Belinda Miller's job can be messy and unpleasant. As director of insurer services for Florida's state Insurance Department, she is charged with putting sick HMOs out of their misery. "It's an absolutely horrible experience," she says.

She's not the only one who hates to go through it. When a health maintenance organization goes bust, policy holders are left scrambling for new coverage, which may not be available at a price they can afford. Some may be told they owe money to surgeons for expensive operations that the insolvent HMO failed to pay for. The physicians may dun them for payment, even though, by law, they aren't supposed to. Some patients get harassing phone calls from a collection agency. "It's quite frightening for them," Miller says. "And it's horrid for providers, too. They haven't been paid. Everybody gets hurt."

It's no wonder, then, that if there's any chance at all for a troubled HMO to survive, Miller and her department will help them find a new investor or try to raise the cash to cure the deficit. And for most of the past decade, they were quite successful at this mission. During the first eight years after Florida's HMO Consumer Assistance Plan was put in place in 1990, not a single liquidation was necessary.

Now, however, that isn't true. Belinda Miller is spending more and more of her time doing the job she desperately wants to avoid. In the past two years, the department has shuttered six HMOs. The most recent was just this spring: SunStar, the seventh-largest HMO in the state with 86,000 clients--was declared insolvent and put out of business.

Florida is far from the only state experiencing HMO fiscal free-fall. One of the most active players in Rhode Island's managed-care market-- Harvard Pilgrim, with 136,000 beneficiaries there--went belly up this spring. Illinois started liquidating one of its largest HMOs a few months ago, forcing 90,000 people to find new coverage. In New Jersey, the legislature passed a bill in March appropriating $100 million for losses resulting from the failure of two HMOs, and Governor Christine Todd Whitman added insult to HMO injuries by announcing that the state would issue new and tougher rules to tackle the problem of slow payments by HMOs to its providers.

These events are part of a larger fact of life: Two-thirds of the HMOs active in the United States today are losing money. One out of five lacks the capital in reserve to meet basic cushioning needs as spelled out by the National Association of Insurance Commissioners.

How could this be happening? Just five years ago, these kings of managed care seemed to rule the world. Whatever they wanted, they got- -be it insurer-friendly legislation, a Wall Street pat on the back or an ever-increasing chunk of business from state and local governments. Patients who were unhappy with HMO-based care, or physicians and hospitals distraught about treatment strictures or pay scales, were dismissed as so much flotsam on the road to progress. After all, in the early 1990s, health inflation had been rampant. Health care costs were eating up nearly 20 percent of state budgets. And HMOs were widely seen as the only practical cost-containment solution.

That was then. Now, HMOs are losing not only money but political clout as well. They're under legislative attack in state after state, with patient bills of rights, longer hospital stays for childbirth, reimbursement of clinical trials for cancer patients, and a whole array of coverage mandates, all passing surprisingly easily. Five legislatures--California, Georgia, Minnesota, Oklahoma and Texas--have given patients the go-ahead to sue their HMOs. Virginia's legislature has passed 21 mandated health benefits since 1997 and has another 11 under consideration.

Clearly, the mighty have taken a political fall. And in that tumble, there are serious implications for states, their health policies and budgets.

For starters, there's the issue of health insurance premiums. They are clearly rising, and at an accelerating pace. After several years of premium increases in the range of 4 to 5 percent a year, many states are seeing increases of 9 percent for some of the large health purchasers--those who can exert the most leverage in the market--with even more dramatic increases for those with less leverage. In Oregon this year, state employees--a very large group--were required to pay 9 percent more than in 1999, but medium-sized businesses experienced rate hikes of 20 percent and small businesses were clobbered with 40 percent increases. What that means, says Hersh Crawford, director of Oregon's Office of Medical Assistance Programs, is that "we are going to see smaller businesses no longer offering insurance to employees or the inability of employees to pay for insurance if the company makes it available. So we'll have a higher number of uninsured."

The instability of HMOs has deep implications for Medicaid. In many states--and in dozens of cities--managed care has become the main means of improving access to quality treatment for lower-income people. Where there were 2 million Medicaid patients using managed care nationally in 1992, there are now 16 million. Moreover, HMO cost- containment features have helped bring the price tag for Medicaid under control. After more than doubling between 1990 and 1995, these costs increased by barely 25 percent in the next five years.

HMO failures in the commercial field inevitably have a ripple effect on Medicaid HMOs, some of which operate in both sectors. When they are unable to continue serving private patients, subsidized patients lose their care as well. Two years ago in Oregon, some form of managed care was available to the poor in all but two counties. Now, there are six that don't offer it.

HMOs came by their problems the old-fashioned way: They earned them.

Once the Clinton administration's health plan failed in Congress in 1994, health insurance companies moved into the vacuum and offered both government and private purchasers a way to tame their spiraling health care costs: The coverage could be financed through a capitated system--a fixed annual amount of money would be allocated for the care of each beneficiary. In exchange for that money, the insurance company would provide whatever treatment was needed.

In small-scale versions around the country, and especially in California, the approach had been working well in the early 1990s. But when managed care went national on a big scale, insurance companies went into overdrive and tried to dominate the market in the regions where they operated. In many cases, that meant competing for business on the basis of price and dropping premiums below cost in order to hold on to customers. As Ronald Bovbjerg of the Urban Institute puts it, "Marketing people won out over actuaries."

Naturally, HMOs worked to bring down their costs as well. They set restrictions on treatment and tried to tame the overuse of health care services. They put the squeeze on physicians, hospitals and other providers to accept lower fees, and to provide service in more cost- efficient ways. In doing so, they eroded any sense of partnership or goodwill between insurers and providers. "HMOs contained costs by pounding on price and utilization," Bovbjerg says. "And they did that in ways doctors found obnoxious."

Even so, it seemed to work at first. The health care system had been marked by decades of waste and inefficiency. And HMOs eliminated a good bit of it. "Taking a few whacks out of that--reducing the use of emergency rooms and cutting down on excessive use of hospitalization-- was an easy way to make money up front," says Jim Vernier, a former Medicaid director for Indiana.

But once this was done, it got harder to keep turning the screws. Physicians and hospitals not only became angry about what was happening, they organized against it. "Providers have become successful in developing cartels that are preventing further price discounting," says Professor Robert Hurley, of Virginia Commonwealth University. "When physicians or hospitals organize into a larger group that says, `unless prices are what we want, we won't join a network,' that frustrates the ability of plans to get further savings."

At the same time, real costs are rising precipitously on another key component of the health care dollar: pharmaceuticals. While prescription medicine is driving up all health insurance premiums, HMOs have covered pharmaceuticals more generously than traditional plans, so they have been hit harder by the cost increases.

While most HMOs have been struggling to cope with all these problems, state legislatures have added new ones: laws mandating particular benefits and levels of patient coverage. "Each benefit may only add a little bit," says Jon Gabel, of the Health Research and Educational Trust, "but half a percent here and half a percent there can add up to a real burden."

Five years ago, it would have been unthinkable in most states for a legislature to tamper with an HMO's cost-containment tools--or to pass any legislation at all that could be described as anti-HMO. No patient bill of rights containing anything as controversial as the right to sue an HMO would have come close to enactment in states such as California, Texas and Georgia. The insurance companies had muscle, and they were able to partner with major local employers to lobby legislatures against imposing regulatory decisions on managed care practices.

That is no longer the case. The corporate lobbyists who befriended HMOs in the old days are keeping their distance now. This is due in part simply to good economic times. With unemployment low, companies are finding it a challenge merely to keep the work force intact. They are more reluctant than in the past to upset workers with a penny- pinching managed care plan. And they are nervous about fighting new coverage mandates imposed by the states. "Employers are still HMOs' strongest supporters," Gabel says. "But every time I talk to an employer group, their principal concern now is finding and keeping the employees they need."

But more is involved than the unemployment rate. HMOs are suffering from a political backlash all across the board: They've become the entity everyone loves to hate. And this has happened, argues Uwe E. Reinhardt, an economics professor at Princeton University, because HMOs made a fatal mistake early on. They failed to establish a convincing and humane public case for managed care. They tended to see their refusal to pay for certain health services not as the withholding of care but as efficient purchasing and smart fiscal stewardship. "In the eyes of patients, however, and in the eyes of health care providers, the media and the courts, that refusal to pay for certain health services tends to be viewed as `rationing' health care outright," Reinhardt wrote recently. He went on to point out that rationing of this type "becomes especially suspect among the public when it is practiced by profit-seeking, investor-owned companies who are easily demonized in the media."

The end result of all this has been managed care regulation. In the current climate of opinion, says Robert Hurley, it is relatively simple for legislatures to say, `Let's lay another requirement on health plans--it's costless and they're robust.' In his view, the inevitable consequence of heavy-handed state regulation will be a return to double-digit annual cost increases--and a new round of recrimination over "who lost the cost-containment battle."

In an effort to avoid further regulations, a few of the major HMOs have been moving voluntarily to weaken some of their unpopular cost- control tools. Last fall, United Healthcare, the third largest insurer in the country, announced it wasn't going to require physicians to seek permission anymore to order service or hospitalize patients. United claimed that such micromanaging was not cost effective--that 99 percent of treatment requests were approved and the company was spending more on its review process than it was saving on care it refused to cover.

Just this April, Aetna U.S. Healthcare, the largest health insurer in the country, settled litigation brought against it by the Texas attorney general's office. In answering the charge that its payment protocols were, in effect, an incentive for doctors to limit patients' medical care, Aetna agreed to provide more information about how it pays doctors and how it determines what care is medically necessary. "To me this settlement means that the new direction in health care signaled by United is becoming broader," says Paul Ginsburg, president of the Center for Studying Health System Change. "HMOs are backing off from some of the practices that have generated political opposition from physicians and consumers."

Meanwhile, the HMOs are trying to recoup their losses by boosting premiums 9 to 10 percent, a higher rate of increase than the rate at which overall health costs are rising. Even though spending on pharmaceuticals is going up at a superheated 12 to 13 percent a year, hospital care and physician service spending is growing much more slowly. In a change of strategy, says Ginsburg, "HMOs are willing to lose market share to restore profitability."

It seems hard to believe that just a few years ago, HMOs were given almost universal credit for breaking the cost spiral that had typified fee-for-service health insurance. From the 1960s through the 1990s, the health-spending component of the U.S. gross domestic product grew on average 3 full percentage points faster than the rest of the GDP. "Had the trend from 1960 to 1990 persisted until the year 2050," says Reinhardt, "about half of the GDP would be spent on health care in that future year." As it turned out, the percentage of the GDP absorbed by health care has remained virtually flat, at about 13.5 percent, since about 1993.

That is a genuine achievement, but it has been largely forgotten even by many in the health care field itself. There is little discussion now of how much money managed care has saved the American taxpayer. The issue is whether the wave of HMO failures is going to gain momentum, threatening the Medicaid program and burdening the state and local governments that are inextricably involved with it.

So far, most of the dropouts from Medicaid have been commercial managed care companies that write Medicaid policies along with their private business. The nonprofit health plans that deal exclusively in Medicaid--and are responsible for most of the Medicaid patients in several states--have been more stable.

But nonprofits can't handle all the Medicaid patients. The commercial HMOs have to be part of the arrangement in order for it to work. The more companies are winnowed out of the commercial market, the more fiscal and managerial strain there is on the system as a whole--and the more money states will have to pump into Medicaid to keep it afloat. "What will happen," says Robert Hurley, "is the states that have gotten good value from Medicaid managed care plans--true believers like Arizona, Minnesota, Wisconsin--will be willing to put money into the program to keep their plans participating. Ambivalent states--or those skeptical about HMOs--will drive them out or lose them."

In Tennessee, where the TennCare program, which replaced Medicaid, is predicated on managed care, there have been ominous signs of instability. Several managed care companies have gone under, and provider groups have quit the program. The problem is the level of state funding. Tennessee simply has not come up with enough money to pay for the rising costs of its program. That lack of cash has undermined providers and HMOs alike.

The bottom line seems to be this: For all its ups and down, managed care has improved access to treatment, quality of treatment, and accountability. The HMO is, despite its faults, a single entity that can be held responsible for providing appropriate care. No such entity existed in old-fashioned fee-for-service medicine. As much as they might complain, most major purchasers of health care--private employers, state employment managers and Medicaid directors alike-- seem to agree with Randall Bovbjerg's observation: "No one wants to go back to blank-check medicine. We can't afford it."