Curing the crisis

Policy experts tell how they’d fix the U.S. health-care system

Insurance premiums are rocketing, the risk of losing health insurance is growing, public-health disasters are brewing. For the past few decades the nation’s health-care system has gone from bad to worse. But it doesn’t have to be this way. Five leaders in health-care policy offer their diagnoses of the system’s ailments and their prescriptions for cures.

Can these proposals work in the real world? Stanford Medicine asked front-line physician Clarence Braddock, MD, MPH, an internist and associate professor of medicine, and legislative expert Ryan Adesnik, Stanford’s director of federal government relations, to read the plans with feasibility in mind. Their queries and the policy experts’ responses follow each proposal.

Victor Fuchs
Jay Bhattacharya
Paul Wise
Alain Enthoven
Alan Garber

Illustration: Riccardo Vecchio


Health economist Victor Fuchs sees economics as “the quintessential policy science” that can help people develop solutions for social problems. He is the Henry J. Kaiser Jr. Professor of Economics and of Health Research and Policy, Emeritus. He is also a senior fellow at the Stanford Institute for International Studies and a core faculty member of the Center for Health Policy/Center for Primary Care and Outcomes Research. He has been particularly interested in studying the role of physician behavior and financial incentives in determining health-care expenditures.

Fuchs’ perspective

The two major challenges facing U.S. health care are to reform the present system of finance and to improve substantially the organization and delivery of care. Neither presents a “crisis” in the sense that failure to meet these challenges in the short term will result in collapse of the system.

Nevertheless, they present serious long-run problems, and every effort should be made to overcome them. Public policy-makers should give top priority to major reform of health-care finance. Without such reform, improvements in organization and delivery will be exceedingly difficult to attain.

The current methods of financing health care — employment-based insurance, Medicare and means-tested programs, such as Medicaid — are deeply and irreparably flawed.

Employment-based insurance, which now covers 55 percent of Americans, is inefficient and inequitable. It distorts labor markets, has high administrative costs and generates discontinuous coverage. Because it is paid through pre-tax dollars, it provides a larger tax break to high-wage than to low-wage workers.

Medicaid and other programs for families and individuals with income below set limits cover about 1 in 6 Americans. These programs require costly eligibility determinations, impose high tax rates on extra income of recipients because the subsidies fall or disappear as their income rises, and encourage evasion or avoidance of reported income. Many who are eligible do not apply — some to avoid the administrative hassle or stigma and others because they expect their income to improve. The programs also generate discontinuous coverage as eligibility changes.

Medicare, which covers about 1 in 8 Americans, while popular, has fundamental flaws. It is an open-ended entitlement that does not consider the cost of technologies relative to their benefits. In an era of rapid technological change, that is a recipe for financial disaster. Despite these programs, 15 percent of Americans have no health insurance; they either cannot afford to acquire it or are unwilling to do so.

Fuchs’ solution

The current system should be replaced by publicly funded social insurance combined with significant market elements. The best way to do this would be through universal vouchers that guarantee basic care to all Americans. A tax paid by business firms on all goods and services would finance the plan. All Americans would have a free choice of health plans and could buy more than the basic plan with their own after-tax income. Regional boards would pay the health plans a fixed annual amount per enrollee, adjusted for their age, sex, medical condition and other pertinent characteristics. Employment-based and means-tested insurance would end. Medicare would be phased out slowly to protect existing beneficiaries.

National and regional health boards, akin in structure to the Federal Reserve System administration, would oversee the system. A dedicated portion of the earmarked tax would fund an independent institute for technology and outcomes assessment to provide up-to-date, authoritative information on the cost-effectiveness of medical interventions.

With this system of finance in place, physicians and hospital administrators would, for the first time, have the opportunity, information and incentive to deliver cost-effective care to all Americans.

Followups for Fuchs

Do you really think these reforms could be accomplished in the near future given the current political dynamic in Washington and the budget constraints that have impacted federal and state budgets? — R.A.

Fuchs: No major reform of health-care finance can be accomplished in the near future. But such reform will be necessary someday. Now is the time to develop a carefully considered, thoroughly vetted approach that will be efficient, equitable and congruent with basic American values.

In what ways does this proposal differ from or have advantages over universal coverage by a single payer? — C.B.

Fuchs: Single payer does nothing to make the organization and delivery of care more effective or more efficient. The universal voucher proposal does.

The state of Oregon attempted a wide-scale assessment of the cost-utility of different health services, in essence dividing “basic” health care from “enhancements.” How will your proposal adjudicate those health services that some see as “basic and necessary” and others see as “enhancements”? — C.B.

Fuchs: The benefits provided by the universal voucher would be determined by an interactive, iterative process involving physicians, other health experts and the public, constrained by the knowledge that the total cost and the yield of the earmarked tax must be equal.

Illustration: Riccardo Vecchio


Jay Bhattacharya is an assistant professor of medicine and a core faculty member of the Center for Health Policy/Center for Primary Care and Outcomes Research. His research focuses on the constraints that vulnerable populations face in making decisions that affect their health status, as well as the effect of government policies and programs on these decisions. He earned both his MD and PhD (in economics) from Stanford.

Bhattacharya’s perspective

At the root of our health-care crisis are two related problems: the rapid increase in medical expenditures and the growth of our uninsured population. The solutions you likely favor for these problems depend on ideological positions on which there is, alas, substantial and irreconcilable disagreement.

If you believe health care is a right, then you are probably most concerned about the large number of uninsured and likely favor a universal, government-provided health insurance solution. If instead you believe that it’s an adult’s responsibility to provide for his or her health-care needs, then you are probably most concerned about rapidly rising medical expenditures and favor increased market competition as the mechanism to address the problem.

Bhattacharya’s solution

My suggestion won’t cure all of our health-care problems. Instead I offer a treatment for one critical concern: We have no mechanism to insure against health risk over the long term. When chronic illness strikes, premiums predictably rise permanently, which is the source of much financial uncertainty. Yet there is no market for insurance against this phenomenon.

To solve this, we need action by both the market and the government. A market for long-term health insurance contracts, in which insurers commit in advance to a fixed schedule of premiums over many years, could fully protect chronically ill individuals from facing a sharp hike in premiums.

An important reason why such insurance products are unavailable in the United States is the problem of unobservable health risk. The premiums for a long-term contract, fixed in advance, would be set to pay the average costs of medical care for the population, assuming that healthy and unhealthy alike all stay with the plan. Of course, the unhealthy would be more costly to insure than the healthy. And the healthy in the plan would realize they could get less expensive insurance elsewhere, and would have strong incentives to switch. If they did so, this would leave insurers in the lurch.

So, for this solution to work, insurers’ contracts would have to impose sharp withdrawal penalties, and the government would have to enforce these provisions. This sounds simple, yes. But in the current legal environment, it is not clear whether such provisions would be enforceable. A simple change in the law could yield substantial benefits for all of us.

Followups for Bhattacharya

What would this plan do to the health insurance premiums of the typical young, healthy worker? — C.B.

Bhattacharya: Because the plan reduces the problem of adverse selection, health insurance premiums will be closer to their actuarially fair levels. Such premiums might still be very high, though, since the plan does not directly change the incentive to adopt new, expensive technologies.

How would this program help the uninsured or underinsured? — C.B.

Bhattacharya: To the extent that the uninsurance problem is caused by adverse selection (which I believe explains a substantial portion of the uninsurance problem), it would dramatically reduce it. That is, relatively healthy people who do not have health insurance because the available products are too expensive will have affordable health insurance products available after the reform. There will still be some uninsurance because some people will be too poor to afford even actuarially fair insurance (but not poor enough to qualify for Medicaid). Medicaid expansions could be used to address the needs of these people.

Illustration: Riccardo Vecchio


Paul Wise is a health policy and outcomes researcher focused on children’s health, and has worked to improve health-care practices and policies in developing countries. Before coming to Stanford in July 2004, he was a professor of pediatrics at Boston University and vice chief of Social Medicine and Health Inequalities at Brigham and Women’s Hospital. Now he is a clinical professor of pediatrics and a core faculty member at the Center for Health Policy/Center for Primary Care and Outcomes Research.

Wise’s perspective

As in cosmology, there seems to be an inherent human impulse to seek a unified theory of why the U.S. health-care system is so messed up. This may be a constructive impulse in cosmology, but in the world of health policy, the multitude of proposed singular causes tends to be debilitating in that they are inevitably linked to a multitude of proposed singular solutions. Not surprisingly, a background noise of sweeping if isolated proposals, a cacophony of prescription, has emerged to dominate the public deliberation of health-care reform. If nothing else, therefore, the main problem with U.S. health-care policy is its profound disrespect for developing processes that could actually lead to meaningful progress. My suggestions, therefore, reflect a fatigue with the current discourse surrounding health-care reform in the United States and, consequently, advocate ways that could provide a more nurturing foundation for informed debate and purposeful action.

Wise’s solution

First, it would be useful to rethink the language of health-care policy. The fractious public debate over health-care reform has tended to attach an array of often unhelpful connotations to the words we use to describe proposed policy options, such as “competition,” “managed care” and “single payer.” Even the word “insurance” no longer works very well because we have come to think of health insurance in ways that differ markedly from how we think about other arenas of insurance, such as car or home insurance. When was the last time we expected our car insurance to pay for an oil change? There is an urgent need for a new language, a kind of health policy Esperanto that transcends the rhetorical baggage and permits a refreshed interchange of ideas and hopes. When the language truly serves the meaning, the commonalities and tensions among the various proposals tend to array differently and more constructively, and a new foundation for progress will emerge.

Second, with a refreshed language, public debate could confront more directly the fact that health care is now widely considered a public good but is being administered as if it were a private commodity. There is currently a profound mismatch of expectation and implementation, which in turn generates a rather incoherent set of incentives and disincentives. The resulting inefficiencies are glaring, particularly for patients, and have led to the paradoxical predicament of accelerating costs and eroding access to care. The sectors of the U.S. health-care system that have functioned the most effectively are those that are administered as if they were a public good. There is a lesson here.

Third, medical capability must be respected. Too often, the impulse to reduce costs can undermine the very thing that is central to any health-care system: ensuring access to the type and level of care that best meets a patient’s needs. Patients in need of general preventive care are best cared for by primary care providers who are readily available and are able to build strong relationships over time. Patients in need of highly specialized or emergency care must have access to specialists in their region that can best meet their critical needs. This kind of regionalized referral system seems like it should be simple to implement. However, it is under growing assault because too many health referral systems are based on financial contracts and not on clinical need. In the end, pressures to keep an extremely premature infant or an adult with a complicated but highly curable brain tumor in a community hospital will not only prove highly inefficient but also unjust.

Followups for Wise

What language would you substitute for the term “insurance?” — C.B.

Wise: I do not know what the best substitute for “insurance” would be; but I do know that the search for such a word is essential.

How will language help resolve the question of whether health care is a “right” or a “commodity”? — C.B.

Wise: I used the term “public good” rather than “right” in opposition to “commodity” because I think it expresses more constructively the general consensus that health care should be provided on the basis of shared, collective interest, like police, rather than on the basis of individual, natural entitlement.

How might new terminology revive the policy debate? — R.A.

Wise: Rethinking exactly what we mean when we say “right” or “insurance” would help invigorate public discourse in ways that would identify more creatively arenas of consensus and true tension.

Illustration: Riccardo Vecchio


Alain Enthoven is is co-creator of the managed competition model. He is the Marriner S. Eccles Professor of Public and Private Management, Emeritus at the Stanford Graduate School of Business, and a core faculty member of the Center for Health Policy/Center for Primary Care and Outcomes Research. He helped establish the Jackson Hole Group, a national think-tank that influenced former President Bill Clinton’s proposed health-care plan. He advocates a financially integrated health-care delivery system that relies on market-based incentives to reduce medical costs.

Enthoven’s perspective

The main problem in health care today in the United States is the high and rapidly growing level of health expenditure per person, reflected in insurance premiums for employment groups, and government outlays for Medicare, Medicaid and health insurance for public employees. The absolute amount of premium increases challenges the affordable increase in total employee compensation. For (simplified) example, suppose an employee has an earning power of $50,000 per year that is paid $40,000 in salary and $10,000 in family health insurance premium. The employer can afford a 3 percent increase in total compensation. That is $1,500. A 15 percent increase in the health insurance premium, which is what Stanford has been experiencing, is also $1,500, leaving no room for a pay increase. If this repeats next year, or if the earning power of the employee is less, the health insurance premium makes the employee unaffordable, giving the employer a powerful incentive to seek to roll back pay or health insurance (by e.g. higher deductibles, outsourcing the job or moving the job to a less costly area). Attempts at rollbacks have led to strikes and labor strife, and more is in store. If recent trends continue, Stanford is likely to be among those employers facing such a dilemma.

All told, 60 percent of health-care spending is a drain on public sector budgets, including health insurance for public employees, and the large drain on federal and state budgets created by the exclusion of employer-provided health insurance from the taxable incomes of employees. In 2004, the cost to the federal budget is estimated at $188.5 billion, with another $21.4 billion cost to state budgets. (John Shiels and Randall Haught, “The Cost of Tax-Exempt Health Benefits in 2001,” Health Affairs Web exclusive, Feb. 25, 2004.) Thus, the public sector in this country is now spending over $1 trillion on medical care. These expenditures and “tax expenditures” increase deficits and crowd out other valuable expenditures for education, criminal justice, welfare, infrastructure and other public benefits.

Enthoven’s solution

If I had the power to set three new policies to address this problem, I would:

1. Require employers to offer their employees a choice of at least three competing health services delivery systems, as Stanford does and as once was required by the HMO Act.

2. Require employers to make their contributions to employee health care in the form of fixed dollar amounts that do not increase with the cost of the plan chosen. Also cap the tax break for employee health insurance at the price of an efficient delivery system. Government would use the savings to make advanceable, refundable tax credits available for low-income people, equal to the price of the lowest-priced plan meeting quality and access standards.

3. Encourage the development of regional health insurance exchanges; that is, institutions set up to broker multiple choices of health insurance plans and delivery systems, so that people could keep the plan of their choice when they change jobs and so that the recipients of the tax credits would have a place to buy coverage.

The consequence of these measures is that people would migrate to what they see as value for money; usually lower-priced plans, as Stanford employees have done. Cost-conscious consumers would drive competition to improve quality and service and reduce cost, making care more affordable. I believe this process would lead to better quality care at about half the cost of fee-for-service care delivered through wide-access, preferred-provider insurance plans.

Followups for Enthoven

What evidence is there that the typical citizen can make a sound, informed decision when choosing a health plan? Won’t most people — especially young people — value low cost over quality or comprehensiveness of coverage? — C.B.

Enthoven: I think consumer choice needs to be structured by a competent, responsible manager, as is done at Stanford, to be sure that the competitors are all good quality, responsible long-term players and that the choices are of roughly equally comprehensive coverage. Then experience shows that people make good choices, are satisfied with their choices and there is not a lot of jumping around. At Stanford, employees have migrated to the HMOs that are the least costly.

Given your cost estimates, isn’t there enough money to support and fund a universal health coverage program? — C.B.

Enthoven: I think there is plenty of money in the system to pay for universal health insurance at the level of efficient health-care delivery systems.

Are there ways to phase in these changes that could have positive impact in the near term? — R.A.

Enthoven: The opening of the market to economical delivery systems ought to happen as fast as possible. It has been much too slow as it is. It will take lots of time for the delivery system to catch up. Subsidies for low-income uninsured and also caps on the tax-free employer contributions can be phased in over at least several years.

Illustration: Riccardo Vecchio


Alan Garber, a Stanford medical school alum, is the founding director of the Center for Health Policy/Center for Primary Care and Outcomes Research at Stanford and a staff physician at the Palo Alto VA Health Care System. He holds the Henry J. Kaiser Jr. Professorship and is a professor of medicine and, by courtesy, of business, economics and of health research and policy. His research focuses on methods to improve health-care delivery and financing. He served as a health-policy advisor to former presidential candidates Bill Bradley and John Kerry.

Garber’s perspective

What makes health-care reform so difficult to achieve in the United States? Economists who specialize in health care tell us that the ubiquity of health insurance is at the root of the problem.

Since most people in the United States have health insurance, few people directly pay the full cost of the care they receive. Most are exposed only to the deductibles (the amount of money they pay before the health plan begins to pay its share) and co-payments or coinsurance (the portion that the patient must pay for each service received after meeting the deductible). Because cost sharing traditionally has accounted for only part — sometimes a small part — of overall health expenditures, most people have been insulated from the costs of the health care they receive. In fact, cost information was hard to come by, especially for hospitalization.

Health plans also tend to leave costs out of decisions affecting individuals’ health care. They rarely consider costs explicitly when deciding whether to cover a new service, medication or procedure. The chief hurdle to obtaining coverage is proof of efficacy. That is, plans only feel obliged to cover a procedure if there is convincing evidence that it improves health outcomes, much as the FDA does when it decides whether to approve a new drug. The standards health plans apply are not as well codified as those of the FDA, and for the most part, the FDA requires more rigorous evidence of efficacy. But this standard is all that any intervention must meet — it can provide a very small health benefit, at great cost, and still be covered by most health plans.

Attempts to change this arrangement in the United States have been unsuccessful. Periodically the leadership of the federal agency for Medicare and Medicaid (now called the Centers for Medicare and Medicaid Services) has asked for the ability to deny coverage for care that does not represent good value — for example, to refuse to pay for a new treatment that is no more effective than a less costly alternative that is already covered. Each time a public outcry has firmly rebuffed them, and the agency has retreated from such efforts. Commercial insurers have somewhat different concerns — for example, the possibility that use of cost criteria would open them to embarrassing and costly lawsuits — but have been equally reluctant to restrict coverage based on the value of the care provided.

With plans unwilling to consider costs explicitly, and with the public largely insulated from the costs of care at the time they consume it, it is little wonder that health care, and health insurance, are so expensive and that the costs rise at a high rate every year. Furthermore, all the key players, including the public, have resisted change since nearly all proposed reforms of health-care financing can lead to a perceived loss of health benefits.

Garber’s solution

Until the general public — not the interested parties and policy experts who have debated these issues for years — becomes engaged in a realistic discussion about the choices available for both publicly and privately funded health care, change will be haphazard, chaotic and potentially disruptive. Many of us have ideas about ways to
change health-care financing in the United States, but no reform effort will succeed without public acceptance and support.

Followups for Garber

What steps might blunt the cost-impact issues you have raised while balancing the political sensitivities on Capitol Hill? — R.A.

Garber: We would need to explore alternatives to business as usual, which would enable Congress to exert some control without taking direct responsibility for the details. The Base Closure Commission is an example of an approach that was explicitly designed to overcome “pork-barrel politics” while allowing for extensive public input and accountability.

Do you think health-care inflation would be controlled if the public were more directly aware of, or even responsible for, the true costs of health care? — C.B.

Garber: Without information about costs, people lack the ability to choose less expensive options. Without responsibility for costs, they lack the motivation to choose care that offers good value.

Comments? Contact Stanford Medicine at

 Back To Contents