Bill Clinton boasts that his plan for universal health insurance will actually save hundreds of billions of dollars. Sorry folks, it won’t. But at least he frames the issue, while Ross Perot has talked only in generalities and George Bush’s plan would leave millions uninsured.

Our basic problem is contradictory expectations. A Washington Post/ABC Poll reports that 63 percent of us think health costs are too high. But we also see health care as a right. People shouldn’t be denied it because they can’t afford it. A survey by the Employee Benefit Research Institute asked how costs might be curbed. Less medical technology? No, said 57 percent. Less generous health insurance? No, said 61 percent. Less care for the elderly? No, said 86 percent. The trouble is that if health care is a right– everyone deserves the best, whenever it’s needed–costs will soar.

All other societies use government to balance these mutually exclusive expectations. Not us. We view the “health-care problem” as a series of discrete defects, all of which should be fixed. By now, these defects are well known. Spending has skyrocketed (from 6 percent of gross domestic product in 1965 to 12 percent in 1990). Too many Americans lack insurance (35 million, by the Census Bureau’s latest count). And increasing numbers of us (26 percent, by one survey) fear we’ll lose insurance coverage.

Every politician who addresses health care must pretend to solve all these problems without curtailing Americans’ right to choose their care. In effect, we demand that our leaders do the impossible. They try to oblige. Take the Clinton plan. It would achieve universal insurance by requiring companies to provide coverage for all workers. This is the easy part. It shifts most of the extra costs to business, which mainly passes them along to workers (in lower wages) or consumers (in higher prices).

The hard part is holding down total health spending. Clinton claims to do this by creating a Health Board that would set a ceiling on national health spending. Unfortunately, the board can’t easily enforce the ceiling. Even if it set rates for all hospitals and doctors, it couldn’t easily control the amounts of services (tests, visits, operations) performed. Similar problems have frustrated government efforts to limit Medicare and Medicaid spending. The huge savings Clinton claims are based on a “study” that simply assumes that the plan achieves most of its goals.

Quite the opposite might happen. Expansive mandates and lax controls could raise total health spending. To be effective, cost controls would have to be tougher. For example, Clinton might require people to join health-maintenance organizations (HMOs). Then he might limit the amount of money that could be paid to each HMO per member. HMOs would have to curb spending to stay within their budgets. But this approach would expose Clinton to the charges that he was depriving Americans of “choice” and undermining health-care “quality.”

The Bush plan aims to avoid such criticism. The president would let the uninsured buy insurance with refundable tax credits-subsidies-but the conditions are sufficiently restrictive that many people would remain uninsured. Meanwhile, Bush ignores the issue of costs and brags he’s preserving “choice,” as if our health care is a bastion of the free market. It isn’t.

We already have socialized medicine. Government pays 42 percent of health care and lavishly subsidizes private insurance through tax breaks. No one has any reason to cut costs because third parties-government or insurers-pay 90 percent of hospital bills and 81 percent of doctors’ bills. Patients expect the best, and doctors and hospitals earn more by ordering more tests and operations.

The choice we have made, unintentionally but logically, is to let health care overwhelm everything else. Since 1973, health-insurance costs have accounted for half of all workers’ increased compensation (pay plus fringe benefits), estimates the Congressional Budget Office. Economist Henry Aaron of The Brookings Institution reports that 29 percent of new government spending in the 1980s went to health care. By the year 2000, CBO expects health spending to hit 18 percent of GDP. And as baby boomers age, cost pressures will intensify.

The easiest way for government to control health spending is to create something like “health stamps.” Everyone would get a fixed amount of money (to be financed by taxes, but replacing employer-paid plans) to buy basic coverage. This would include coverage for catastrophic illnesses and might require copayments for routine care; people could still choose their own coverage. Anything beyond the basic package, people would pay for themselves without tax breaks. Ideally, insurers, doctors and hospitals would organize groups to compete for members. The groups with the best services and lowest costs would presumably attract the most members and be most profitable.

Some waste-unnecessary tests or operations-might vanish. But inevitably, strict spending limits would limit health services. Waiting times might increase for visits and surgery. Meager funds might impede valuable new technologies. The dread word rationing would be heard. The practical, political and ethical problems of such a system would be horrendous but, perhaps, less horrendous than tolerating today’s costly chaos. Every plan, including doing nothing, has immense drawbacks.

Only government can force us to see and confront these choices. The task of the next president is to change the terms of public discussion. It is to make us face our contradictory expectations. It is to compel us to compare drawbacks and to construct a better imperfect system. Our ideal system is make-believe, and if the next president pursues it–by perpetuating our illusions–he ensures failure.