The medically uninsured have always been with us. Yet, they have not always been a cause for civic alarm. As a national issue, their plight has drifted into view only episodically, rarely with tangible results. Many thought this would change when President Bill Clinton proposed his ambitious health plan in 1993. But within a year, his proposal had been crushed and the issue seemingly forgotten--the fate of countless efforts to achieve universal coverage in the past.

Yet, here we are in the lead-up to the 2000 elections, and the issue of the uninsured has reemerged. The two leading Democratic presidential contenders, Vice President Al Gore and former Sen. Bill Bradley, have both endorsed universal health coverage for children and new insurance tax breaks for adults. Bradley has even proposed that the federal government spend $ 65 billion a year to cover 95% of the uninsured by allowing all Americans to buy into the private health plans offered federal employees, including members of Congress.

The issue is hardly exclusive to Democrats. In Congress, a bevy of conservative Republicans--among them, the former arch-nemesis of the Clinton plan, House Majority Leader Dick Armey (R-Texas)--want to create tax breaks for private insurance. These efforts follow on the heels of the unexpected health-care compromise of 1997, when Democrats and Republicans agreed to release more than $ 20 billion over five years for state programs to cover uninsured kids.

The standard explanation for renewed interest in the uninsured is that the problem is getting worse. It is. Since the demise of Clinton's call for "health care that's always there," health insurance has ceased to be there for about a million more Americans each year. The ranks of the uninsured, roughly 39 million in 1993, reached 44.3 million in 1998. The reasons for this continuing erosion of coverage are complex, but two stand out: the continuing decline in the proportion of small and low-wage firms that sponsor or subsidize coverage, and the simultaneous movement of many former welfare recipients onto payrolls of just such firms.

The profile of the uninsured is similarly complicated: A quarter are children, a quarter are poor, a sixth are noncitizens, an eighth are well off economically (earning more than four times the poverty level) and roughly 5% are unemployed. Most, however, are workers earning modest incomes: men and women with a high-school degree or less who work full or part time and yet still have difficulty making ends meet. These are the workers who have fallen behind in the information economy, their wages essentially stagnant for two decades, their workplace benefits eroding.

What is most surprising about the continuing increase in the uninsured is that it is occurring during a time of remarkable prosperity, when government coffers are flush and medical costs are widely seen as under control. Clinton's plan, constructed in the harsh budgetary climate of 1993, was basically a zero-sum proposition in which tough new regulations and mandates would permit expanded coverage. By contrast, today's leading legislative proposals mainly promise to shower more money and special tax favors on Americans' already heavily subsidized employment-based system.

What's also crucial about the current situation is that the partisan complexion of health care has changed since 1994, in part because of the failure of the Clinton plan. After the plan's defeat, the public began to worry less about intrusive cost-cutting by government and more about intrusive cost-cutting by aggressive private insurers. This turnaround has posed particular dilemmas for Republicans, generally opposed to strict regulation of private health plans. By proposing new tax breaks for private insurance, many Republicans want to shift the focus from health-plan regulation to the problem of the uninsured.

Yet, many Democrats also have moved to embrace tax-credit proposals, though largely for different reasons. The most important influence has been the tempered sense of the possible that followed the demise of Clinton's plan. Democrats have spent the last five years wondering if they could have achieved significant change if they had sought more modest goals. Reluctantly, most have concluded, as Gore recently put it, that health-care reform must be accomplished on a "step-by-step basis."

Gore's proposal is, in fact, exhibit A for this new strategy. In 1992, his collection of tax breaks and modest program expansions would have been dismissed as inadequate. Actually, it was--remember President George Bush's health plan? But at the time of its release, the plan was heralded as brave. Even Bradley's new campaign proposal, the most ambitious yet, has no mandate on employers or individuals to ensure coverage and no provisions for cost control.

Indeed, most of the current plans either paper over these problems or pretend they don't exist.

Take expanded coverage, the primary goal of many proposals. Since most Americans already have insurance through employers or public programs, incremental coverage expansions need to tread gently around existing arrangements, lest they encourage further reductions in coverage or end up merely reshuffling public accounts. The children's health programs authorized in 1997 have had disappointing results for just this reason. Even as states have reached out to uninsured kids, employers have continued to shed dependent coverage and hundreds of thousands of children have left the Medicaid rolls.

What's more, if coverage is less than universal, government officials always have to worry that new or expanded public programs will attract the sickest and hence most costly patients. This is a major fear raised by Bradley's proposal to open the federal-employee health program to all Americans.

Insurance regulations are also often touted as a way to help the uninsured, but when employers or individuals are free to drop coverage, restrictions on what insurers can do will raise rates for poorer or healthier Americans--the very groups already most likely to be uninsured.

Nor will tax credits alone do the job. Because the vast majority of the uninsured are working adults of modest incomes who have little or no federal income-tax burden, any credit to help them would need to be refundable. A tax credit alone would also mean little unless affordable plans were available, as they are not for most who buy insurance outside the workplace today. Yet, until now, the major proposals with any chance of passage would merely extend the existing tax deduction for health insurance to those who purchase insurance on their own, such as the self-employed. This would benefit few of the existing uninsured, while providing a generous tax cut for many well-off Americans.

Finally, consider cost containment. Some economists have argued that the amount we spend on medical care shouldn't be any more of a concern than the amount we spend on, say, computers. But once government commits itself to underwriting coverage for a significant portion of those without it, then the cost of health insurance cannot be so blithely ignored. Indeed, throwing billions in new federal subsidies into health care is a tailor-made recipe for medical inflation, as was learned after the passage of Medicare. The current proliferation of no-strings-attached subsidies for private insurance is a distinctive byproduct of the recent low rates of medical inflation. When rates begin to rise again (and they already have), the dilemmas of passive intervention will return.

None of this is to say that meaningful changes are impossible. Much will depend on the direction the debate takes and the outcome of the next election. Perhaps the most promising idea is Bradley's proposal to open up the federal-employee health program to every American. Not only does it have a populist ring, but if subsidies were linked to enrollment, it could serve as a back-door route to universal coverage within a common public framework.

Another idea with promise is universal coverage for children. The goal has deep affinities with Medicare, which also concerns a sympathetic group that lacks strong ties to the work force. But without a strict mandate on parents to insure their children, millions of kids will fall through the cracks--as more than 15 million do now under our existing patchwork of public and private programs.

That is the real challenge of step-by-step reforms: to think about how they will fit together and to anticipate any unintended consequences. This will require caution, creativity and foresight. It will also require some basic agreement on the direction incremental reforms should take, which has historically proved elusive. With the return of the uninsured to public consciousness, America has an opportunity to learn from its past failures. But nothing guarantees that the issue will not disappear again--as it has so often before.