Musings of a Philadelphia Physician who has served the community for six decades

Return to Home

Related Topics

Health Savings Accounts, Regular, and Lifetime
We explain the distinction between Health Savings Accounts, Flexible Spending Accounts, and Lifetime Health Savings Accounts. Sometimes abbreviated as HSA, FSA, and L-HSA. Congress should make it easier to switch between them. All three are superior to "pay as you go", health insurance now in common use, only slightly modified by Obamacare. It's like term life insurance compared to whole-life. (

The Outlines of Lifetime-Funded Health Insurance

It is misleading to make precise predictions, about almost anything, eighty or ninety years in advance. However, predicting the average of millions of people is more accurate than predicting any individual future, whereas mathematical principles like compound interest are precise, forever. But let it be clear; what follows is rounded off, estimated, and largely based on projecting past experience into future performance. You must do so, if you want to talk about it, at all.

Investments are more predictable than health costs. At 10% they will double in seven years; at 7%, doubling investments takes ten years. Ten in seven; seven in ten. From birth to age 91, there will be time for thirteen doublings of investments. At seven percent, only nine doublings. With a focus on health economics, Americans divide into four groups: children from birth to 26, working people from 26 to 65, retirees over 65, and poor people of any age. We assume only people from 26-65 are able to deposit money in Health Savings Accounts; children and poor people are dependent on working people to help them, while retirees must live on money they earned while they were 26-65. Businesses and governments are pass-throughs which sign checks, but in our way of thinking, only individuals make and spend money in the national accounting of it. These are the assumptions, please read them twice.

1. Investment predictions are based on Ibbottson's compilation of actual market performance since 1926 of all investments in all classes. It's safe to assume index funds, now available in the trillions of dollars, will follow Ibbottson's patterns for the next hundred years, only because they were remarkably steady during the last century. Two major depressions and a dozen minor ones, one World War and a dozen smaller ones, were unable to shake the long-term trends for more than a blip in the lines.No such prediction can be guaranteed, of course. Highly diversified, Index funds have management fees of about a tenth of a percent, making them steady passive investments for people who have little investment experience, and probably equalling performances of most people who do. Using Ibbottson's raw data, half the population will do better and half will do worse--by managing their own money, even with professional advice. But if everybody buys the index fund without advice, everybody will perform the same. Collectively, the average common stocks of "small" American corporations (but nevertheless greater than one billion dollars of market value each) achieved a ninety-year performance of 12.2%, which we here discount to 10% by using diversified ETF (index funds) of really big stocks with familiar names.

2. At present, the only realistic source of deposits into a Health Savings Account is by individual investors within the age group 26-65, except for investment income. Contributions made on behalf of children derive from money earned by working parents or by -- somebody else aged 26-65. Retirees invest, but the core of what they invest was earned earlier, again 26-65. We ignore exceptional cases. The population 26-65 supports their own costs, and those of everybody else. Nevertheless, it is impossible to made precise predictions about the time and amount of shortfalls in individual Accounts when sudden withdrawals must be anticipated. For the most part, transfers are made from accounts in surplus to accounts in deficit, but particularly during the phase-in period, supplements may be necessary. However, everybody's Health Savings Account is a separate piece of property and not a pool. This difficulty is managed by slightly overfunding everything to keep transfers to a minimum, and pooling these surplus amounts by agreement to reimburse them at some later time for some specific purpose. Furthermore, the principle is announced in advance that if shortfalls are unavoidable, the accounts to be billed for it are to be limited to the working-age group from 26 to 65. The ultimate fallback is the full faith and credit of the taxing power of the U.S. Government; but we hope to avoid using it except in dire emergencies like a national nuclear attack or something else of this order of severity, eventually establishing a reputation of a self-funding program. Within that program, the real fallback is to the 26-65 generation who are earning a living. They are expected to care for their children, and aging parents, but by individual agreement. Since the plan is to stop collecting 6.5% payroll deductions from this age group, and anticipated deficits are of the nature of 0.5% of income, assessments should be comfortably met, although it is too much to expect them to be cheerfully met. A whole chapter is later devoted to this sensitive topic.

2.5 Transfers are necessary, however. Because of the security risk, it is probably wise to introduce the extra step of transfer into and out of an insurance or insurance-like pool, so that transfers between Health Savings Accounts can be performed by a tightly controlled security organization which maintains permanent transaction records as its main or only function. Pooling would actually ease the accounting burden of linking every account in surplus with every account which runs a temporary deficit, when actually it is only necessary to account for the balances between individual accounts and the pool. If newborns have individual accounts, they will have to be linked to their parents or guardians, and perhaps transferred from their parents accounts at age 26. Although making health insurance a personal rather than a community activity is a step forward, there will be much occasion for reducing individual volatility while the accounts are still too small to provide their own liquidity reserves. This is also the place to put subsidies for the poor, and payroll tax assessments on the 26-65 age group, replacing the 6.5% payroll tax for Medicare pre-payment, which has been eliminated out of consideration for dropping later Medicare coverage. After the transition phase is complete, the pool will be less necessary, but it may take decades before money spent on obstetrics comfortably matches up with money pre-funded for cancers and strokes.

3. Medical costs essentially do not matter for lifetime plan design, since this is "found money". Rising costs are of course the main concern, and of course we should pay all of them, but not necessarily by investment income, entirely. We strive to generate as much new revenue as possible, and are confident it will raise appreciably more than the present system. If more is needed, additional sources must be sought. It might, if all went as planned, generate half of the cost of healthcare in the far future. But it will never seem like that much, because we are already outspending our revenue, and borrowing the shortfall. Only after our books balance on a current basis, will the public notice any difference. Congress will notice it sooner, and be tempted to spend it. If it generates more that we need for healthcare, then if we are wise we really should spend the surplus on retirement costs for an aging population.

However, the outlines of what is possible can be made out. Likely, future medical costs for younger individuals under the age of 65 should remain constant, or even decline in the future. However, medical costs of the elderly are assumed to rise in the future, as people live longer and get more expensive chronic diseases. CMS says 5% of the elderly generate 50% of costs for their age group. Conditions related to obesity are a new source of such costs, while Managed Care has had no effect. Exceptions will appear, but predicted cost curves seem likely to assume the shape of a dumb bell, bulging at the ends, but shrinking in the middle. Since working cash for inter-person transfers and unexpected illness are laid on the working age group, it is a lucky happenstance that future predictions almost always show a dumbbell or wasp-waist shape to the cost curve, making it possible to design budget shortfall levies to concentrate on this level. The biggest threat to future healthcare financing may well lie in the likelihood that people who now die at the age of 60 will live to be 85, and be afflicted with the same high expenses as we now see in people aged 85. If present trends continue, the rising costs after age 85 contain a mixture of falling sickness costs, hidden within rising domiciliary costs, or nursing home costs, which possibly belong in a different budget. This outcome seems more likely that the present rate of longevity extension, which is more likely to level off. However, the original point is the strongest on: it is a mistake to pretend to predict a future which cannot be predicted.

4. We assume average health costs for a lifetime to be $300,000, based on a Blue Cross of Michigan study, confirmed by AHRQ and CMS to be of that general magnitude. It is a critical number, since it is the burden workers age 26-65 must carry for the whole medical system at every age--averaging $7800 per year apiece for the working person. It is important to know how it is calculated, to understand what it means -- and what it doesn't.

Calculated as described, the $7800 pays for one working person, plus averaged contributions for dependents and charity obligations. Because of cost-shifting, the proportion of redistribution is unclear. But, remember, these are lifetime costs, using current prices. If costs remain otherwise identical, a 3% inflation rate means the answer, calculated the same way next year, will be $309,000. This point is made to convince the reader, that even if we do not know the precise costs, we can be fairly sure that costs will soon outrun our ability to cope with them.

In order to include present costs and present practices, a hypothetical person is constructed from current costs at each level, reassembled in order to reflect current costs for current treatments, as if they all occurred in the year of death. It therefore includes 3% inflation over the time span from birth to each particular age. The modest costs of childhood are thus inflated the most, while the expensive last year of life is not inflated at all. Since it will be adjusted in the next paragraph, it is probably not a serious error.

Predicted Future Healthcare Costs. If the $300,000 we spent on each person's health in the last 90 years should merely increase at 3% inflation, lifetime costs will become $4.5 million, 91 years from now. That's sobering enough. But if medical costs increase as much as they actually did in the past century, lifetime costs will come to an unthinkable $1.5 billion dollars a person. Therefore, we accept the present hypothetical lifetime cost, including inflation, to have been $300,000, and assuming no change other than 3% inflation, will be $4.5 million, 91 years from now.

Nevertheless, we know in retrospect that a solitary deposit of $55 in 1926 at the 10% rates which actually prevailed last century, would have kept even with it without later additions.Today, to keep up with the costs of a newborn great grandchild could apparently be accomplished with a deposit of $796, over twelve times as much. It's still an achievable goal, but it's drawing away from us. Remember, $800 will only pay for present prices, plus 3% inflation. Unlike the last century, the next century cannot add thirty years to life expectancy, or eliminate thirty diseases. In fact, only five remaining diseases account for half the cost, and life expectancy has no room left to increase by another thirty year extension. The medical profession has the scientific tools to make it work, provided the financial and political professions create the right environment. The present prospects are for science to deflate disease costs in every age group except the oldest, but to quantify is impossible. Since 1913 when the Federal Reserve was founded, a dollar then is worth a penny, now. The medical profession can't help with inflation. Perhaps no one can, but at least a monitor exists to make mid-course corrections of the currency.

More than that, notice the difference between $300,000 and $796. The difference, although roughly estimated, suggests the savings possible by switching to lifetime costs, and investing the difference between "whole life" and "pay as you go" annual payments. It is unnecessary to come even close to actual costs to see the savings from financing the medical system longitudinally, outstripping anything imaginable in extra administrative costs, or price escalation from moral hazard. Cut it in half, or take only a tenth, the savings are so appreciable you begin to wonder if they might upset the stock market. There are even safeguards from miscalculation, remaining inherent in other approaches to cutting the cost of medical care.

For example, we fervently hope, but make no assumption in the example, for an extension of working life, both down and up, to 24-75. That is, we favor a reduction of the two great vacation periods in American life, by a parallel extension of the lifetime of significant work. We recognize most Americans do not agree, and in a democracy that's how decisions are made. But this safety valve remains available to those with bad luck or bad timing; it's how you recover your finances if they have slipped along the way.

Let's cut wasteful practices, particularly the habit our government has of making hospitals into welfare programs, or our insurance administrators have for making them into banana republics, and the habit the public has of wanting everything for free. Let's structure costs in such a way that if an individual doesn't overspend for healthcare, the money saved gets applied to better retirement. It gives the individual some skin in the game, which is the essence of bringing costs down by competition.

Right now, however, it is necessary to examine how we might extract the savings from Health Savings Accounts, gradually transitioning from one-year term to whole life with investment, without upsetting the system. And examining what useful things might be done with a cash windfall, before too many extra noses push into the trough. After all, you cannot spend the money after you are dead.

Let's start backwards from an assumed guess of $300,000 average lifetime expense, from the viewpoint of someone aged 90, which is also only a guessed-at future longevity, to the day of death. To have $300,000 at age 90, you must have $30-40,000 set aside at age 65 in index funds. Remember, in the elderly we are talking about the period of greatest health costs by present projections, in an age group where few people are working and thus must entirely depend on investments and pensions. It can be done but it's a stretch. In many ways, the greatest obstacle would be the mind-set of elderly people themselves. We are talking about buying common stock for elderly people, who must overcome the main reason they buy high and sell low. Left to themselves, they will lean toward the safety of low-yielding bonds.

We have repeatedly alluded to The Monitoring System, which will take time and experience to design. Whether the monitor resides within the Treasury, the Department of Health, or some independent agency is a political question that others probably feel they have a better right to decide. Such an agency might have many functions, but since it must have the power to make myriads of mid-course adjustments, it probably requires a self-balancing oversight board like that designed for the Federal Reserve, and we favor that approach. At least once a year, that monitoring body will have to recalculate the estimates of the emerging trend of the balances between costs and revenues, and the distribution of the balances among each yearly cohort from birth to age 91.

Those yearly recalculations would set the price of entry into the system for latecomers, calculating what it would cost to make up for failing to pay for it all at birth. And if the system makes a revision for new information about trends in motion, everybody will in a sense be a newcomer, subject to a late deposit levy. And since the working adults 26-65 will be picking up the extra costs for birth to 26, plus charity cases at all ages, there will have to be secondary and tertiary adjustments in the levy. Furthermore, there may be recalculation of the cost of a particular age cohort for current medical expenses, and that will have to be set as an additional deposit requirement for that age cohort. Meanwhile, the investment managers will report on how close they came to their target, and further adjustments made. The Federal Reserve will make a report on current inflation rates, leading to more adjustments. The ultimate goal is to set a price for late entry at each year, so that continuing future income distributions will be equal, for current entrants, as for those who made a lump-sum investment at birth. This monitoring system will also be responsible for smoothing out short-term volatility, as in an influenza epidemic, and possibly long-term readjustments of internal lending and borrowing which were not anticipated at the outset of the program.

The Elderly Investor. Although the Libertarian view is that people ought to be able to do what they please with their own money, this is one case where it probably would be advisable to mandate the pooling of investments, in spite of the obvious introduction of political risk. The argument runs: it might be possible for most people to save $30-40,000 by any number of ways before the age of 65. But after 65 it becomes a little unwise for a growth fund to place trust in the investment judgment of a class of people who rightly prize security over growth. They will predictably have a very hard time shaking the perceptions of their age group. On the other hand, if there is ever a chance people will accumulate $45,000 in savings, it would be at the time they stop working. Let's hurry on; our present purpose is to illustrate the principle we are driving at.

Working people 26-65. Between the time they get their first job and the time they retire, working people have children, send them to college, buy a house, and try to come up in life a little. They get dozens of claims for their support, so in our example, to have perhaps $35,000 to surrender at age 65, using our system they might alternatively have to have $500 available at age 27, from Santa Claus. And then let it grow, untouched, to the next goal of $35,000 when they reach 65. That sounds easy, but it often has its problems. If somebody, say their parents, gives them the $500 as a present, it's all pretty easy. But if they have to work for it, then somebody has to give them $35 at birth, because the daisy chain is connected from start to finish. That's right, $35 turns all the way into $300,000 at age 90 if each step is coordinated. It pays for an entire lifetime of health costs. But it doesn't need to. If just about everything goes wrong, a quarter of that would still amount to a big chunk of money. Are you going to tell me no one could afford to give $7 to a newborn? There's no rule against making a partial contribution to your own care. There are practical problems to be addressed, but the power of compound interest isn't one of them. In fact, you might easily find that no investment house would accept a $7 deposit for a 90-year forward account.

Children. After the elderly, the second subsidized group is composed of children, including obstetrics and pregnancy. There is overlap here between child and two parents, and for conceptual purposes there is nothing to do but be arbitrary. The addition of 26 years of compounding is too tempting to quibble about ambiguities, which might be solved by giving it to any of them, or to all three. That heightens the unfairness to those who do not have children, but it also creates an incentive for the mother to have her first child younger. Medically, that would be a desirable thing. Our society, perhaps even our biology, has created a tradition that the parents subsidize the health costs of children. The Health Savings Account system formalizes that tradition, or at least does not conflict with it. For the surprisingly small amount of one single payment of $150 at birth, the child would have $40,000 at age 65, assuming a 10% investment return. Investment advisors might rebel at their costs for accepting amounts that small, but a single-payment zero-coupon bond or credit might be created. That would ease the mechanics, as well as reduce the outcry against federal subsidy to people who might be indigent when the child is born, but are far from it later in life. The disadvantage is a bond makes no provision for the health care of children even though it pays for it, so some patchwork is still needed. A birth deposit of $150 would be worth about $2000 at age 26, and average childhood medical costs might be somewhat greater, so a transfer of ownership could imply a net liability.

The Poor. The third and last category of subsidized people consists of those who are both poor and sick at the same time. Unfortunately we have tried and rejected two methods of dealing with their problem. The first was defined by the original Good Samaritan: "Take care of him, and I will repay thee." And the second method was almshouses, now a relic of the past. The disadvantages of both approaches are now obvious. The third method was to eliminate poverty, which has worked pretty well. Fifty years ago, sixty percent of the hospital beds in Philadelphia were "ward" beds. Nowadays, there are few enough of them to scatter among paying patients. But the disadvantage soon appeared that the public became determined to prevent the inevitable rationing from spilling over to more fortunate components of society, in an era when hospitals are fearful of discrimination. Mindful of the long history of charity for the sick poor, and the spotty history of using government to cover the costs, we propose that governmental charity be paid out of the pool for interaccount transfers. That preserves an independent audit of just how much is paid by whom, and it is linked to an assessment process on people who must pay the bill. That will not prevent government from discounting its contribution, as it does not prevent Medicaid from discounting hospital bills. But it widens the audience who are instantly aware of it, all of whom will be heard from in the November elections. Individuals are compassionate, governments only pretend to be. You would almost have to say it is the one remaining good feature of having a King -- to symbolize the nation's simultaneous aspirations, of opulence and compassion.

Since America has rejected the obvious approaches to caring for the sick poor (almshouses and blank checks), our institutions are in some disarray. We even seem to be rejecting a mixture of the two, which was the hospital reaction to the 1965 entitlements. Until we identify and concentrate the sick poor in some way, we cannot even measure the size of the problem. But at least concentrating the rest of the population's sickness on paper allows us to measure their cost and (by subtraction) estimate a health budget for the sick poor. It will inevitably cost more than average, and result in worse outcomes. But only after we measure it, can we even decide how much we can afford.

What you have, including the three demographic subsidies, is what it seems to cost working people in today's environment to care for themselves and their obligations. It's distributed over forty years of working, but not everybody works that whole period of time. If you wish, you can contribute $100 a year from age 25-65, a surprisingly small amount which after compounding at 10% should pay the lifetime costs of one person (yourself). Calling it $150 to be safe, it is no more than a tenth of what most people suppose they pay for annual health insurance. Therefore, it is safe to suppose a family of four could afford to pay for ten poor people (in addition to themselves) at the cost they are already spending. Remember please, our goal is not to pay for all health care to the last penny. Our goal is to devise ways to pay for as big a chunk of it as we can.

And by the way, devising some method to get the latent money out of these accounts for medical care, since $300,000 does no good in a frozen account of somebody aged ninety. Please read on.


Please Let Us Know What You Think

(HTML tags provide better formatting)

Because of robot spam we ask you to confirm your comment: we will send you an email containing a link to click. We apologize for this inconvenience but this ensures the quality of the comments. (Your email will not be displayed.)
Thank you.