them has unexpectedly useful results, for example, in providing retirement income if you have been lucky with your health.

On that foundation then was devised New Health Savings Accounts (N-HSA), combining six more innovations, each of which is easy to explain, but in combination utterly transform the basic design. The extended longevity of the 21st Century makes compound interest in passive investing into a powerful investment tool and is used to reduce healthcare costs to the consumer, markedly. Secondly, improved healthcare for the working years has unbalanced the employer-based model, so sickness costs are getting crowded into retirement years. For this, the accounts permit extraction of the first year and last years of life, transferring their heavy costs to the working generation where employment-basing still makes sense. And so on. With very little new legislation, most of this package is ready to go.

Lifetime Health Savings Accounts (L-HSA), patterned on a whole-life model. L-HSA won't work without some new legislation to edge around recent regulations and some outmoded premises. Multi-year coverage is cheaper but requires a longer commitment, so it needs to be precisely designed. Starting fresh, it directly addresses a host of problems hiding behind a century of habit. Its flexibility accepts a range of designs, stretching from self-insurance out of a bank's safe deposit box, stretching all the way to letting life insurance companies run everything.

We conclude insurance of every sort "shares the risk" for expensive problems but generates extra expense when little problems get insured that don't need to be. Little medical problems swamp the fixed overhead of insurance unnecessarily. When you only insure essentials -- as true catastrophic insurance does -- it costs far less than insuring everything. So, here's an outline of a major variation, adding features, one by one. It's a Health Savings Account, on steroids.

Proposal 25A: Let's combine the high cost of the first year of life with the really high-cost last year of life, as a basic foundation of minimum health insurance.

The dual combination could surely include everyone. What I am technically trying to achieve, I admit it, is to combine one life situation, which sometimes generates a surplus it can't use, with another situation, where every baby creates a difficult debt for someone else to pay. And whereas 100% of the population experiences birth and life at the two ends of life, there is ancient uneasiness about sharing liabilities outside of families, and even, how far the boundaries of families will stretch. Ancient fear of violating obsolete family boundaries sometimes hinders useful proposals. Because of increased longevity, grandparents are now real people, not just a picture on the wall. And by no means are they all senile.

A sly feature of all this is, people still alive are willing enough to overfund the costs of the terminal care lying ahead of them, but few still alive have their own birth costs on their minds, even if they were never paid. That lopsided initial generation might generate sizeable reserves for a circular program, if we imaginatively link accounting between generations, carrying those birth costs forward. Another unrecognized feature is the costs of these first and last years of every life are in fact both paid in retrospect, and often not by the patient.

Furthermore, most of the heavy expenses of both ends of life are created within a hospital, which often delays final billing until the issue of responsibility is settled, creating blind spots in which final responsibility is unclear. Nevertheless, the hospital aggregates many services in a total hospital bill, so bulk payments by diagnosis or even by age, are tempting but as yet crudely perfected. When all these matters finally get standardized, reimbursement from one insurance entity to another should become commonplace. The final transaction in both end-of-life insurance as well as the beginning of- life insurance easily and more naturally evolves into who reimburses some other entity who (temporarily) paid the bills.That's about all, there is to say about the funds' flow, which should become very simple, but come in larger lumps. At such esoteric levels, no one cares whose money it is, so long as it gets paid. At the local patient level, family transfer issues can be troublesome. It can be recalled we have gone into detail about grandparent/grandchild transfers in the section on New Health Savings Accounts, and indeed it emerges as the main innovation of that effort. Still earlier, we described the compound interest hidden in the background, paying for a great deal of it. As an aside, most people will eventually find it is desirable to use overfunded accounts as a basis for tax-sheltered retirement costs, and will therefore often die with a small surplus, which is the basis for closing the loop between generations. That may take time to evolve.

Proposal 25B: And then, add catastrophic coverage regardless of age, less the cost of overlaps. Title: Tri-Challenge Basic Coverage.

Here's the universal catastrophic coverage we promised in the first chapter, minus the first and last years of life overlaps.

Overlaps. One passing word about overlaps. Reducing overlaps is one of the main mechanisms for reducing costs, but the insurance entities will fight fiercely over who gets the reductions. For example, what about a six months old child who dies with an expensive hospital bill? My suggestion is that the beginning of life insurance pays first, the end of life insurer pays second, and then the catastrophic insurer pays for the balance. This gives the savings to the catastrophic carrier, but most of the cost is given to grandpa, who is dead, and anyway, most of it is investment income, leading to fewer complaints.

We thus design the basic coverage to include two universally-unavoidable costs, plus the universally-inescapable risk of unpayable health costs at all ages in-between. It is clearly superior to less universal, and more unaffordable, coverages.

Proposal 25C: Require the unused birth coverage to be transferred, either at the time of death or optionally at other times, to either a designated grandchild by inheritance or to a designated pool of unassigned third-generation recipients. The grandchild's Health Savings Account would begin at birth, capturing 21 extra years of compound interest, compared with employer-based insurance.

The purpose of beginning an HSA at birth is to add 21 years to the compounding while staying within the laws of perpetuities.

A 1:1 ratio of national grandparents to national grandchildren do exist, but so do multi-children families, no-children families, unmarried families, divorces, etc. There is definitely a need for a pool within which, mis-matched funds are administered.

The childhood transfer feature adds about $28,000 to the coverage, but it would only require $42 a year (added to the last year of life premiums, at 6.5% interest compounded from one year of age). Because catch-up transition at age 40 would require $200 annual contribution to catch up with newborns, rising to $28,000 for someone aged 65, it might be better to add $25, or so, at birth to reduce the even greater cost of late joiners (over age 40) to the plan. They must be enticed, however, because they are the ones closest to activating the transfers, and hence are the most important support to enlist to an innovation.

Proposal 25D: This is voluntary Tri-Challenge Basic Coverage. Whether to make it mandatory, whether to subsidize it for the poor, and whether to replace Obamacare with it -- are political decisions, not questions of insurance design.

This coverage probably approaches half of the entire health costs of the nation, depending on how you treat the cost of prison inmates, the unemployable and illegal immigrants. If science should ever succeed in eliminating every major disease and therefore every other cost of healthcare, 18% of present costs would very likely persist. And in fact, the cost of prisoners, mentally retarded and illegals show no signs of changing much, either.

They are costs most likely to be permanent, but what we spend on the rest will depend on where we place the limits on "Catastrophic" care insurance. That cost depends on the size of the deductible, and the upper limit of coverage. You can readily predict the debate: the higher the deductible, the lower the premium, so the out of pocket cost depends on the deductible, too. But while controversy would remain, the great beauty of this design is the lessened political resistance from every voter who would likely benefit, which is 100%.

There is no escaping these realities, and in the meantime, the rest of health costs will dwindle down to the cost of birth and death, which change their nature very slowly. Therefore, although it is not traditional in the health insurance field, I propose it has long seemed an entirely natural thing, for the costs of childbirth to be an obligation of some other generation, usually but not always of the same family.

The problem is indeed complicated by the unusual concentration of malpractice claims in obstetrics, as well as the marginal finances of parents at this stage of their careers. But the hidden effect is to create more, or less, valuable babies, depending on how you judge the impact of emotions versus supply and demand. It was almost an unknown issue a century ago.

Any resistance to this proposal is thus likely to be more instinctive than rational since we have had so many generations of channeling such issues within the family unit. The relatively trivial cost of funding children's health care through 104 years of compound interest does generate a temptation to overfund the program, which if necessary can be frustrated by requiring one or more zero balances per lifetime. That is particularly true in this particular instance when compound interest could generate almost any amount of leveraged money at the death of a grandparent, simply by increasing the modest amount impounded at the grandparent's birth, and waiting long enough for it to grow. People in charge of managing the currency are then drawn into the discussion. Indeed, the relatively confiscatory level of estate taxes (60-70%) is a sign our society is not comfortable with perpetuities, although available remedies are fairly simple. Indeed some states also levy inheritance taxes on the recipients as well as the donor's estate taxes.

Whatever the traditional resistance, it's permanently true that the costs of the first year of life will always greatly exceed what a newborn is able to pay. And therefore it surely follows that this obstacle has hindered health financing for a very long time. There must be some mechanism for inter-generational transfer of funds, or life cannot continue. At present, it's called a family, and families are under strain. Conditions of modern life have evolved to the point where interference with some generational transfers will cause more suffering than the relaxation of such attitudes. There will be resistance, but it must be persuaded to reconsider its position and bilateral compromise must result.

Warning: Start Saving While You are Young. Health Costs for the first year of life are reportedly 3% of the total, and while the last year of life (15%) is worse, those costs laid on young families are particularly disruptive because of still higher costs later if neglected. At best they compete with college, housing, and automobile costs, and probably reduce the birth rate of the middle class. The enclosed graph shows a family of curves based on subsequent investment income from different invested external-contribution levels at birth, to help readers judge how much surplus might likely be generated at the end of a lifetime of average costs. Our goal is to estimate the cost of overfunding grandpa's health costs at birth, so there would remain an incentive for him to spend wisely, but still generate enough surplus to fund a grandchild's juvenile health costs. That is, to estimate the cost of transferring the obligation of grandchild costs from parents to grandparents. But it must ultimately be recognized that the full consequences of such a basic change, are unpredictable.

First, however, we must determine the size of grandchildren's costs. 3% of all costs ($10,500) for the first year of the child's life sounds plausible. But 5% of costs from the first birthday cake to the 21st birthday ($17,200) sounds surprisingly high and needs to be challenged if only to defend it properly from rapidly escalating educational costs. That's one of the great advantages of starting with a demonstration project; you can see where the bills are coming from. But that's mostly a quibble, because even $17,000 is manageable, and the legal boundary of age 21 is strongly defended.

Let's take just a moment to examine the laws of perpetuities, which mostly focus on intergenerational inheritance. Established about three hundred years ago as common law, they permit transfers for 21 years after the birth of the last person to be alive at the time of the bequest. I'm not a lawyer, but those do not seem like a handicap for what is proposed.

The pool would also pay for multiple children in a family or those newborns who do not have a willing grandparent. I never met any of my four grandparents, so I don't know if they would have been willing or not. As the father of fourteen living children, one of my two grandfathers would surely have wanted some adjustments. If it becomes an issue, the government could easily afford to donate the $7 yearly required to avoid the issue. During the early transition phase, it might possibly be necessary to have government backup if temporary mismatches appear, eventually repaid by adjusting the initial cash deposits. It happens the birthrate is 2.1 per mother, which easily matches one-half per grandparent, greatly relieving but not eliminating the occasional mismatches. It would seem a fairly simple task to charge 1:1 (grandparent to child) for the first three or four years, and gradually re-adjust if more precise data becomes available. It could even be possible to pay for the child-generation completely, but this is not entirely desirable. It should not require dynamic scoring to understand that making healthcare free would increase its cost.

Not everything which is desirable to do is desirable to do in a big hurry.

The political advantages are wide-spread. Much tax resistance to paying for healthcare would be deflated by knowing that almost all cost was absolutely essential, and for the most part universal. The parent generation would be relieved of the cost of the healthcare of children. During transitions, there probably will be problems with overlapping coverage. Not only is Medicare reducible for overlapping costs, but the U.S. government is relieved of some of the embarrassment of borrowing 50% of Medicare from foreign countries in order to pay for it. But, sorry, we're getting down into the weeds.

If desired, the 1.6% salary withholding for Medicare can be reduced, relieving working people and their employers of about one-quarter of Medicare costs. The tax inequity between employees and self-employed should be eliminated anyway, but this somewhat reduces the disparity. The doughnut hole was a good idea, but it is part of copayments which are a bad idea. It should be self-evident that making hospital insurance free but doctor payment subject to Part B premium, creates unmanageable distortions. Many healthcare financing problems are like the fable of Columbus and the egg -- once explained, anybody can do it. Nevertheless, it would seem much better to proceed slowly according to a defined plan, using demonstration projects and experimental trials, mid-course adjustments, and careful monitoring. Because not everything which is desirable to do is desirable to do in a big hurry.

Finally, someone does need to calculate the cost of adding catastrophic stop-loss insurance to birth-and-death insurance. It isn't possible for an insurance outsider to calculate the overlaps between the two types of insurance, which are probably considerable. But, particularly if there is a lot of overlap making it relatively cheap, that combination would constitute the kind of basic insurance which covers what everyone needs, and very few would be able to fabricate. If you linked it to a more sensible diagnosis related code, as a basis for DRG for inpatients, and firm association with market-based outpatient costs, you might get a firm basic package. It then only requires a relative value index for items which do not overlap, and a firm rule that the same items be charged the same amount, for helpless inpatients and not-so-helpless outpatients.

What are vital but uncovered by this basic package are new scientific discoveries, so self-evidently essential they create temptations to exact extortionate prices. I'd say it would be shooting yourself in the foot to go hard on such discoveries for their first few years. Overall, that which is left uncovered by this hybrid tri-insurance may be hated for its commercial motives, but nevertheless remains something we clearly want to encourage. Managing costs of that sort ought to be left to the Food and Drug Administration, the Patent Office and competition. And silently endured by insurance.

Originally published: Tuesday, March 10, 2015; most-recently modified: Sunday, July 21, 2019