President Clinton’s Health Security Act is the most important piece of domestic legislation since the Civil Rights Acts, and because it is an attempt to reform one seventh of America’s economy, it could prove the most significant economic change since the New Deal. Though no congressional vote is expected for about a year, the act has already been comprehensively analyzed and debated by politicians, doctors’ associations, health-care economists, insurance companies, small and large business groups, and journalists. Very little attention has been given to the most profound issue it raises. How do we decide whether the Health Security Act, or any other structure for medical care in the United States, is fair?
Everyone agrees that the United States now spends too much on health care. Medical services accounted for 14 percent of our gross domestic product in 1991—France and Germany spent 9 percent and Japan and Britain 8 percent—and economists predict that without reform medical expenses would grow to 18 percent by the year 2000. But how much should a nation like ours spend on its citizens’ health? How do we know that other nations are not spending too little, rather than that we are spending too much?
Most people also agree that health care is unjustly distributed in America. Forty million Americans have grossly inadequate medical coverage, or none at all, and many who do have adequate insurance now will lose it, because they will lose their jobs or develop a disease or condition that makes them uninsurable. In all, without reform, a quarter of all Americans will be without health insurance for some period during the next five years. But how much health care should a decent society make available for everyone? We can’t provide everyone with the medical care that the richest among us can buy for themselves. How do we decide what lesser level of care justice demands even the poorest should have?
Clinton’s act is long and elaborate—1,342 pages and 240,000 words. I shall summarize its main features. It aims to achieve universal coverage by requiring every resident of the United States to participate in some form of health plan. To reduce overall costs it would require people not to negotiate directly with a plan as individuals, but to join a large purchasing cooperative (called a regional health-care “alliance”) which would use its size and bargaining power to secure the lowest possible rates for health-care coverage. Each alliance would enter into contracts with a variety of plans and would offer its members a choice among them.1 Under the Health Security Act, state governments must create these alliances by 1998, and each state has considerable leeway in deciding how many to create and what territory within the state each will administer.2
Alliances must enter into contracts with any state-certified health-care “plan”—whether it is sponsored by an organization of hospitals and doctors similar to the health maintenance organizations (HMOs) that many Americans have joined in recent years or by a more traditional private health insurance company. But an alliance need not make a contract with a plan that charges more than 120 percent of the average premium of the other plans in the alliance; and all plans must offer the basic medical coverage the act defines as the “comprehensive benefits package.”
Plans may charge different premiums for that basic coverage, and may choose one of three arrangements, set out in the act, by which patients share some part of the cost of particular medical services: (1) a “low-cost sharing” plan under which individuals join an HMO and pay little or nothing when using doctors or facilities that are associated with it, but pay 40 percent of the cost when using other doctors or facilities; (2) a “fee-for-service” plan that allows patients to choose their own doctors, but provides for a substantial deductible amount and also provides for patients to pay 20 percent of the cost of any treatment themselves, subject to limits on the total amount any patient may be required to pay in any one year; and (3) a “combination” plan that includes membership in an HMO, higher charges than what the low-cost sharing basis imposes for the use of the plan’s own doctors and facilities, but much lower charges than it does for the use of outside doctors or services. These options allow individuals to judge how important it is for them to have a wide choice of doctors and hospitals when they are sick, and to pay higher charges for such a choice if they wish.
Each alliance must offer at least one “fee-for-service” plan, and doctors reimbursed under such a plan must adhere to fee schedules negotiated between the alliance and the region’s doctors, who will negotiate collectively under a special exemption from anti-trust laws. Alliances must review all plans they offer to assess their quality. They must collect and distribute information about the relative success of the care and treatment each has provided, and collect payments and premiums from members on the plans’ behalf. Plans must accept any applicant, unless they are oversubscribed, and they must fix their charges on the basis of “community rating,” which means that they may not charge more for those who are at greater risk of disease or accident.
Employers must pay, on behalf of their employees, 80 percent of the average annual premium of plans within the alliance jurisdiction, which means that an employee who chooses the cheapest plan will pay less than 20 percent of its annual premium, while one who chooses the most expensive plan will pay more.3 People who are self-employed or unemployed must pay the full cost of the plan they choose, but the national government would subsidize these costs, on a sliding scale, for anyone who could not afford them. (The act accepts a limit on such subsidies, in order to reassure fiscal conservatives who worried that the size of the federal commitment would potentially be very great.) Insurance companies may offer “supplemental insurance” policies that cover treatment beyond the basic package, provided that these policies are offered to everyone within the same health plan on the same basis. The act alters the tax law so that after the year 2002 payments for supplemental insurance are no longer deductible, and employees must count as taxable income any contributions that firms make to its cost.
The great bargaining power of the alliances should substantially reduce health-care costs, as I have said. The act nevertheless provides further financial safeguards. It creates a new national agency—the National Health Board whose seven members are to be appointed by the president—which is charged with reducing the annual rate of increase in health-care costs to the rate of inflation by 1999.4 The board would notify each alliance of a target rate of increase for its average premium costs in each year, and alliances would have the power to reduce the premiums set by the more expensive plans if that proves necessary to meet those target rates. In that way, the act provides a limit on the rate by which insurance premiums may rise, and that provision has, not surprisingly, provoked strong opposition in the insurance industry and among congressional conservatives.
If Clinton’s health care act were passed in its present form, it would provide coverage for everyone which no one could lose by falling ill or getting old or losing or changing a job. It would also provide mechanisms for slowing the rate of growth of aggregate medical costs. Much of the already intense debate over the health care act has centered on economic and medical issues. Would the cost-containing features actually work to hold down health-care costs? Would those features oblige doctors to compromise on medical care, rushing patients through treatment, or withholding desirable diagnostic studies or procedures, to meet their reduced budgets? How many people would be able to afford fee-for-service plans that allow them to choose their own doctors? How many would find their health-care costs actually increasing under the new plan? Would the new regulatory roles of states and the national government produce more rather than less bureaucracy, and make American medicine less rather than more efficient? Would forcing all employers to buy insurance for their employees cause many small firms to go out of business? These are crucial issues, and the public’s opinions about them—as well as pressures from interest groups—may decide whether the act is adopted and in what form.
But I shall concentrate on the issues of justice I described. Though the administration is careful not to mention the word, the act plainly rations health care when it defines the basic coverage which all plans must provide, and which will be guaranteed by the government for everyone. The act defines part of the basic package in detail: it describes, for example, a comprehensive schedule of immunization for infants and children, and routine screening and physical examinations by age groups: physical checkups, for example, every three years from age twenty to thirty-nine, every two years from forty to sixty-five, and every year thereafter, and mammograms to detect breast cancer for women every two years starting at age fifty. Some benefits—in dental, eye, and mental health care—would be limited at the start, but more generous after January 1, 2001. The act also explicitly excludes some kinds of treatment from the basic package altogether—most cosmetic surgery, for example.5
The act’s most important rationing provision, however, is not detailed but extremely abstract: it provides that medical treatment is part of the basic package only if it is “necessary and appropriate,” and it assigns the National Health Board the responsibility of determining what kinds of treatment are necessary and appropriate and in what circumstances. That board would have to decide, for example, whether bone-marrow transplants or other experimental forms of treatment are necessary and appropriate for particular diseases; it would determine whether most people can have such expensive procedures when doctors tell them that is their only chance, slim though it may be.
Does the act—or would the board—ration health care fairly? Should women have routine mammograms before age fifty? When should expensive magnetic resonance imaging be covered? Should the board ever include speculative bowel and liver transplants? If we cannot buy all the tests and treatments that everyone might want, how shall we decide, as a nation, how much we should spend, collectively and on each of us?
Some critics deny that health-care rationing is really necessary: they argue that if the waste and greed in the American health-care system were eliminated, we could save enough money to give men and women all the medical treatment that could benefit them. But though administrative expenses account for a significant part of hospital costs,6 and American doctors’ salaries are extremely large by other nations’ standards,7 the greatest contribution to the rise in medical costs in recent decades has been the availability of new, high-tech means of diagnosis, like magnetic resonance imaging and new and very expensive techniques like organ transplants and, on the horizon, monoclonal-antibody treatment for cancer. America is not paying all that much more for the medicine it formerly bought more cheaply; rather it now has so much more medicine to buy.
Many politicians and some doctors say that much of the new technology is “unnecessary” or “wasteful.” They do not mean that it provides no benefit at all. They mean that its benefit is too limited to justify its cost, and this is an argument for rationing, not an argument that rationing is unnecessary. There is an emerging consensus among doctors that routine mammograms for women under fifty, which are expensive, do not save many women’s lives. But they do save some.8 Heroic transplants that rarely work do work rarely. So we cannot defend the decisions the act makes, or the board would make, as simply avoiding waste. We cannot avoid the question of justice: what is “appropriate” medical care depends on what it would be unfair to withhold on the grounds that it costs too much. That question has been missing from the public debate, perhaps because the most emotionally laden decisions have been assigned to a board that would not be created until after the act is passed. But it is time the public began to take the question seriously.
For millennia doctors have paid lip service, at least, to an ideal of justice in medicine which I shall call the rescue principle. It has two connected parts. The first holds that life and health are, as René Descartes put it, chief among all goods: everything else is of lesser importance and must be sacrificed for them. The second insists that health care must be distributed on grounds of equality: that even in a society in which wealth is very unequal and equality is otherwise scorned, no one must be denied the medical care he needs just because he is too poor to afford it. These are understandable, even noble, ideals. They are grounded in a shared human understanding of the horror of pain, and, beyond that, of the indispensability of life and health to everything else we do. The rescue principle is so ancient, so intuitively attractive, and so widely supported in political rhetoric, that it might easily be thought to supply the right standard for answering questions about rationing.
In fact, however, the rescue principle is almost wholly useless for that purpose, and the assumption that it sets the proper standard for healthcare reform has done more harm than good. The principle does offer an answer to the question of how much America should spend on health care overall: it says we should spend all we can until the next dollar would buy no gain in health or life expectancy at all. No sane society would try to meet that standard, any more than a sane person would organize his life on that principle. In past centuries, however, there was not so huge a gap between the rhetoric of the rescue principle and what it was medically possible for a community to do. But now that science has created so many vastly expensive forms of medical care, it is preposterous that a community should treat longer life as a good that it must provide at any cost—even one that would make the lives of its people barely worth living.
So the rescue principle’s answer to the question of how much a society should spend on health care overall must be rejected as incredible. Once that answer is rejected, the principle has no second-best or fallback level of advice: it simply is silent. That is worse than unhelpful, because it encourages the idea that justice has nothing to say about how much a society should spend on health care, as against other goods, like education or controlling crime or material prosperity or the arts.
The rescue principle does have something helpful, though negative, to say about the other question of justice, which is how health care should be distributed. It says that if rationing is necessary, it should not be done, as it now largely is in the United States, on the basis of money. But we need more positive advice: What should the basis of rationing be? The egalitarian impulse of the principle suggests that medical care should be distributed according to need. But what does that mean—how is need to be measured? Does someone “need” an operation that might save his life but is highly unlikely to do so? Is someone’s need for life-saving treatment affected by the quality his life would have if the treatment were successful? Does the age of the patient matter—does someone need or deserve treatment less at seventy than at a younger age? Why? How should we balance the need of many people for relief from pain or incapacity against the need of fewer people for life-saving care? At one point the procedures of an Oregon commission appointed to establish medical priorities ranked tooth-capping ahead of appendectomy, because so many teeth can be capped for the price of one operation. Why was that so clearly a mistake?
We need a different, more helpful statement of ideal justice in health care, and we should start by noticing one problem that seems to make reform mandatory. Why does America spend so much—so much more than other nations—on medicine? In large part because individual decisions about how much health care to buy are made by patient and doctor but paid for by a third party, the insurance company, so that those who make the decisions have no direct incentive to save money. Insurance premiums are tax-deductible, moreover, and an employer’s contribution is not treated as part of the employee’s taxable income. So health insurance makes patients insensitive to cost at the moment of decision, and the real price of that insurance is subsidized by the nation. People would probably spend less on their own or their family’s care if they had to pay the actual cost themselves, at the expense of other goods and opportunities they might also want or want their families to have.
Of course, in the long run most people do pay the true costs of their health care, but they do so indirectly and unwisely, because employer contributions and tax funds could be used to buy what they would choose to have if they made the choice themselves: better schools for their children, for example, or economic investments and programs that would improve America’s competitiveness and give them greater job security. Our medical expenditures are therefore irrational: the system makes choices for people that they would not make for themselves, and the result is that our collective expenditures are too high—measured, as they should be, by how much care we really want, taken together, at the price we really want to pay.
Conservative economists seize on this fact: they say we should create a free market in health care by removing all tax benefits and subsidies so that people can have only the care they can afford. While that is, of course, an unacceptable solution, it is important to see why. It is unacceptable for three reasons. First, wealth is so unfairly distributed in America that many people would be unable to buy any substantial health insurance at market rates. Secondly, most people have very inadequate information about health risks and medical technology; they do not know what the risk of breast cancer is before the age of fifty, for example, or how much, if any, routine mammography before that age would add to their life expectancy. Third, in an unregulated market, insurance companies would charge some people higher premium rates because they were greater health risks (as, indeed, many insurance companies now do) so that people with a poor health history, or who were members of ethnic groups particularly susceptible to certain diseases, or who lived in areas where the risk of violent injury was greater, would be charged prohibitive rates.
This analysis points to a more satisfactory ideal of justice in health care—the “prudent insurance” ideal. We should allocate resources between health and other social needs, and between different patients who each need treatment, by trying to imagine what health care would be like if it were left to a free and unsubsidized market, and if the three deficiencies I have just described were somehow corrected. So we should try to imagine that America is transformed in three ways. Suppose, first, that the distribution of wealth and income is as fair as it possibly can be. In my view, that means that the resources people can initially command, in making their decisions about education, work, and investment, are as nearly equal as possible;9 but you should imagine an economic distribution that is fair according to your own views, whatever these are. (I shall assume, however, that on your views, as on mine, the wealth of everyone in a fair society would be much closer to the average than is true in America now: the great extremes between rich and poor that mark our economic life now would have largely disappeared.)
Second, imagine that America has also changed so that all the information that might be called state-of-the-art knowledge about the value and cost and side effects of particular medical procedures—everything, in other words, that good doctors know—is generally known by the public at large as well. Third, imagine that no one—including insurance companies—has any information available about how likely any particular person is to contract any particular disease or to suffer any particular kind of accident. No one would be in a position to say, of himself or anyone else, that that person is more or less likely to contract sickle-cell anemia, or diabetes, or to be the victim of violence in the street, than anyone else.10
The changes I am asking readers to imagine are very great, but they are not, I think, beyond the reach of the imagination. Now suppose that health care decisions in this transformed community are left simply to individual market decisions in as free a market as we can imagine, so that doctors and hospitals and drug companies are free to charge whatever they wish. Medical care is not provided by the government for anyone, nor are medical expenses or health-insurance premiums tax-deductible. There is no need to subsidize medical care in any such way, because people have enough resources to buy, for themselves the medical care they decide is appropriate. What kind of health-care institutions would actually develop in such a community? Would most people join health maintenance organizations that provided care by staff doctors at a relatively inexpensive rate? Would any substantial number choose more expensive insurance arrangements that allowed more freedom of choice in doctors or hospitals? Would the average plan or policy provide coverage for routine medical examinations or diagnostic screenings? What kind, how often, and at what age? How many plans or policies would provide, at appropriately high rates or premiums, experimental or very expensive or high-risk or low-expected-benefit procedures of different kinds? How much of its aggregate resources would the community devote to medical care through these various individual decisions?
It is impossible to answer these questions with any precision.11 But we can nevertheless make two crucial claims about justice. First, whatever that transformed community actually spends on health care in the aggregate is the morally appropriate amount for it to spend: it could not be criticized, on grounds of justice, for spending either too much or too little. Second, however health care is distributed in that society is just for that society: justice would not require providing health care for anyone that he or his family had not purchased. These claims follow directly from an extremely appealing assumption: that a just distribution is one that well-informed people create for themselves by individual choices, provided that the economic system and the distribution of wealth in the community in which these choices are made are themselves just.12
These important conclusions help us to decide what health care we should aim to provide for everyone in our own, imperfect, and unjust community. We can speculate about what kind of medical care and insurance it would be prudent for most Americans to buy, for themselves, if the changes I imagined had really taken place; and we can use those speculations as guidelines in deciding what justice requires now—in deciding, for example, which medical tests and procedures the National Health Board should decide are “necessary and appropriate” if the Health Security Act is passed.
Consider a twenty-five-year-old with average wealth and prospects, and state-of-the-art knowledge of medicine. Suppose he can choose from a wide variety of possible arrangements to provide for the health care he might want, under various contingencies, over the course of his life. What arrangements would it be prudent for him to make?13 He might be tempted, initially, to buy insurance providing every form of treatment or care that might conceivably be beneficial for him under any circumstance. But he would soon realize that the cost of such wildly ambitious insurance would be prohibitive—he would have nothing left for anything else—and decide that prudence required a much less comprehensive insurance program.
Of course, what is prudent for someone depends on that person’s own individual needs, tastes, personality, and preferences, but we can nevertheless make some judgments with confidence that they would fit the needs and preferences of most contemporary Americans. We can be confident, for example, about what medical insurance it would not be prudent for most people to buy, because some insurance would be a mistake no matter what happened in the future, including the worst outcome. It would be irrational for almost any twenty-five-year-old to insure himself so as to provide for life-sustaining treatment, to be provided if he falls into a persistent vegetative state of permanent unconsciousness, for example. The substantial sum he would have to spend in insurance premiums, year by year, to provide that coverage would be much better used in other ways to enhance his actual, conscious life. Even someone who lived only a few months after purchasing the insurance before he fell into a vegetative state would have made, in retrospect, a mistake, giving up resources that could have made his short remaining conscious life better, in order to buy a longer unconscious state.
We can enlarge this claim to include dementia as well as unconsciousness: it would not be prudent, for almost anyone, to purchase insurance providing for expensive medical intervention, even of a life-saving character, after he entered the late stages of Alzheimer’s disease or other form of irreversible dementia. The money spent on premiums for such insurance would have been better spent, no matter what happens, in making life before dementia more worthwhile. Of course, most prudent people would want to buy insurance to provide custodial care, in conditions of dignity and adequate comfort, if they became demented; providing for such care would be much less expensive than providing for life-saving treatment—for example renal dialysis or an organ transplant—if it were needed.14
Now consider a somewhat more controversial suggestion. In most developed countries, a major fraction of medical expense—over a quarter of Medicare payments in the United States, for example—is spent on people in the last six months of their lives. Of course, doctors do not always know whether a particular patient will die within a few months no matter how much is spent on his care. But in many cases, sadly, they do know that he will. Most young people on reflection would not think it prudent to buy insurance that could keep them alive, by expensive medical intervention, for four or five more months at the most if they had already lived into old age. They would think it wiser to spend what that insurance would cost on better health care earlier, or on education or training or investment that would provide greater benefit or more important security. Of course, most people would want to live those additional months if they did fall ill: most people want to remain alive as long as possible, provided they remain conscious and alert and the pain is not too great. But prudent people would nevertheless not want to guarantee those additional months at the cost of sacrifices in their earlier, vigorous life, although, once again, they would certainly want insurance to provide the much less expensive care that would keep them as comfortable and as free of pain as possible.
We can use these assumptions about what most people would think prudent for themselves, under fairer conditions than those we now have, as guides to the health care that justice demands everyone have now. If most prudent people would buy a certain level of medical coverage in a free market if they had average means—if nearly everyone would buy insurance covering ordinary medical care, hospitalization when necessary, prenatal and child care for their children, and regular checkups and other preventative medicine, for example—then the unfairness of our society is almost certainly the reason some people do not have such coverage now. A universal health-care system should make sure, in all justice, that everyone does have it.
On the other hand, if even under fair conditions very few prudent people would want to insure themselves to a much higher level of coverage—if, as I said, very few people would insure to provide life-saving care when demented, or heroic and expensive treatment that could prolong their lives only by a few months, for instance—then it is a disservice to justice to force everyone to have such insurance through a mandatory scheme. Of course, any judgment about what most prudent people would do is subject to exceptions: some people have special preferences, and would make very different decisions from those many other people would. Some people might think, even on reflection, that guaranteeing a few extra months of life at the end was worth great sacrifice earlier, for example.15 But it seems fair to construct a mandatory coverage scheme on the basis of assumptions about what all but a small number of people would think appropriate, allowing those few who would be willing to spend more on special care to do so, if they can afford it, through supplemental insurance.
If we substituted the prudent insurance approach for the rescue principle as our abstract ideal of justice in health care, we would therefore accept certain limits on universal coverage, and we would accept these not as compromises with justice but as required by it. Expensive treatment for unconscious or demented or terminally ill patients would be relatively easy cases to decide if we adopted that approach. Other decisions would be more difficult to make, including, for example, heart-wrenching decisions about the care of babies born so deformed or diseased that they are unlikely to live more than a few weeks even with the most heroic and expensive medical intervention. Last summer doctors in Philadelphia separated newborn Siamese twins who shared a single heart, though the operation would certainly kill one baby and give the other only a one in a hundred chance of surviving for long, and though the total cost was estimated to be a million dollars. (The twins’ parents had no medical insurance, but Indiana, where they lived, paid $1,000 a day toward the cost, and the Philadelphia hospital absorbed the rest.) The chief surgeon justified the procedure by appealing to the rescue principle: “There has been a unanimous consensus,” he said, “that if it is possible to save one life, then it is worth doing this.”
But the different standard I am defending would probably have recommended against the operation.16 Suppose people of average wealth, when they marry, are offered the opportunity to buy one of two insurance policies: the first provides that if any of their children is born with a life-threatening defect, neo-natal treatment will be covered only if it offers a reasonable (say 25 percent) chance of success, and the second—much more expensive—provides that such treatment will be guaranteed even if it offers only the barest hope. Most potential parents would decide, I believe, that it would be better for them and their families to buy the first policy, and to use the premiums they would save each year to benefit their healthy children in other ways—to provide better routine medical care, or better housing, or better education, for example—even though they would be giving up the chance for a desperate gamble to save a defective child if they ever had one.
If the act is adopted, the National Health Board, as I have said, will have to decide what medical procedures are “necessary and appropriate” and thus should be part of the comprehensive package of benefits everyone is guaranteed. Some of these decisions would be particularly difficult: deciding when very expensive diagnostic techniques or experimental organ transplants with a low chance of success are appropriate, for example. Such decisions must of course be based on the best and latest medical evidence, and must constantly be reviewed as that evidence changes. But they, too, should be guided by the standard of individual prudence: Would it make sense for someone to insure himself when young to guarantee a vastly expensive blood test which would improve the diagnosis of a heart attack by a very small percentage of accuracy if he should ever have doubtful symptoms of cardiac disease? Or to provide a risky, expensive, and probably ineffective bowel and liver transplant if doctors decided it would give him a small chance to live?
The rescue principle insists that society provide such treatment whenever there is any chance, however remote, that it will save a life. The prudent insurance principle balances the anticipated value of medical treatment against other goods and risks: it supposes that people might think they lead better lives overall when they invest less in doubtful medicine and more in making life successful or enjoyable, or in protecting themselves against other risks, including economic ones, that might also blight their lives. An agency such as the National Health Board might well decide that while prudent people would provide their family with the prenatal and well-child care that so many Americans lack, and would insure against serious medical risks at all stages of their lives by providing tested and reasonably effective treatment should they need it, they would forgo heroic treatment of improbable value if they needed it in return for more certain benefits like education, housing, and economic security, for example. If so, then justice demands that a national health scheme not provide such treatment.
In summary, the prudent insurance test helps to answer both questions of justice I mentioned at the beginning: How much should America spend overall on its health care, and how should that health care be distributed among its citizens? The test asks what people would decide to spend on their own medical care, as individuals, if they were buying insurance under fair free-market conditions, and it insists, first, that we as a nation should spend what individuals would spend, collectively, under those conditions; and, second, that we should use that aggregate expenditure to make sure that all have now, as individuals, what they would have then.
Of course some of the decisions I have been discussing would be made differently by different people trying to apply the prudent insurance test. It is very important that any agency charged with those decisions, like the National Health Board, should be made up of representatives of different groups that might be expected to make such judgments differently; it should have doctors and health-care specialists, of course, but it should also have ordinary people of various ages drawn from different parts of the country and, if possible, different ways of life. Such an agency could draw on the experience of countries with “single-payer” government-run health services which have had systematically to ration health care.
In Britain, for example, doctors in the national health system have been forced to allocate scarce resources like renal dialysis machines and organs for transplant, and they have worked out informal guidelines that take into account a potential recipient’s age, general health, life quality, and prospects, as well as prospects for adequate care by family or friends. Though this supposed cost-benefit test is different from the prudent insurance test, the decisions doctors have made under the former presumably reflect their judgments, guided by experience, about the relative value of different kinds of treatment at different ages and in different circumstances, and these are also judgments that a prudent insurer would be required to make.
The prudent insurance test also makes plain why it is so important to consult public opinion before rationing decisions are made. Since rationing should reflect not just technical cost-benefit calculations but also the public’s sense of priorities, consultation is essential. When Oregon sought to establish priorities in health care under Medicaid, it organized a series of “town meetings,” and a “parliament” to discuss the matter, and though the meetings were criticized by some because they were attended by very few of the poor whose health care was being debated, the meetings were nevertheless valuable sources of information about what those who did participate thought would be prudent insurance decisions. Clinton’s Health Security Act should make plain the importance of such discussions at the national level as well.
Still, no matter how much information an agency seeking to apply the prudent insurance test is able to gather, its results must be provisional, open to revision on the basis of further evidence of public preference as well as of medical technology and experience. In that respect, the Health Security Act’s decision to allow supplemental insurance at market prices, with no tax deduction or subsidy, fits the prudent insurance approach particularly well.17 If a very substantial number of people of average income buy supplemental insurance, in spite of its expense, that would suggest that the act’s definition of the basic package, or the board’s elaboration of that definition, was in error and should be changed. If most men bought supplemental coverage providing for routine prostate examinations every year beginning at an age earlier than the act specifies, for example, this would presumably reflect people’s informed judgment that more frequent examinations were prudent in spite of their cost, and this would suggest that the act should be amended to provide them.
Some of the Health Security Act’s explicit coverage decisions seem plainly right according to the prudent insurance test: to exclude most elective cosmetic surgery, for example, since the cost of insuring for wholly elective procedures is often prohibitive, and to include prenatal care and routine child-hood immunizations and checkups. Other specific decisions I mentioned are more debatable, like the possibly ungenerous schedule of routine medical tests and checkups for adults. The prudent insurer test poses another, more complicated, question, however. Does the act make some people pay too much for the coverage it provides? It requires most people to pay for their health care in three ways: through the impact on their wages (and perhaps on their jobs) of requiring their employers to pay 80 percent of the average premium; through their own contributions to those premiums and co-payments and deductibles their plan includes; and through the direct and indirect taxes they pay to provide government subsidies for the unemployed and others who cannot pay for themselves.18 Estimates (even those supplied by the administration) vary about how many people will pay more for health care than they now do if the act is passed. Among those who will pay significantly more, if Clinton’s proposed 75 cent increase in cigarette taxes to help finance the plan is adopted, are smokers; that seems just, on the prudent insurance model, because free-market insurance under fair conditions would presumably include higher premium rates for them. Most of those who will pay more are in higher tax brackets, and that, too, seems fair since they would presumably have less wealth if circumstances were more just.
Still, the danger remains that if the act, or any similar plan, is adopted, some low-wage earners or other relatively poor people will pay more for the health care they are forced to buy—through taxes, the impact on their wages and jobs, and their own contributions—than, given their stringent circumstances, they might prudently wish to spend. That is a genuine problem, but since poor people would have better coverage than most do now, and since those who now pay for coverage would presumably pay less, it is not a practical objection to the act that they would pay less under a fairer plan that is now politically impossible. (It would be a graver objection if the limit on federal subsidies led, as some liberals fear it would, to compromising universal coverage by withholding the full standard package of benefits from some of the poor.) The problem rather underscores what we already know: that even after extensive reform in health care, America will remain a deeply unjust society.
In any case, the Health Security Act seems better, to judge by the prudent insurance test, than any other scheme of reform with as much chance to succeed, and plainly better than the status quo. Some legislators, led by Representative Jim McDermott of Washington and Senator Paul Wellstone of Minnesota, continue to support a national single-payer plan, but almost no one thinks such a radical plan could be adopted. Right-wing Republicans favor plans, like Texas Senator Phil Gramm’s proposal, that leave health care largely as it is, with minimal extension of coverage to the poor. Senator John Chafee of Rhode Island, a liberal Republican, has offered a plan similar in many ways to Clinton’s, but it does not require employers to pay for their employees’ health care, and it proposes that any extension of coverage to those now uninsured be delayed until it can be paid for out of savings from other aspects of reform, including reduced aggregate premium charges and greater efficiency. Representative Jim Cooper, a Tennessee Democrat, has proposed a plan that subsidizes health care for the poor—a family of four earning $21,000, for example, would receive half the cost of the $4,000 annual premium that adequate coverage would cost—but even the Cooper plan’s supporters concede that many families would still be unable to buy adequate coverage on this scheme.19 Clinton’s Health Security Act will undoubtedly be changed, if it is adopted at all, probably in the direction of compromise with supporters of these other suggestions. But its general structure is more promising for justice than any realistic alternative yet proposed.
Some of the most potent criticisms of the Clinton plan, politically, have little to do with justice, but they must nevertheless be acknowledged. Critics insist that the act would create great new bureaucracies in each state and increase rather than decrease inefficiency in health-care provision. They also warn that a state’s decisions about how to divide its population into health-care alliances would be a source of bitter controversy and potential unfairness. An alliance whose jurisdiction includes high-density urban areas would have greater overall health costs than one that does not—it costs $63,000, on average, to treat a baby born with a narcotics addiction—and since an insurer must charge the same premium to everyone within the same alliance, people will be anxious that the state be divided, if possible, in such a way that their alliance includes no large city. The act imposes some restrictions on a state’s freedom to draw alliance boundaries—a city cannot be divided between two alliances, for example—but the broad discretion given to states would almost inevitably produce great political friction and resentment, and might result in substantial injustice as well.
These are also genuine problems, but the twin, competing, goals of universal coverage and cost control cannot be achieved without substantial government intervention, and hence without new federal agency powers and responsibilities, and though dividing the nation into different insurancecost regions does require arbitrary divisions many people will resent, the alternative is either to continue the unfair system of “experience rating,” which means that people pay rates reflecting their individual health risks and that those who need insurance most cannot afford it, or to establish nationwide premium structures which would be extraordinarily inefficient and give vastly more power to national bureaucracies.
The Health Security Act has also been criticized as threatening the health-care research and experimentation that have produced such extraordinary medical advances in recent decades. It is true that many valuable life-saving medical techniques that are now routine were once experimental, and that if expensive experimental procedures are not treated as appropriate under a scheme of universal coverage, many fewer may be performed. That is a further reason for allowing supplemental insurance, which would provide some opportunity for testing new procedures. It also underscores the importance of subsidizing medical research in other ways.20 Not everyone who wants government-subsidized experimental treatment will be able to receive it—the choice should be made on the basis of which treatment provides the most appropriate experimental test—but it is important for the nation (and the world) that some receive treatment rather than none at all.
The political battle over Clinton’s plan will be intense. We should be aware that even more is at stake, in this battle, than a more just system of health care, important though that is. Justice has had a bad time in America in recent decades—too many voters are unwilling to make any significant sacrifice for a more just society. Health-care reform has a reasonable chance of success, nevertheless, because so many voters are frightened by their prospects under the current system. Even those who have adequate insurance now fear losing it in the future, and though universal coverage is redistributive—many people will pay more, directly or indirectly, so that everyone can be covered—it seems appealing because almost everyone can imagine circumstances in which a universal plan would help them in case of disaster.
If, in spite of this political opportunity, America proves incapable of achieving any substantial reform—if the political power of insurance and drug companies and medical associations who think reform would be bad for them proves decisive—then the conventional wisdom that the public cares nothing for social justice will solidify, and presidents will be unlikely to risk challenging that wisdom again for many years. But if reform succeeds—if we finally end the national disgrace of being alone among prosperous nations in cheating citizens of life and health—we may be able to build on this success. We may rediscover a national sense that justice is worth its price in tax and government. We might then win greater justice in education, housing, jobs, and all the other goods and chances we now distribute so unfairly; we might begin to end the even greater disgrace that dooms some of us, at birth, to a bleaker life than the rest of us would think endurable. Abandoning the rescue principle, which insists that medicine is different from everything else, would improve the chance that health-care reform could have this consequence in the long term. Once we accept that health is one among many goods that make life worth living, we will have no justification for denouncing savage inequality in medicine and tolerating it anywhere else.
—December 16, 1993
January 13, 1994
People now covered by Medicare would continue to be covered, but could join alliances later. States would have the option to integrate Medicare patients into the alliances, if they provided coverage as good as Medicare does. Medicaid patients would be absorbed into the alliances, and receive the standard comprehensive benefits. ↩
The act gives each state the option, however, of establishing a “single-payer scheme,” under which the state itself pays all the costs of the medical care of its residents, including the salaries of the doctors and other professionals who provide that care. Many nations, including Canada, Germany, and Britain, have such single-payer schemes, and Clinton has been criticized for not pressing for a national single-payer scheme here. He did not do so, he has said, because he thought there was no chance Congress would approve one. ↩
Employers would not be required to pay more than 7.9 percent of their payroll in the aggregate, however; government subsidies would make up the difference. Large firms, which employ at least 5,000 full-time employees, are given the option of forming their own alliance for their own employees. ↩
Some economists argue that healthcare costs must inevitably rise at more than the average of inflation, unless the quality of medical care is seriously compromised, because the rate of medical productivity cannot rise as fast as that of other products and services. See William J. Baumol, “Health Reform Can’t Cure High Costs,” The New York Times, August 8, 1993, p. 13, and “Do Health Care Costs Matter,” The New Republic, November 22, 1993, p. 16. ↩
The act, as now drafted, includes abortion in the basic package of coverage. Section 1116 provides that a doctor or facility need not provide any service if he or it “objects to doing so on the basis of a religious belief or moral conviction.” ↩
According to a New England Journal of Medicine study, administrative costs were 25 percent of hospital costs in 1990. See Erik Eckholm, “Study Links Paperwork to 25% of Hospital Costs,” The New York Times, August 5, 1993. ↩
The average medical salary in the United States in 1992 was over $160,000. Salary varies dramatically by medical specialty: the average salary of a cardiovascular surgeon was $574,769 and a family practitioner $119,186. See “Health Plan Would Hurt Most the Doctors Who Make the Most,” The New York Times, November 7, 1993, p. 1. ↩
See Gina Kolata, “Mammogram Guideline is Dropped,” The New York Times, December 5, 1993, Section 1, p. 30. ↩
I explain this conception of economic fairness in “What Is Equality?” Part 2, which appeared in Philosophy and Public Affairs, Fall 1981. ↩
I am ignoring an important issue that my argument raises, but that I will not pursue here. Is it right, in the hypothetical exercise I am constructing, to exclude information relating risk of disease to voluntarily chosen behavior? Should insurance companies be in a position to charge cigarette smokers or mountain climbers higher premiums, for example? That seems reasonable. But, if so, what counts as voluntary behavior? Should sexual behavior of a particular kind be treated as voluntary for this purpose? It would seem wrong for insurance companies to charge active male homosexuals higher premiums because they are considered more likely to contract AIDS. Is this because sexual preference is less under people’s control than nicotine addiction? Or because the sacrifice in giving up sex is so much greater than giving up smoking? ↩
It does seem likely that even though the members of the imagined community would begin by making individual insurance decisions, they would soon develop, through these individual decision, collective institutions and arrangements like cooperative insurance purchasing agencies or pools, because these would provide economic advantages in a free market among people of roughly equal wealth. The result of the process might very well be something functionally very close to that proposed in Clinton’s plan. ↩
My claim needs minor qualification. Even in the imagined community, some paternalistic interference might be necessary to protect people from imprudent insurance decisions, particularly when they are young. And some constraints might be necessary to provide adequate resources for later generations. ↩
That is different from asking what any particular twenty-five-year-old would in fact do, because many people, particularly when young, do not make prudent decisions. They do not, that is, make the decisions that best serve the plans, convictions, tastes, and preferences they would find, on reflection, that they already have. It is prudent, of course, to provide for change—any prudent long-term insurance policy is written in general clauses rather than precise details of treatment, and is open to revision year by year. ↩
See my recent book, Life’s Dominion (Knopf, 1993), Chapter 9. ↩
People with certain religious convictions might make that choice, for example. But it is worth noticing that even those Catholics and others who think that it is always wrong to refuse available life-prolonging treatment do not necessarily think that all possible sacrifices must be made, in advance, to insure that very expensive life-prolonging treatment is available. Someone who thought it would be wrong, as a matter of religious principle, to decline an expensive and arduous operation that he could afford in order to extend his life a few months might nevertheless, with perfect consistency, think it prudent not to pay for the insurance, over his lifetime, that would enable him to afford it. He might think it more sensible to use the money that such insurance would cost to provide better medical treatment for himself or his family earlier, or better education, or some other goods or opportunities his convictions also deemed important. If so, then the prudent insurance test offers no reason why a national health scheme should make available for him what he would not have insured to provide for himself. ↩
At this date, one twin survives, but she remains in serious condition and is still breathing with the help of a ventilator. (See The Washington Post, December 12, 1993.) Even if she is able to lead some form of life, it would not follow that the decision to perform the operation, at a time when the chance of survival for long was remote, was a just one. ↩
The rescue principle might hold it unjust for anyone to buy better medical care than everyone can have, and that medical care outside a system of universal coverage should therefore be abolished or, as it is in Canada, seriously discouraged. But the prudent insurance approach begins in a different idea: that no one can complain, on grounds of justice, that he has less of something than someone else does, so long as he has all he would have if society were overall just. Suppose, for example, that if wealth were fairly distributed no one, in view of the cost and the prospects, would prudently insure to provide a liver and bowel transplant for himself even though it might conceivably one day save his life. If so, then justice does not require that such transplants be provided for everyone now, even when some people, unfairly rich, can buy them for themselves. See my article “Justice in the Distribution of Health Care,” McGill Law Journal, Vol. 38 (1993), p. 883. ↩
Clinton has tried to sharply limit the new taxes that his plan will require, and has delayed some benefits for that reason. For a criticism of this decision, and an appeal for more direct taxation for health-care reform, see Rashi Fein, “A Dangerous Year,” The New York Times, October 28, 1993, p. A27. Some health-care economists believe that the Health Security Act would in fact require substantially more, in new taxes, than the administration has so far conceded. But a recent study by an independent research group concluded that the plan is financially sound as it stands, adding that, “It meets the president’s requirement of providing universal coverage, and it does so without relying on an increase in broad-based income taxes.” See Spencer Rich, “Health Plan Funding Passes Muster,” Washington Post, December 9, 1993, p. A4. ↩
See The New Republic, December 6, 1993, p. 8. ↩
For a summary of the Health Security Act’s initiative in supporting research, see the White House publication The President’s Health Security Plan, Chapter 19. ↩