Detoxifying the Debate

As an art form, caricature is fun. The caricature of ideas, however, does not have the same appeal. And when the caricaturists seek to arouse fears and anxieties by distorting unfamiliar ideas into misshapen and threatening images of insidious evil and betrayal, they do public debate and even their own case a great disservice.

In an article in the previous issue ("Healthy Compromise: Universal Coverage and Managed Competition Under a Cap," TAP Winter 1993), I presented a proposal for universal health insurance through new, regional health insurance purchasing cooperatives, which would represent the consumers and employers who pay for health care and be responsible for keeping the growth of health spending within a nationally set budget limit. The purchasing cooperatives would negotiate with competing health plans over costs and services, provide consumers information about their alternatives, conduct an open enrollment, adjust the total payments to plans in line with the risk of their enrollees, and monitor the performance of plans to ensure that they provided the services to which their enrollees were entitled. These are some of the main responsibilities involved in "managing competition."

The American Prospect article was a further development of ideas I outlined in a short book, The Logic of Health-Care Reform, published last October. A still more detailed discussion appears in an article that Walter Zelman and I published in a recent special issue of the journal Health Affairs, which includes additional papers on the same reform strategy. (As deputy to John Garamendi, California's elected commissioner of insurance, Zelman helped to craft a proposal for publicly financed health coverage through competing plans that set off much of the recent debate.) I invite any reader disturbed by the two previous critiques of managed competition to turn to these sources for a more accurate picture of the approach, as some of us have conceived it. Here, speaking solely for myself, I wish only to restate some key points about this strategy and its relation to single-payer proposals for national health insurance.

Reading these indictments by advocates of Canadian-style single-payer system, one would hardly guess that the single-payer model and the kind of managed-competition approach I've supported have much in common. Both would decouple health insurance from employment--that is, everybody would have a right to health coverage regardless of whether they worked. No one could be excluded or required to pay a higher cost because of any pre-existing condition. Everyone would be entitled to the same broad benefit coverage, according to standards set in federal law.

In their sources of revenue to pay for those benefits, managed-competition and single-payer approaches face similar choices and political constraints. Both approaches could be financed by premiums or taxes, set at the local, state, or national levels. The Garamendi plan relied on a payroll tax. The proposal I outlined in the Winter issue called for the purchasing cooperatives' premiums to be divided between employers and employees, with the employers' share capped as a percentage of payroll and the employees' share as a percentage of family income; similarly, the premium obligations of the self-employed and those out of work would also be capped as a percentage of income. The effect of this approach is to preserve continuity with the current premium-based system (hence to avoid the odium of taxes) but to make the incidence of costs more progressive through the lower and middle parts of the income distribution. The revenue for health coverage could also come in whole or part from excise, value-added, or income taxes at the national level. Although the managed-competition approach has the political advantage of routing the funds through regional cooperatives and directly to the plans that people choose (thereby giving them a sense of direct control over the use of their insurance payments), both approaches are compatible with any of the various potential revenue sources.

The resemblance goes further: A purchasing cooperative that embraces all consumers in a region is a single payer--that is, a single payer of health plans. Moreover, the proposal I presented would not preclude a state from organizing a single purchasing cooperative to pay doctors and hospitals. Some of the more rural states--Vermont and the Dakotas come to mind--might well do so because their low population densities make it difficult to support effective competition among plans.

Subscribe to The American Prospect

Conversely, while a single-payer system might pay some doctors and other providers individually by fee-for-service, it might pay a fee per enrollee--that is, a "capitation" payment--to other providers grouped together in integrated health plans. These plans would likely compete with each other and with fee-for-service providers. Indeed, in states like California and Minnesota, which have many integrated health plans and multispecialty group practices, it is difficult to imagine any other result under a single-payer system. Inevitably, the single payer--presumably the state--would have to determine the appropriate payment levels to different types of plans and providers and how to manage the competition among them. Thus while a managed-competition system might become a single payer in states without integrated health plans, a single-payer system would have to manage competition in regions where integrated plans are plentiful.

I emphasize the likely overlap between these approaches to detoxify the discussion, to explain why many proponents of single-payer and managed-competition models in Washington agree on the basic framework of legislation as long as it provides for state flexibility, and to suggest that the real choices here are not drawn out of a Manichean universe. Curiously, like some right-wingers, the caricaturists of the left have posed a black-and-white choice between a market model on the one hand and a government program on the other. But the division is not so neat. Not only does the managed-competition model call for government to spread the costs of health care over the community, to set new rules for health insurance, and to use a new vehicle--the purchasing cooperatives--to give consumers greater power in bargaining with the plans and providers. It also calls for a National Health Board to evaluate new technologies (and, just as important, to reevaluate old ones), to set guidelines for quality of care, and to require uniform reporting to assure comparable data for assessing performance. This approach hardly expresses a fealty to a "textbook" image of a market.

A single-payer system, on the other hand, would scarcely banish markets in health care. It would leave in place private hospitals, private imaging centers, private laboratories, commercial drug companies, and so on. Even nominally nonprofit medical institutions, moreover, have been known to respond to economic incentives. I am told that physicians are not entirely unresponsive to the signals and on occasion the seductions of the market. But, in the eyes of the caricaturists, the evil of "price" enters the picture not with any of these veterans of the marketplace but only with a private health insurance plan. For mystifying reasons, they are as soft on private fee-for-service providers as they are shrill on health maintenance organizations.

To be sure, a genuine and important difference between these approaches is that managed competition is designed to reconstruct the economic incentives in health care to encourage providers and consumers to make cost- and quality-conscious choices. For while a single-payer system might pay for enrollment in integrated health plans, it is not designed to set up a clear comparison among such plans or between those plans and conventional insurance, much less to reward consumers for choosing a plan that succeeds in controlling costs.

Here is where the two models deviate critically from one another. Under the version of managed competition I've supported, the purchasing cooperatives would use their clout to bargain aggressively with plans over their premiums. The plans would range from staff-model HMOs to prepaid provider networks to preferred provider plans to a conventional, free-choice-of-provider insurance plan. (For lack of any ability to exclude high-cost providers, the latter would be virtually certain to be more costly.) All plans would compete on the same benefit package, but some plans could offer that package for a lower price because of their mode of organization. The revenue flowing into the purchasing cooperative from employers, employees, and government subsidies would pay a "benchmark" rate. That benchmark premium could be set at the price of the lowest cost plan, or the average of the three lowest, or at some percentile--say the thirtieth--of all premiums. A consumer that chose a plan that cost more than this benchmark rate would pay the extra out of pocket. On the other hand, a consumer that chose a plan that cost less than the benchmark could get a rebate (this is, in fact, what happens in the employee health benefit program at Xerox).

Perhaps the core disagreement between the critics and supporters of managed competition on the left concerns the significance of this incentive for consumers to enroll in less costly health plans, which in turn encourages groups of providers to organize plans that attract consumers by keeping down their costs. The caricaturists charge that this incentive will "force" Americans into "stripped down" health plans. But let us consider the actual proposal, the principles, and the evidence.

The proposal I advanced--like other well-known managed-competition proposals--calls for a comprehensive, uniform benefit package for all. The benefit package in the Garamendi proposal, for example, covers physicians' services, hospital care, laboratory tests, home health care, and prescription drugs (to give a partial list) without annual or lifetime limitations. It has no deductible and limited copayments. This is no "stripped down" plan. Compare the Garamendi package to Medicare--the Garamendi plan is far more generous.

Indeed, the critics have the story about benefits exactly backward. A proposal for universal health insurance that does not have strong incentives for consumers and providers to reduce unnecessary and inappropriate care will be driven to downgrading the benefit package because there will be no other way to protect the public treasury. The Garamendi proposal could offer more generous benefits than Medicare does because its reference point was a health plan paid by capitation with internal incentives to control costs rather than a fee-for-service plan with incentives to drive up volume and technological "intensity."

The critics suggest that health plans will achieve lower costs by compromising the quality of care. But that is not borne out by the research. Indeed, quite the opposite: where there are measured differences between higher cost fee-for-service insurance and HMOs, the studies favor the HMOs. Deborah Stone writes that "the more money you have to spend" under managed competition, the more likely you will receive early diagnosis. Actually, the evidence is the opposite: Americans enrolled in HMOs receive more preventive and ambulatory care than do those in fee-for-service plans. There is no evidence to bear out her charges of inferior quality; nor does the evidence show, as she suggests, that the savings of HMOs are due to advantages in risk selection.

Repeated studies have shown, however, that there are enormous variations in health care costs from one region to another, and from one community to another within a region, without any relation to underlying disease patterns or outcomes of treatment. Physician practice styles--the rules of thumb about whether to recommend tests, further visits, surgery, hospitalization, and so on--are critical here. The rules often depend less on scientific evidence than on local custom, professional culture, and available time and resources. We know that the greater the supply of available beds, the greater the likelihood of hospitalization for conditions where there is little or no evidence that hospitalization improves the result. We know that if a specialist rather than a primary-care doctor treats the same condition, more tests and procedures will be ordered. And we have many more specialists in America than we need.

In other words, health plans that work with their physicians to encourage high-quality, conservative practice styles, that carefully manage the supply of hospital beds and technology, and that achieve a better balance between primary care and specialists among their doctors have a much better chance of controlling health care costs responsibly than does the fee-for-service system at large.

Encouraging such plans makes sense. But the managed-competition approach doesn't assume as a point of faith that they will achieve desired improvements in quality. Rather, it calls for the plans to agree to new rules of accountability and for regular monitoring of their quality of care by the National Health Board and purchasing cooperatives. Consider two of the issues raised by Deborah Stone: early diagnosis and waiting times. The stage at which various cancers are diagnosed is a readily quantifiable indicator of the quality of treatment. So, too, are waiting times. These and other indicators can be regularly measured, used by the purchasing cooperatives to set objectives for quality improvement by the plans, and provided to consumers to help inform their choices. Indeed, one considerable advantage of the managed-competition approach, unlike Canada's system, is its capacity to focus accountability on organizations that have comprehensive responsibility for costs, quality, and consumer satisfaction--and to give consumers a chance to walk if they don't like the results.

A comprehensive system for quality improvement does not yet exist in the United States. But we do have enough experience with employee health benefits programs that offer competing health systems to answer the charge about stripped-down plans. The programs that most closely approximate a managed-competition model are those that exist for public employees in states like California and Minnesota and at a few large corporations like Xerox. (The federal employee system also offers multiple health plans, but it does not bargain with plans over their rates.) The California system offers more than twenty HMOs and two different fee-for-service plans; Minnesota has one fee-for-service plan and five HMOs. Xerox deals with over 150 plans across the country. The HMO options tend to be lower in cost, and their costs have been growing more slowly, but they are by no means stripped down.

The caricaturists suggest that Americans will be on the verge of revolution once they learn about this kind of framework. I am not convinced. State and federal employees have good health benefits; a systemoffering all Americans those choices would hardly be unpopular. The fiscal crisis that hit California in the past few years would have led most other states or private employers to cut back health benefits. That didn't happen in California because the managers of the employee health program, acting like a purchasing cooperative, were able to drive down the rate increases while protecting the benefit package. Other Americans should have been so lucky.

The critics on the left point to Canada and suggest that we could cut our costs through administrative economies alone. But the problem of controlling health care costs is far greater and more difficult in the United States today than in Canada or other countries because of the excesses of recent decades. In addition to training too many specialists, we have built up a huge overload of capacity and costly technology in our hospitals and freestanding diagnostic and treatment centers. Compared with doctors elsewhere, American physicians have adopted more resource-intensive, high-cost patterns of practice. Were we to combine a Canadian-style, fee-for-service insurance system with America's patterns of health care investment and medical practice, the likely result will not be Canada's level of costs, but America's--indeed, quite possibly worse. I cannot think of a quicker way of bankrupting our country--and ultimately producing a backlash against universal coverage--than by agreeing to pay on a fee- for-service basis the costs that our health care system is now capable of generating.

Some critics have questioned how much savings we can expect to achieve through health plan competition alone. That is a reasonable question, but this approach does not rely only on competition. It has other mechanisms to contain costs as well--the concentration of countervailing economic power and expertise in the purchasing cooperatives, giving consumers a strong, informed agent in bargaining; a radical restructuring and consolidation of the insurance market, with the elimination of underwriting and other market-segmentation practices; a federal board to analyze the cost-effectiveness of technologies and procedures and to prescribe uniform enrollment, reimbursement, and other procedures for administrative simplification; and a budgeting system enforced by regulating the flow of funds through the purchasing cooperatives and, where necessary, by setting provider rates.

A comprehensive approach to reform will necessarily be eclectic and pragmatic. Before those on the left respond to what they take to be the ideology of managed competition, they ought to look closely at the framework and the key policy choices that lie within it. In one sense, to insist on nothing less than a government takeover of the health insurance system is too great a demand. But in another sense, it is too little. Our system does not need a mere takeover so much as it needs a thorough turnaround.

You may also like