Philadelphia Reflections

The musings of a physician who has served the community for over six decades

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. (

Admonitions: Using the Transition to Lifetime Health Insurance as an Inflation Restraint

Inflation Protection. Let's imagine the typical individual has reached the point where he is writing his will at the age of 90. He has followed our advice, created an HSA, dutifully funded it, and reached the point where his medical expenses are mostly behind him, but -- he has a meaningful amount of money left in the HSA. The existing legislation is pretty relaxed about that, allowing him to convert his HSA into an IRA, and follow its rules for inheritance. That's fine because it has created an incentive all these years, to save just a little extra money in the HSA for contingencies; after paying taxes, he can spend it as he pleases.

The Escrow Fund. It may be fine, but it eventually comes to an end in the face of terminal care costs; at that point, the future is damned. Since most people never know for certain which episode will be the last, indifference to insured costs is fairly general. There needs to be additional restraint against medical cost inflation. We propose that a compartment of Medical Savings Accounts be designated as a single-purpose escrow fund, adhering to the model of buying a life membership in a club, except in this case, it can only buy lifetime health coverage from Medicare. Annually, his fund manager transfers a sum to the individual escrow fund, calculated to reach a buyout price for Medicare coverage at some later age, and assuming the investment income achieves a stated goal. The individual may borrow against the escrow to pay current medical expenses and may need a subsidy to do so, but the escrow may not be spent down. (It continues to generate investment return to the fund which in normal circumstances would exceed the loan interest). At any time he has enough money, our Medicare subscriber can make a voluntary deal with Medicare as follows: If he will turn his escrow fund over to Medicare at his death, then Medicare will no longer collect his full Medicare premiums, starting today. That's a good offer or a bad one, depending on his life expectancy and how much is in the escrow fund; as of today's rules that would typically be several thousand dollars. That's a bad deal for the government if there is inflation. However, for individuals at any age down to age 26, it would seemingly always have made a better deal if he had only made it a year or two earlier. If it had been offered at age 65, it would have made a tremendously better deal than at age 90, because so many more premiums would lie ahead. And before that, if the deal were offered to a working person it could extend to skipping the 6% Medicare payroll tax deductions, which could be a stupendous deal. So let's go back and make a counter-offer using this inflation restraint. It's called the accordion plan, where both the government and the individual must agree on the best year to clinch it, depending on how everything is going. Unfortunately, any system like this requires an unimpeachable monitor.

Buying Out of Medicare. The average person over age 65 is haunted by the possibility that his living expenses will someday exceed his income, so he likes to have as much of his anticipated expenses pre-paid as possible. (He likes to be offered life membership in a club, for example.) So, he proposes to Medicare that they do the arithmetic and tells him at what age they think he could stop paying payroll taxes and/or Medicare premiums to Medicare and pay them into his escrow fund so that he becomes paid-up and no longer cares about medical cost inflation. And if there is no point in time when the two are clearly equal, then how much would he have to supplement the escrow fund from his savings to reach the goal. Since the law of large numbers enables Medicare to predict its average costs with greater precision than the individual can predict his own, a difference between the two prices represents the individual's fear that he might incur substantially higher than average costs. Half of the Medicare beneficiaries will, and half won't, but any individual's actual chances are largely a lottery. So, a substantial number of people would take the deal on terms favorable to Medicare. This isn't exactly the proposal we plan to make, but it illustrates the principle.

Part of the secret of the current proposal is that the individual can have the advantage (and bear the risks) of investing in the equity of private companies, whereas we would squirm if the government owned a big chunk of American private business. Notice for example how quickly the government sold back the stock of General Motors after it had bailed it out. Private individuals might indeed make 10% return on index funds of the entire U.S. stock market, given a 90-year horizon, (and under the discipline that you can't buy high and sell low, because we won't let you sell other than for medical costs). Furthermore, the government can't sell it for you, because you own it independently. This deal would fall apart if inflation unbalanced it, but the value of stocks and the cost of medical care will respond to inflation at about the same rate, providing you wait long enough and use big enough numbers. Nevertheless, it would only seem prudent to appoint an independent agency to monitor and control matters, particularly because stockbrokers are not considered to be fiduciaries, putting the client's interest ahead of their own. In spite of that fact, most people would be astonished at how fast a fund will grow at 10%. Medicare can't get such investment income, because in 1965 it was decided to use the "pay as you go" system, but it is clear that Medicare would sustain much higher debts if we abruptly cut off its payroll tax and premium income. So this process should require each individual to take at least ten years to switch completely, holding each person's "paid up" goal as ransom if he participates in reckless medical spending, and delaying the government's acquiring the escrow if they permit such spending. We are apparently never going back to using gold bars to frustrate government-endorsed inflation of the currency, so we have to devise other self-balancing restraints like this one.

In these days of burdensome health insurance costs, it is useful to consider how health insurance might emulate what is normally done with "whole-life" life insurance. Most life insurance clients to understand that total premiums amount to less than the face value of the policy, while considerably more than the face value is often paid out to the beneficiary. The apparently miraculous appearance of extra money is accounted for, by the ability of the insurance company to invest the premiums until they are needed for benefits. True, life insurance has also prospered from the stretching of longevity by improved healthcare, but that windfall also applies to health insurance. In both cases, improved longevity is hoped-for but not guaranteed, and adds to the safety reserves. The issues to be pondered are how to set final rates so far in advance. Or, if you wait to see how costs actually develop, how to give a useful benefit to someone who by that time is already dead. The life insurance companies have devised their way of managing this awkwardness, which requires public trust in the good faith of their counterparty. Results vary between companies, but in the long run, it doesn't pay to cheat.

The Need for Transparency and the Image of Fair and Square. Transition from an old system to a new one is a familiar problem for legislators. In our case, it may actually facilitate matters by restraining the impulse to take on too much difficulty, all at once. Every citizen is covered for hospital costs in Medicare Part A, and the great majority are covered for physician and outpatient costs by Medicare Part B. Part C is only partial and voluntary, Part D is still on trial. For this discussion, we need not describe the varying ways that Medicare Part A, B, C, and D collect premiums or the historical reasons why they differ. It should be emphasized early, however, that overall direct income falls short of covering overall Medicare benefit costs by 50%, so Medicare is 50% subsidized, and therefore not nearly as stable as the public assumes. It would help a lot, for example, if debt just stops being called an asset on the balance sheets. Everybody enjoys getting a dollar for fifty cents, so the program is more popular than it would be if euphemistic revenue descriptions were discontinued. It is particularly worrisome, that the popular alternative of a "single-payer system" implies simply extending Medicare to persons of all ages. But it also implies extending the 50% hidden tax subsidy to all ages so single-payer consolidation would add an even more unsustainable burden to the national deficit. This apparently irrelevant comment helps explain why the public expected Obamacare to be cheaper than it proved to be, and adds considerably to the urgency to find other revenue sources during a protracted economic recession, for what are proving to be unexpectedly high prices. Hence, the need for more subsidy than was anticipated. The consequent income redistribution is widely resented. Time and again we return to George Washington's central maxim as president: honesty is the best policy.

{top quote}
How To Recycle the Income {bottom quote}
Terminal Care is Mostly Medicare
Go With the Flow. To return to the point, almost everybody now dies with terminal expenses covered by Medicare. The medical industry waits for about six months, and eventually gets paid for its services by Medicare, incompletely perhaps, but for the most part. From this is derived the insurance shorthand concept of the "last year of Life costs" which largely represents terminal illness. Along with the obstetrical costs of being born, these two costs are the only two we can safely assume will continue forever. Everything else is like the Federal Reserve's Quantitative Easing. We can be certain it will stop, but we have no idea how long it will take. While we can perceive that medical progress is largely a matter of removing diseases from the list, the allied perception is that younger people are always going to be somewhat healthier than older ones. There will be exceptions like HIV/AIDS, but the perception is probably permanent. Since progressively older people are supported by savings and wealth transfers, the concept of investing the savings of younger people in order to sustain them when they get older, seems a dependable one. It also seems safe to assume that people will resent it less if it is their own money, rather than drawn from a public pool. That is, savings will be resented less than taxes, incentives will be resented less than coercion, and the final outcome will be the accumulation of more savings within private hands than within national treasuries.

Now add the idealized extra specifics: if subscribers by contribution, gift or subsidy create a Health Savings Account early in life, and Medicare can be induced to reduce its own premiums out of recognition of equivalent reserves in the funds, the future payment of (at least) last-year-of-life costs could be assured -- and current premiums for Medicare could be accordingly reduced, putting the money back in people's pockets. In this way, Medicare and the subscriber would adjust to the benefits of a major new revenue source, the investment proceeds of the Health Savings Accounts. A whole bundle of uncertainties absolutely do remain -- the zig-zag of interest rates, the volatility of the stock market, the elimination of some diseases, the creation of expensive new treatments, the actual longevity of the population, and the constant menace of inflation -- but one certainty survives. To a significant degree, a new source of income would effectively lower the cost of health care, even though it may not have paid for all of it precisely. No rationing, no income redistribution, no great change in how medicine is practiced. The opportunity seems too attractive to dismiss, but it must be continuously and openly monitored.

{top quote}
Income is fairly predictable, Future Costs are not. {bottom quote}
Balancing the Books
Simplify. Since the ultimate outcome is uncertain and will surely vary from predicted, adequate provision must be made for both the possibility of surplus and deficiency. Because of possible surplus, residuals should be allowed to pass through inheritance, or to be spent for non-medical purposes, or both, as incentives to compliance rather than circumvention. Because of potential shortfalls, some process for early detection and freezing of shortfalls should also be created, leading to subsidies, and restitution of subsidies, under defined circumstances. To the maximum degree feasible, the "accordion" principle should be employed, whereby benefits are expanded or contracted to reflect surplus and deficiency in the individual Health Savings Account, and indirectly, the growing success or failure of the scheme. Furthermore, the use of average costs rather than specific ones is encouraged, thus making it easier to deal with average lifetime health costs. In a computer age, it is almost as easy to report 300 million individual accounts as to measure by average performance, but it leads to intolerable public confusion about what is happening to the program. If achievable, a transformation from annual bills to lifetime costs would allow the elimination of pre-existing condition exclusions almost without effort, since lifetime liability would remain unlinked to HSA balances, and even largely unaffected by individual catastrophic illness if pooling is added. It might require considerable research, however, to detect the creation of unforeseen loopholes to game the system. Ultimately, that is a gigantic undertaking. The more we can simplify it with self-enforcing incentives, the more likely we can perform the essentials well, side-stepping all the work and aggravation of playing cops and robbers.

Get Started. The two quickest ways to induce large numbers of people to create Health Savings Accounts would be to add a permanent rollover feature to Flexible Spending Accounts, which currently contain a use-it-or-lose-it feature. Because of the cost to employers, it would be a useful opportunity to remind them of the inequity they have enjoyed from seventy years of Henry Kaiser tax exemption. And the second accelerant would be to eliminate the income tax discrimination against it, by allowing health insurance premiums to qualify for purchase by Health Savings Accounts, and thus to become tax-deductible like almost everybody else's health insurance.

Originally published: Wednesday, February 12, 2014; most-recently modified: Monday, May 13, 2019