The scene is Capitol Hill. It's the year 2035. Thousands of elderly protesters assemble outside the Capitol building. Inside, the House Ways and Means Committee meets to enact huge cuts in both Medicare and the national health-insurance program. Members are reluctant to take this step, but there's no choice. Although Congress raised the Medicare payroll-tax rate to 15 percent a decade ago, the Medicare program is still woefully insolvent, consuming 40 percent of the $10-trillion federal budget. Because the total burden of payroll taxes for all social programs has reached 45 percent, the Congressional Budget Office estimates that half of the U.S. economy has gone "underground," like Latin American economies in the 1980s. Another hike in payroll taxes would only drive more of the economy underground. But elderly voters, who now make up a majority of the electorate, have swarmed to Washington demanding "fairness," since they have paid into the system for so many years.
Sound farfetched? The numbers suggest that this scenario could actually come to pass as soon as the next decade, not way off in 2035. The Medicare program is heading for a smashup, yet our political leaders speak only of instituting new federal health-insurance programs that would cover everyone.
The rhetoric and specific policy claims being made on behalf of the various universal government health insurance reforms today are remarkably similar to the promises made on behalf of Medicare a generation ago. Indeed, when Medicare became law, many of both its supporters and opponents predicted that it was the first step toward universal health care provided by the federal government. When Medicare finally passed in 1965, Rep. Phil Burton (D–Calif.) expressed the sentiments of many among its Great Society enthusiasts when he said, "I am equally certain that before many years Congress will choose to extend comprehensive medical care as a matter of right to every man, woman, and child in this country."
So the history of Medicare and the outlook for the program over the next generation provide a sobering lesson for today's would-be designers of national health insurance. Unfortunately, no one seems to be paying attention to what the Medicare experience has to teach.
Today's proposals for a universal national health-care policy typically divide into either a government-run "single-payer" system like Canada's or a "play or pay" scheme that would require employers to provide health insurance for every worker or pay a payroll tax into a government insurance program. Advocates of such policies claim these programs won't cost much because significant savings can be had through cost containment and other efficiencies of scale. The experience of 25 years of Medicare says otherwise.
The two primary lessons of Medicare are the chronic problem of woefully underestimating program costs and the impossibility of genuine cost control. A closer look at Medicare shows why these two problems are certain to plague a government-administered universal health-care plan.
The cost of Medicare is a good place to begin. At its start, in 1966, Medicare cost $3 billion. The House Ways and Means Committee estimated that Medicare would cost only about $12 billion by 1990 (a figure that included an allowance for inflation). This was a supposedly "conservative" estimate. But in 1990 Medicare actually cost $107 billion.
This is a mere bagatelle compared with "conservative" projections for the next generation. The Congressional Budget Office estimates that Medicare will cost $223 billion by 1997. Constance Homer, deputy secretary of Health and Human Services, warns that "by the year 2003, at the current rates, we will be spending more on Medicare than we do on Social Security."
The news gets even worse for the "out years" after that. The Health Care Finance Administration has given up making long-range projections of budget outlays of Medicare. Instead, HCFA makes calculations about the "actuarial balance" of the program—how much of the nation's payroll will be required to pay for the program.
The 1992 annual report of the Federal Hospital Insurance Trust Fund, which pays for the hospital-insurance portion of Medicare, warns that the Medicare program "is severely out of financial balance" and could go bust as soon as the year 2000. The report says expenditures from the hospital fund represented 1.3 percent of the nation's gross domestic product in 1991 and will grow to 4.7 percent by 2065. To cover the cost, the Medicare payroll-tax rate will have to more than quadruple, from the current rate of 2.9 percent to 13.79 percent.
The full narrative of Medicare's enactment in 1965, a classic tale of legislative and interest-group infighting, is too long to recount in detail here. Proposals for Medicare-style programs began surfacing in Congress during World War II but didn't have a serious prospect of passage until the Kennedy administration.
Repeatedly in the early 1960s a coalition of Republicans and conservative Democrats defeated Medicare. The key figure in this perennial drama was Wilbur Mills, the legendary chairman of the House Ways and Means Committee. Mills refused to pass a Medicare bill out of that key committee, supposedly out of concern that Medicare would threaten the integrity of the Social Security program (to which Medicare is attached).
Following the Democratic landslide in the election of 1964, which gave Democrats a 2-to-1 majority in both houses of Congress, President Lyndon Johnson exerted his influence to stack the Ways and Means Committee with new Democrats sympathetic to Medicare. Wilbur Mills changed his mind and embraced Medicare. "Mills can count" was the explanation given for his flip-flop. This new political landscape virtually assured that Medicare would sail through Congress with huge majorities.
When it became apparent after the 1964 election that Medicare's passage was likely to be a slam dunk, the American Medical Association and Republicans scurried to put forward an alternative. The AMA foolishly tried to exploit public confusion over the fact that the Medicare proposal covered only hospital expenses but not costs for doctors, surgeons, dentists, and other outpatient services.
Arguing now that Medicare didn't go far enough, the AMA sought to outflank Medicare with an alternative program that would include outpatient services as well as hospital expenses. "Eldercare," as it was called, provided for a voluntary comprehensive insurance program, administered through the states and financed through means-tested premiums from recipients and federal matching funds. Not to be outdone by the AMA, House Republicans, under new leader Gerald Ford, offered their own alternative, which was similar to Eldercare except that it would be administered by the federal government.
The crafty Wilbur Mills responded to these alternatives by saying, in essence, thank you very much, we'll just add features from both of these onto the existing bill and create an even bigger program. Mills called the result "a three layer cake." Supplemental Medical Insurance was added to cover costs of doctors and other outpatient services. Finally, Medicaid was created to provide medical care for the nonelderly poor. (The costs of the Medicaid program, which requires state matching funds, now threaten to bankrupt many states.)
The hospital-insurance portion of Medicare was to be supported through a payroll tax shared equally by employers and employees. The voluntary Supplemental Medical Insurance was to be financed by premiums paid by the participants with dollar-for-dollar federal matching funds. The mechanism for increasing revenue for the hospital-insurance plan, when necessary, was the raising of the taxable earnings base.
To keep solvent, the Supplemental Medical Insurance system would adjust the insurance premium until premiums and matching funds covered expenditures. Congress generally dismissed fears of cost overruns. Rep. Claude Pepper (D–Fla.) said: "The cost will not be greater than our present efficient [sic] and wasteful fee-for-service system. According to experts the charge to the average family under a national health-insurance program will actually be less than it pays now, partly because the employer and government will contribute to the fund."
The deliberations about the cost of the hospital-insurance program make for a fascinating and almost comical story. Because it is hard to predict changes in medical technology and hospital costs, it was decided that the program could only be projected out 25 years, instead of the 75-year horizon that is used for Social Security projections. Even with this shorter time horizon, the projections turned out to be wildly inaccurate.
The most serious error the planners made was the assumption that hospital costs wouldn't rise faster than wages. Hospital costs had been rising 2.7 percentage points a year faster than wages over the previous decade. But the House Ways and Means Committee said it was a "reasonable" and "conservative" assumption that the difference between the rates of increase for wages and hospital costs would disappear by 1975, after which wages and hospital costs would rise at the same rate. Obviously, if hospital costs rose faster than wages, there would have to be a sharp increase in either the payroll-tax rate or the wage base against which it was levied.
A 1965 House Ways and Means Committee report on the actuarial basis for the hospital-insurance program proudly declared that "Congress has very carefully considered the cost aspects of the proposed hospital insurance system" and that "Congress very strongly believes that the financing basis of the new hospital insurance program should be developed on a conservative basis." The report acknowledged that hospital costs were rising faster than wages. But it dismissed the alternative scenarios that have turned out to be closer to what has in fact happened.
"It is inconceivable," the committee report says, "that hospital prices would rise indefinitely at a rate faster than earnings because eventually individuals—even currently employed workers, let alone older persons—could not afford to go to a hospital under such cost circumstances….Quite obviously, it is an untenable assumption that there can be a sizable differential between the increase in hospitalization costs and the increase in earnings levels that will continue for a long period into the future." This airtight logic didn't consider the effect of the increased demand that Medicare set off.
Anticipating a 3.5-percent annual inflation rate, government actuaries predicted that the cost of a day's hospital stay by 1985 would be $155 and that the hospital insurance portion of Medicare would cost $9 billion by 1990. The actual average cost of a hospital day by 1985 was over $600; instead of $9 billion, the hospital-insurance program cost $63 billion in 1990.
By the time Medicare passed the cost assumptions had been a subject of controversy for several years, especially with Ways and Means Chairman Mills. In a 1963 Ways and Means hearing, Mills clashed with the Social Security Administration's chief actuary, Robert Myers, about the accuracy of cost estimates made for previous Medicare proposals. Mills pointed out that if any of the previous bills had been passed in the late 1950s or early 1960s, they would already be underfunded. A few excerpts from the transcript tell the story:
"The Chairman: I am concerned about your estimates of the cost of the present  program as I look back to see what happened to your estimates of cost with respect to these other programs….Isn't it a fact that it would be about 100 percent underfunded today if we had enacted it in 1958 and provided exactly the tax then suggested as appropriate…?
"Mr. Myers: Mr. Chairman, I can't answer that question exactly now. [A week later Myers submitted a memorandum to Mills concluding that the program would have been about 50-percent underfunded.]
"The Chairman: What would be the situation had we enacted the so-called Kennedy-Anderson program in 1960 and continued the payroll tax then suggested as being appropriate….l think today you would find it would be about one-third underfinanced.
"Mr. Myers: It is possible that that is right. [Myers's subsequent memorandum confirmed Mills's view on this second bill as well.]
"The Chairman: What do you do with hospital costs? Do they remain constant, or do you contemplate in your estimates an increase in those costs?
"Mr. Myers: The cost estimates really can be looked at in either of two ways.
"Either it is assumed that hospital costs will remain level in the same way that wages do, or that, if they rise, eventually wages will catch up with them….[A]lthough in the past, hospital costs have been rising faster than wages, it seems reasonable to me to assume that over the long run these two elements will move at about the same rate.
"The Chairman: What is there in the program to give us any assurance that hospital costs can be so contained that they will not continue to rise by 3 or 4 percent more rapidly than earnings levels in these future years?
"Mr. Myers: This, as you point out, is matter of judgment."
Mills continued to press Myers on the subject, with a series of questions aimed at preventing Myers from wiggling out of the issue with bureaucratic equivocations. Myers finally had to admit that a doubling of the projected payroll tax rate was possible.
"Mr. Myers: My difficulty [with Mills's scenario] is that if the increase continued indefinitely, hospital costs would pyramid so high that they would be eventually as much as the total of all wages [i.e., the wage base against which the payroll tax would be levied], but let me try to make an assumption that is a little more finite in this respect.
"The Chairman: All right.
"Mr. Myers: Suppose that this situation went on for, let's say, 20 years and then the hospital costs caught up to general wages. Under an assumption like that, I think that what you said is quite correct—that the required combined employer-employee contribution rate would be somewhere around 1 percent of payroll, instead of 1/2 percent of payroll."
In 1965, Medicare architects declared that the initial tax rate of 1 percent of income after 1967 would be sufficient to fund the program for 25 years with only slight upward adjustments in the taxable income base of $6,600. (Income above that level wasn't taxed.) But it was clear at the outset that the fiscal projections for Medicare were essentially wild guesses, and the subsequent history of the program has shown little improvement in the ability to forecast the cost of the program.
The Medicare hospital-insurance program was already out of actuarial balance and about 50-percent underfunded by the early 1970s, requiring in 1972 the first of several increases in both the payroll tax rate and the wage base against which it was levied. The 1972 amendments to the Social Security Act also expanded what the Medicare program covered, including expensive renal dialysis.
Chastened by the high inflation of the early 1970s, the Social Security Administration developed new actuarial projections that substituted "dynamic" assumptions about wages and costs in place of the "static" assumptions of the 1965 projections. But these projections still expected that hospital costs would soon stop soaring at a faster rate than wages. "Public influence toward reducing the rate of increase in hospital expenditures is assumed in the cost estimates," says a 1973 Ways and Means Committee report. (This was at the time of President Nixon's so-called Economic Stabilization Program, better known as wage and price controls.)
Recent Hospital Insurance Trust Fund annual reports and forecasts for the Medicare hospital-insurance program are more sober but have drawn surprisingly little attention for how alarming they are. Forecasts for the last several years have employed three alternative scenarios for income growth over the next 75 years. Even under the most optimistic assumption, the Medicare payroll tax will have to nearly double, to 4.38 percent, shortly after the year 2000.
Under the pessimistic assumption, the payroll tax will have to soar to 13.79 percent for the program to stay solvent. The increases will come quickly; the tax will have to increase to 5.62 percent by about the year 2000. Under any of the assumptions, the Medicare trust fund will go broke within the next 20 years, possibly as soon as the year 2000.
But even the supposedly pessimistic assumptions behind these projections aren't realistic, and herein lies another small drama. Roland King, the chief actuary of the Health Care Financing Administration (HCFA), has been blowing big holes in the projections. King has to sign off on the "actuarial soundness" of the projections in the annual report each year. Acknowledging the "garbage-in, garbage-out" quality of the projections, King has noted in his statement that the projections assume unprecedented growth in average wages over the next generation.
The projections assume real income growth (pay increases adjusted for inflation) substantially above the trend line of the last 25 years. Calling this "unjustified optimism," King notes that even the "pessimistic" projection assumes that real income will grow at an annual rate that is faster than the cumulative rate for the last 25 years.
King concludes in a memorandum he circulates with copies of the Medicare annual report: "Indeed, the assumptions are so optimistic that even the pessimistic assumptions project real earnings increases during the next quarter century, and each quarter century thereafter, will be many times the increases of the last quarter century….The Trustees' assumption that real earnings growth rates will suddenly accelerate to levels that substantially exceed the real earnings growth rates of the last quarter century must be viewed as unreasonable." King believes that the actual actuarial deficit will be 60-percent higher than even the pessimistic forecast.
The Hospital Insurance Fund trustees have had a prickly reaction to King's observations. In last year's annual report, the trustees offered this grumpy dismissal: "We believe that the comments on real-wage gains by the HCFA Chief Actuary also represent an expression of a preference outside the bounds of the legally required actuarial opinion." In an appendix of the latest annual report, the trustees take aim at King again, but without offering any substantive refutation of his arguments: "It is perplexing and disconcerting that an actuarial opinion with unjustifiable qualifications has been allowed to be repeated for several years in the HI reports." In other words, shut up.
Nobody in Congress is paying much attention to this looming fiscal catastrophe. An HCFA official told REASON, "People keep thinking that if we just push back little by little the date of the trust fund's insolvency, we'll be all right." Congressional liberals thought it would make a big difference to more than double the taxable wage base—from $51,300 to $125,000—as a part of the 1990 budget deal with President Bush, but, as the HCFA official told us, "Some politicians were surprised to find that this will only push back the date of the trust fund's bankruptcy by about two years."
The other half of the Medicare program, Part B Supplemental Medical Insurance, is a similar tale of unexpected soaring costs. Its funding was, in theory, established on a more stable base than the hospital-insurance program. The supplemental program was to draw 50 percent of its revenue from insurance premiums paid by beneficiaries and 50 percent from federal matching funds from the general treasury. As program costs grew, however, adjustments were to be made in the monthly insurance premiums (which started out at $3.00 a month in 1966; today's Medicare part B premium is $31.80 per month) to keep the 50-50 ratio.
This plan ran into trouble by the early 1970s, again because the cost of medical care rose faster than income. Elderly enrollees, many of whom lived on fixed incomes, were paying an increasing share of their income for supplemental-insurance premiums. In the 1972 amendments, Congress stipulated that these premiums could rise no faster than Social Security benefits, regardless of program costs. Since 1972, the supplemental program has required ever larger subsidies from the general fund. The result: The Treasury now coughs up $3 for every $1 paid in premiums. In typical Orwellian doublespeak, Congress continues to describe the program as "self-supporting."
It seems never to have occurred to anyone during the debates about Medicare that increasing the demand for health services would generate huge pressures on hospital costs. The planners seem to have overlooked the fact that if you shift the demand curve sharply outward without moving the supply curve, prices will go up. The original 1965 cost projections allowed for a 10-percent increase in hospital-admission rates among the elderly, but in fact hospital-admission rates among the Medicare-eligible rose immediately by 25 percent, the rates for surgical procedure by 40 percent, and the number of hospital days by 50 percent.
The federal government has approached the problem of cost containment gingerly, not wanting in the early years to jeopardize the political support for Medicare within the health-care industry with stringent reimbursement policies or among the elderly by restricting benefits. But with the cost of the Medicare program growing at more than 15 percent a year (in 1975 Medicare outlays jumped a whopping 30 percent), reality has forced the government to try to impose cost containment. The parade of imaginary horrors the AMA predicted in 1965—such as price fixing for specific procedures and limits on doctors' fees—is finally coming to pass.
The results of cost containment, like the history of the fiscal projections for Medicare, have generally been disappointing and provide another sobering precedent for designers of universal national health insurance. In the 1970s, the federal government adopted the "Certificate of Need" program, which required states to review and approve capital investments of health-care providers. While correctly recognizing that Medicare and other government health programs provided perverse incentives for hospitals to increase their overhead by buying new capital equipment, the Certificate of Need program was a crude attempt at government rationing. Even the Congressional Budget Office, in a recent report, concludes that the program turned the capital equipment procurement process into a political process, dependent more on influence and political clout rather than on actual medical need.
A more serious strategy was adopted by the Reagan administration in the mid-1980s: the Prospective Payment System, which reimburses hospitals according to a fixed-price structure based on Diagnostic Related Groups. For example, if the average gall-bladder surgery costs $10,000 and requires four days of hospitalization, that is what the government will pay the hospital for a Medicare patient, regardless of how long that particular patient may need to stay. During the aggressive implementation of this policy in the mid-1980s, the cost of Medicare's hospital-insurance program grew at single-digit rates for the first and only time in its history.
The attempt to contain costs of Medicare's supplemental insurance program by freezing physician fees from 1984 to 1986 was less successful. That program cost continued to rise by more than 10 percent a year. A Congressional Budget Office review of the fee freeze and other attempts to fix prices notes that doctors adapt simply by increasing volume. A 1991 CBO study concludes: "Studies of the effects of fee freezes or controls on physicians' prices indicate that they result in a pronounced volume offset." The same inherent weakness diminishes the long-term prospects that the PPS/DRG system can reduce the rate of health-care cost inflation.
Hospitals have adapted to the system by qualifying patients for more tests and procedures, by shifting patients to different DRG designations that carry a higher reimbursement rate, and especially by shifting patients from hospital to outpatient settings, which shifts the cost to the supplemental-insurance program. An HCFA official told us: "There is simply no way to contain costs, and nothing on the horizon to contain costs."
The lesson of both PPS and the physician-fee cap is that government efforts to control health costs through price fixing are like trying to press the bumps out of a tube of toothpaste. Press in one area, and the tube bulges in another. To smooth out all the bumps by force, you have to apply pressure to all areas of the tube. The cost-containment logic of fixing prices for government health programs is leading inexorably to government price fixing for the entire health-care system. It is already happening in two ways.
First, private health insurers, wary of having costs shifted onto them, are adopting the Medicare PPS rates as their own reimbursement policy. "Over time," says Anthony Masso, director of managed care for the Health Insurance Association of America, "you will probably see [the Medicare reimbursement rates] being used by almost all private-sector insurance carriers."
Second, the latest attempt at Medicare cost containment is an even more ambitious price-fixing scheme that not only restricts the amount the government will pay for hospital procedures but even sets fees individual practitioners should receive in hospital and clinical settings. In the best socialist style of centralized decision making and price fixing, HCFA has commissioned a multimillion-dollar study to develop a ranking of the "relative value" of medical procedures on which to base the new reimbursement rates.
Having learned from the failure of the fee freeze in the mid-1980s, the government is attempting to head off the possibility of increased volume by reducing physician-reimbursement rates by as much as 30 percent in some cases. But instead of increasing volume, this time doctors and clinics are cutting back on their Medicare practices. A 1990 survey by the Association of American Physicians and Surgeons found that nearly 50 percent of physicians would respond to a cut in Medicare fees by seeing fewer Medicare patients.
"Private doctors," said Dr. Jane Orient, a past president of AAPS, "if they remain in business at all, will probably be cutting back on their Medicare practice, or dropping it altogether."
Under a universal "all payer" health-insurance policy, the Medicare price-fixing policy would be applied to the entire health-care system. The fears the AMA expressed at the time of Medicare's enactment 25 years ago that it would lead to socialized medicine seem fated to be realized.
As the nation barrels toward some kind of universal health-care policy, there is an ominous political parallel shaping up between 1965 and 1993. What finally tipped the balance in favor of Medicare in 1965 was the huge turnover in Congress. The next Congress will have more than 100 new House members. Despite some losses, the Democrats will have a solid majority, and even some Republican congressional candidates adopted as boilerplate the slogan that the federal government must take a role in providing some kind of universal health coverage. Bill Clinton has vowed to make health care a top priority, and Congress isn't likely to hesitate much over the legacy of Medicare.
In light of the alarming actuarial projections for Medicare's hospital-insurance program, it is no wonder that politicians are loathe to propose a national health plan that explicitly includes a new payroll tax. But the halfway house of "play or pay" that Democrats are advocating will quickly fall into the same trap as Medicare.
First, Medicare is simply one example of the fact that, as an HCFA official put it, the federal government "continually underestimates the cost of social insurance programs." This problem would be compounded under a "play or pay" scheme because the alternative payroll tax would initially be set low to entice participation.
As a consequence, the government would end up with a disproportionate share of high-cost enrollees (similar to what has happened with the state-run "assigned risk" pools for auto insurance) because companies with high health-insurance costs would find it to their cost advantage to dump the burden onto the government program. Thus, the "pay" portion of the government program would certainly be underfunded. "It is guaranteed to increase the federal deficit," says a source at the HCFA. "It is impossible to set a payroll tax that will cover the cost."
A recent study by the Urban Institute found that if the payroll-tax rate of a "play or pay" policy were set at 7 percent, 112 million nonelderly people, or 52 percent of the insured population, would be covered through the "pay" option. This number includes 59 percent of all private-sector workers. If the rate is 9 percent, an estimated 84.8 million would still be enrolled in the public plan, or 39 percent of the nonelderly insured population. These numbers suggest it will be a massive, and massively underfunded, program.
Providing significant funding through cost containment is equally chimerical, unless the government is prepared to administer an onerous "all-payer" regime of price controls on the entire health-care system, and then employ a rigorous scheme of rationing on top of that. Although a few liberals speak openly of price controls and "all-payer" regulations, they are loathe to entertain the necessity of rationing because of the inequities it would generate.
There is one decided difference between the politics of Medicare in 1965 and today. In 1965, all the opponents of Medicare had to offer was a watered down version of that program, Medicare on the cheap. Today, opponents of increased federal involvement in health care have a true market-based alternative to offer. The combination of "Medi-save accounts," in which people would be encouraged to save their own money for health insurance and direct payment of medical expenses, and tax credits to enable low-income families to purchase health insurance, offer a viable alternative to centralized big-government health insurance.
Although this isn't the magic answer to soaring health-care costs, it is a policy that rests on the one proven strategy for holding costs down: making the individual responsible for first-dollar costs of health care. If the critics can portray the issue as big government versus individual empowerment, there just might be a chance for positive reform of the health-care mess.
Contributing Editor Steven Hayward is research and editorial director for the Pacific Research Institute in San Francisco. Erik Peterson, a law student at George Mason University, is a researcher at PRI. REASON intern Grant Thompson provided research assistance for this article.