What If Technology Never Stops Improving? Medicare’s Future Under Continuous Cost Increases

What If Technology Never Stops Improving? Medicare’s Future Under Continuous Cost Increases

Pauly, Mark V

I. Introduction

In discussing the future of Medicare, it is commonplace to remark on the ominous demographics that will begin to affect the financing of the program as the baby boom generation retires. Even after allowing for immigration, the population of beneficiaries is projected to grow at twice the rate of the workforce after the Depression-era Baby Bust works its way through, and temporarily brightens, the trust fund balance.1 The number of under-sixty-five worker-taxpayers per beneficiary will fall nearly in half by mid-century,2 and so financing of both the Part A trust fund and Part B general revenues will be a challenge, and there may have to be “cuts” in some aspects of the program.

This gloomy conventional wisdom, however, misses a major point. The fact is that, adjusting for economy-wide inflation, there is very strong reason to believe that the Medicare program could avoid cuts in benefits and still be able to finance the cost of the baby boom generation with little or no increase in taxes. The increasing productivity of workers and the associated economic growth could, in principle, offset the demographic time bomb. The trend on taxable Medicare payroll per worker has inched upwards over the 1990s because of the expansion of the tax base to all non-capital income and the striking improvement in the productivity of American workers.3 Improvements in tax base per worker are, if anything, even stronger for the general revenues that fund Part B because those taxes could tap both profits and capital gains as well as higher worker wages.4 Roughly speaking, estimates of tax base growth per worker have increased to 1% per year, up from forecasts of 0.7% or less in the early 1990s.5 Because the Medicare population will grow at about 2% per year after 2015 and the worker population will grow at about 1%, the trend in tax base per worker is almost exactly what is needed to offset the difference.6 One should perhaps add an additional adjustment for growth in medical input prices in excess of overall prices (or wages), but this addition is relatively small on a per-year basis and is sustainable with either good luck in economic growth or modest tax rate increases.7

Why, then, is there so strong a belief that the sky will eventually fall? The answer is that it seems quite unrealistic to forecast zero growth in real Medicare spending. Historically, real expenditures (deflated by the CPI) per beneficiary have grown at rates of 4% to 5%.8 Even the seventy-five-year projection by Medicare’s actuaries puts the rate at 3%, 1% faster than the rate of growth of real Gross Domestic Product per capita.9 The new technology is presumably beneficial, at least in the aggregate.10 Cost trends in total spending are similar for those over and under age sixty-five, with slightly slower Medicare growth attributable primarily to lower growth in Medicare physician payments.” If Congress included prescription drug coverage (the segment of medical spending with perhaps the steepest increase in technology and (in recent years) spending)12 as a covered benefit, then Medicare spending growth per beneficiary that accommodated new technology would probably be at or above the top of its historical range.

The real policy questions for Medicare in the long run then are: (1) What rate of future growth in technology and associated cost do the citizens want Medicare to achieve?; and (2) can Medicare feasibly raise the resources to fund that growth? A benchmark here would be the rate of growth in the private sector, so another version of the question this Article addresses is whether one can expect Medicare to continue to keep up with the private sector in both technology and cost.

I will argue that, as currently structured, Medicare definitely should not, and probably will not, be able to keep up with the private sector given plausible political models. Instead, monetary concerns will compel Medicare to adopt a policy more similar to Medicaid in those states that are budget constrained, with either explicit administrative decisions to ration out beneficial but overly costly technology, implicit decisions about reimbursement that have the same effect, or a delegation of that dirty job to contracted private insurers, HMOs, and other entities.13 In contrast to the Medicaid population, however, the elderly are not predominantly poor, so limiting middle class people’s access to technology to the same extent as indigent people’s access also may not be stable.14 There is no painless solution to this dilemma. This Article will discuss what I regard to be the least painful and most graceful types of solutions.

II. Medicare and Private Insurance Spending Trends

The rate of growth in total spending, especially the rate of growth in spending per insured person, amongst private insurance and Medicare has historically been quite similar.15 Although there are periods when one or the other grows more rapidly, over time these deviations disappear.16 The similarity between Medicare and private insurance growth in the first eighteen years of Medicare should be no surprise because Medicare had the same type of coverage and the same type of provider payment system as did private insurers.

More recently, there has been a divergence. The private sector has moved much more aggressively toward managed care, while the traditional Medicare plan is wielding its enormous economic and political power to set administered prices for most services.17 The shift to managed care in the private sector led to lower private insured spending growth in the first half of the 1990s, but Medicare’s exercise of its buying power in carrying out the Balanced Budget Act (BBA) has led to slower Medicare growth recently.18 In addition, the absence of Medicare coverage for prescription drugs has shielded the Medicare plan (but not the Medicare population) from the rapid growth in drug spending.19

Near-term prospects for low Medicare spending growth (excluding the effects of any prescription drug program) are fairly bright. The growth rate in the elderly population is one of the lowest in years as the smaller late-Depression cohort moves through.20 Moreover, the spending growth reductions in the BBA, though already watered-down to some extent, will continue to moderate growth, especially in Part B.21 Additionally, the campaign against Medicare fraud has doubtlessly scared some providers into requesting less than they deserve.22

This period will correspond to one of positive growth in the accounting entry known as the Hospitalization Insurance Trust Fund (unless proposals to cut payroll taxes as taxes for the non-rich come to fruition).23 But there is virtually unanimous agreement that this situation cannot last. Eventually, probably by 2020 if not sooner, expenditures will quickly exceed revenues by a very large amount.24 In Part B the spending growth will rekindle even sooner.25 Thus tax increases will eventually be required to maintain benefits. Depending on the assumption made about the growth of spending, some analysts see a relatively moderate tax increase, but most commentators, and some Medicare trustees, see a substantial tax increase in the offing.26

III. What’s So Bad About High Taxes?

Currently, the tax for Medicare Part A is 2.9% of payroll (combining the employer and employee shares because workers ultimately pay both).27 General revenue taxation finances three quarters of the cost of the Part B program, and premiums beneficiaries pay (set at $58.70 for 2003) finance one quarter.28 Part B spending has been growing. A shift to ambulatory care on the part of beneficiaries and a shift of home health spending to Part B by politicians has augmented spending.29 Calculated as a percentage of payroll, the Part B tax probably amounts to another 1.5 to 2%.30 This total Medicare tax, now less than 5% of payroll, is likely to at least double, and possibly triple, by approximately 2030.31 Combining this with a payroll tax for Social Security that is expected to increase moderately, the percentage of payroll going solely to social insurance could be in the range of 25% to 30% by that time.32 It will surely rise from today’s combined rate of a little over 15% to at least 20%. These social insurance payments, largely taking the form of transfers, primarily represent redistribution of wealth within the economy, from workers to seniors. Because those seniors are parents and grandparents of workers, and because there is social concern about the well-being of all elderly citizens, these transfers are not wasteful-but they are difficult.

Voluntary premiums paid by or on behalf of beneficiaries will not finance the great bulk of this increased spending.33 Rather, about 90% of funding will come from compulsory taxes, either in the form of payroll (actually non-capital) income taxes for Part A or general revenue taxation for Part B.34 Although the Part A taxes are earmarked for hospitalization insurance, at the individual level (in contrast to Social Security) they bear no direct relationship to the benefits the individual would expect to collect.35 In the case of general revenue taxation, the disconnect is complete.

Economists argue in theory, and empirical research shows, that this disconnect has a cost. Because a person’s tax burden rises in relation to the higher values of the tax base (more aggressively so if the tax is progressive), people have an incentive to make choices that reduce their share of the tax base.36 There is a whole panoply of distortions, ranging from excessive do-it-yourself repairing to sophisticated financial manipulations, that only occur because of the taxes on financial income that behavior can alter. Of course, if total tax collections are set in a budget, these distortions do not, in the aggregate, reduce taxes; rather, they add an additional cost to any funds collected through taxation.

Economists term this phenomenon the “excess burden” or “deadweight loss” of taxation.37 It is well documented that taxes do affect behavior, although the magnitudes are sometimes measured imprecisely. Measures of this excess burden (or “tax on taxes”) are known to be positive and thought to be relatively large (if also imprecise).38

Not only is it more costly to raise funds for the public sector through compulsory taxation than through voluntary purchases in the private sector, but the excess burden per dollar of tax collected multiplies with the tax rate.39 A 1% increase in the tax rate costs much more at high overall tax rates than at low rates; therefore, the excess burden of a tax increase for something specific like new technology for Medicare beneficiaries depends on the mix of taxes Congress uses and the level of tax rates overall. Nevertheless, recent research suggests that the excess burden as a percentage of taxes collected in the United States with today’s level of tax burden (which rose during the 1990s) is probably at least 30% and perhaps more.40

Although the research on the existence of the excess burden of taxes is definitive, research on what it should or does mean for tax-financed spending decisions is much more ambiguous. The common-sense intuition-if something costs more, it should be bought less-may still come through, but must this happen, and if it must, how will it happen? And if the excess burden cost per dollar of revenue collected rises over time as Medicare and Social Security tax rates rise, how would and should public spending respond to this ever-growing price?

IV. Financing New Medical Technology in Old Age

Although the significant intergenerational redistribution Medicare accomplishes somewhat attenuates its precision, the fundamental question is simple: How do or should people pay for what they want and need in their old age when one important component of consumption-medical care-will be increasing in both quality and cost? For private goods (retirement housing is a good example), the economically-motivated agent knows what to do: forecast the cost and consequent amount you want to spend after retirement, consume less and save more to accumulate funds for that expense, and then dissave and consume after you retire.

The source of the problem is the socialization of medical spending. Had we set up a tax-financed program to pay for retiree housing, we would be as upset about higher housing spending as we are about rising medical spending and as gloomy about prospects for improvement in access to improving quality housing in the future.

Socialization of medical insurance provides substantial benefits to accompany its undoubted excess burden costs. But those costs are real. In addition to the excess burden of taxation, the fact that an anticipated availability of public insurance obviates the need for private saving tends to diminish the incentive to save and thus adds another kind of distortion.41 Less tangibly, but more importantly, the necessary and creaky process of public sector decision-making in a bureaucracy also adds a potential cost.42

To fully understand the issue, one needs a brief sketch of what might happen and what should happen. Because this is politics, it is even more guess work than economics. Begin by assuming that spending will grow in Medicare at the same pace as in the private sector. Given the certain increase in beneficiaries relative to workers and the economy, higher levels of public financing will be required. What will that kind of financing mean? One possibility is to increase taxes to higher levels than at present. These tax increases mirror the already higher premiums for employment based insurance.43 Even if the burden of Medicare taxes rises to 10% of salaries, so that, in combination with increased Social Security taxes, fully 30% of a worker’s income or salary is taxed for transfers to seniors,44 the overall U.S. tax rate depends on what happens to other public spending, tax policy, and fiscal policy.

Such a tax increase would, at a minimum, cause political distortions, possibly along generational lines and political recriminations, including blaming providers of care (or anyone else in the neighborhood) for inefficiency. Another possible reaction would be a reduction in other government spending. The cost of this activity is the foregone benefit from the curtailed expenditures, which also could be substantial.

If political decision-making were rational, the optimal reaction would be to limit Medicare spending instead. Strictly speaking, the cost to Medicare of paying for all technology, new and old, will have risen. Therefore, the government should, to some extent, reduce all benefits. But because it may be difficult to take away benefits already provided, it will be both more preferable and more likely that some new technology will be cut relative to what Medicare has done in the past. This cut can take the form of either delay, omission of adoption, or imposed limits on the quantity of use.45 The key point here is that the Medicare payment for new technology should be cut, and my assumption is that, sooner or later, the technology somehow will be cut. This does not mean that total Medicare spending actually will be reduced or that technology will be lost. It only means that the rate of addition of beneficial, but costly, new technology in Medicare will have to be cut.

Cut relative to what? Although the case is not airtight, a benchmark for cuts is likely to be coverage for people under age sixty-five (and a good comparison is likely to be people near that age with private coverage). If the growth in income, relative prices, and taxes are approximately the same in both sectors, the comparison will be most apt.

Because some people think that the objective of the Medicare program is to provide seniors with coverage equivalent to that provided privately for people under age sixty-five, there is likely to be conflict-the efficient course of action, which is to limit spending more in the coverage with higher marginal cost, will clash with the postulated objective. Something will have to give; while there is no necessity (or even expectation) that political decisions will be efficient, the inevitable high fiscal pressure of ever-increasing Medicare spending and taxes will be hard to resist. It would be useful to examine the feasible and logical goals for the future of Medicare and discuss policy while we still have time to adjust.

V. What’s Ahead for the Private Sector?

So what is this private sector growth likely to be? The private sector is currently under fire for outrageous premium increases, provoking employers’ private efforts to curtail spending.46 Although there is currently much talk and a little action toward more aggressive private cost containment, the growth rate is still high relative to its long term trend and certainly high relative to GDP growth.47

At this point in time, the only thing we can say is that the newer health insurance plans, primarily involving greater patient cost sharing, will at best reduce spending and improve satisfaction (compared to the managed care alternative) as a “one-time” event.48 There is no evidence that any of these plans are going to be more effective over the long run in holding down the rate of addition of costly new technology.49 Therefore, unless there is a prolonged and severe period of macroeconomic downturn, it is most plausible to assume that privately insured spending will grow at approximately its long term rate, which is somewhat in excess of the rate of growth of real GDP. One of the substantive reasons for expecting a modest tailing off in spending growth is that a major driving force for growth, a reduction in the percentage paid out-of-pocket, cannot proceed much further for hospital care as inpatient and outpatient care approaches 100% insurance coverage (for those who have insurance).50 The other reason is that growing premiums tend to cause some people to become uninsured and then to spend and cost less.51

Nevertheless, my expectations are that the share of the privately uninsured population’s income devoted to medical insurance will continue to increase moderately and that with, at most, a more intensive scrutiny of new technology than in the past to rule out useless or very low-value “lifestyle” technologies, insurers will continue to supply new technology that is costly but beneficial. Yet, even though health plans do evaluate new technologies,52 no plan has declined to cover enough of the costly new technology to offer lower premium growth to consumers. That is, there is no market evidence that consumers want a less inflationary “low tech” health plan that sacrifices beneficial but costly new technology in order to hold down premium growth.

VI. Can Medicare Stay in Step?

What can and should Medicare do to deal with new technology in this environment when its real cost of adding that technology rises above that of private sector plans? The first, and in some ways the most difficult, question to answer is whether realization of this unavoidable higher Medicare cost will penetrate the collective decision-making process.

On the one hand, citizens’ (especially those in the middle class, who now dominate the elderly demographic) demands for new technology will not disappear just because people turn sixty-five. The elderly and their lobbyists, for good reason, will continue to value and demand new technology.

Medicare has traditionally dominated private insurers in decisions on technologies directed at the illnesses of older people.53 Although I would not rule out a temporary continuation of this trend at a substantial cost in spillover effects on private insurance, my forecast is for eventual divergence. The only alternative is severe financial distress, not just for Medicare, but for Social security and, eventually, for the ability of the federal government to provide public goods in total.

There are other political advantages for Medicare over private plans that may delay this difficult choice. Because it is virtually compulsory and offers a uniform package, Medicare has lower administrative costs than private insurers who offer voluntary plans with greater variety.54 This advantage, however, is of the “one-time” character and need not extend to coverage of costly new technologies.55 The marginal cost of administering benefits for new technology is unlikely to be lower for Medicare than for private insurance.

Perhaps more importantly, the evidence of the BBA suggests that the traditional Medicare plan (“Old Medicare”) has substantial monopsony power which it can use to limit its expenditures.56 Recent increases in payment rates to hospitals and doctors, however, suggest that this “buying” power is limited politically, if not economically. The desire on the part of some to avoid creation of monopsony power for pharmaceutical coverage is one reason why the design of such a benefit has been such a difficult matter for consensus.57 Again, however, it is likely that any monopsony advantages are one-time.

Is there a political template for a government-operated insurance that rations new technology? Of course there is; it is the federal-state Medicaid program. Most Medicaid rationing is implicit.58 For instance, unusually low provider payment rates restrict the supply of higher quality care that embodies new technology. Such indirect rationing by (low) price is more politically acceptable than explicit rationing by clinical, demographic, or social criteria. Is it possible that Medicare will come to behave more like Medicaid, using low provider prices and selective coverage to limit technology and cost growth?

VII. How Might/Should Medicare Ration Coverage of New Technology?

At one level, the task of constraining Medicare coverage and the cost of new technology is simple. Reasonably well-developed technology evaluation methods for determining or forecasting the effectiveness and cost of new technology exist.59 A number of managed care plans and countries implement, to some extent, these methods for coverage determination.60

Nevertheless, despite several failed or stalled attempts, Old Medicare has not implemented transparent and definitive coverage standards. The rationale under which Old Medicare permitted reimbursement was that the new technology would be better and no more expensive. Congress has not approved, much less implemented, the practice of explicitly denying coverage for new technology with positive cost and benefits because the value of benefits was less than the cost.61 Its position is that “there is no dollar value at which Medicare would decline to pay.”62 What accounts for this failure, and what lessons does this failure offer for a future in which the goals of technology assessment will become many times more important than they have been in the past?

Susan Foote has recently addressed this question, attributing the Medicare “regula mortis” to successful industry opposition.63 Although one would expect those firms whose products would not be covered under a more restrictive coverage policy to oppose that policy-how could one expect otherwise-this explanation does not tell us why opposition has been successful. I think there is more to the issue than business as usual.

First, politically, one should expect current Medicare beneficiaries with legendary political clout to oppose limits; in this matter the elderly and the investors are natural allies. Foote also notes Medicare’s interest in lower costs,64 sometimes attributed to the “trust fund,” even though much device-related technology is potentially covered under Part B, which has no trust fund.65 My guess is that, on a routine basis, the concentrated interest of beneficiaries trumps the diffuse interest of taxpayers that Medicare’s governmental leadership weakly represents.66

More importantly, no model of “transparent and definitive” coverage standards with face validity and intellectual depth exists. It is therefore easy for industry to oppose a half-baked policy that assumes a consensus on facts that does not exist. It is easy to challenge standards based on flimsy or tortured logic.

More specifically, the cost effectiveness model, which could serve as the basis for standards, requires agreement on the monetary value of health (as life years added and quality of life) and on beneficiary convenience and satisfaction. None of these as-yet-unsuccessful efforts to promulgate methods has even begun to establish that value. Instead, beyond the obvious and brave stance against dominated options, the approach has been one which relies on indefinable adjectives: A technology that raises cost but does more good must do a “reasonable,”67 “appropriate” or “significant,”68 or “substantial”69 amount of good-but how much is that? Both the 1989 draft rules70 and the 2000 Notice of Intent71 will be insufficient in the future when limitations must be stricter than before and stricter than in the private sector. Medicare does not know how to limit the growth in spending that arises from new technology.

In the absence of this straightforward, cost-effective approach, Old Medicare has proceeded in two ways. One way is by ad hoc muddling through, with much of the heavy lifting fobbed off on a consensus of carriers assisted by clinicians who do not have the foggiest idea of how much health is worth.72 This approach, despite its long history, cannot provide the discipline that will be needed in the future.

The other approach is the opposite of evaluating new technology and then examining its costs. Instead, this approach sets a rate of growth in spending based on other more-or-less arbitrary benchmarks (a “sustainable growth path” based, for example, on GDP growth), accepts whatever new technology that funding will pay for, and considers the value of new technologies that go over the limit when, and if, they arise and cause an outcry. In any of these cases, however, the future will eventually require a more clearly specified, rigorously administered, and persuasively argued approach.

Foote suggests some possible administrative solutions. She imagines that Congress will itself define criteria and set limits.73 I cannot see Congress voting on the value of life. Another possibility is negotiated rule making.74 But who will represent the taxpayers? A third approach is to find precedent in previous decisions-a common law approach.75 This does not help if the previous period was one of unsustainably permissive approval.

A fourth possibility is to ignore the explicit problem and attempt to control spending in ways less obviously directed at new technology. For at least the next five years, Medicare’s current-period finances are actually quite bright; the Baby Boom does not replace the Depression-era Baby Bust until then. In the meantime, Medicare could indirectly ration technology by limiting, at least, the most expensive variants to various types of “Centers of Excellence,” especially if (as is likely) the most expensive new technology is also complex and dangerous in the wrong setting. The political pressures on Medicare to pay enough so that every provider a beneficiary would choose can render any service will, however, limit this strategy. In addition, the political pressure from local providers on their federal legislators to provide broad and equal access will further limit this strategy.

To a considerable extent, this type of clinically based limitation on technology depends on the new technology being strongly characterized by a positive relationship between volume and outcome, the so-called “practice makes perfect” phenomenon. Although this relationship holds for some technologies, it does not hold for all. Moreover, suppliers of technology that are interested in having a wide market may alter the form of technology to make it easier to use and less risky at low volumes.

In short, although there is some hope for such a clinically-motivated de facto PPO strategy for Medicare, it cannot be a complete or permanent solution. Muddling through may work for a decade or so, but not forever.

VIII. Some Market Alternatives

Most of the elderly are not poor (though they are not by any means rich).76 One possible policy would be to require them to finance some of their own new technology. Medicare could continue to make a contribution toward coverage that would be enough to maintain quality (or allow it to increase a little) and even offer a policy that provides that coverage. But non-poor beneficiaries that wished to do so would pay for quality improvements over and above what Medicare pays. Beneficiaries could supplement the public payment if desired.

Current Medicare payment policies make supplementation very difficult. Both Part A and Part B are subject to limits on balance billing so that payments for costly activities require that the physician classify the treatment as medically useless and entirely at the patient’s discretion. In contrast to people with private insurance, people on Medicare cannot pay with their own money for something that is more medically valuable to them than it is to the Medicare bureaucracy.

This approach could also permit greater competition that might curb cost. I am profoundly skeptical that such an outcome will emerge. This arrangement, however, makes it possible.

IX. Conclusion

Medicare faces a difficult future, one likely to be far more difficult over the long run than that of private insurers. Not only will Medicare’s financial needs increase if it continues to add coverage for beneficial but costly technologies, but the economic distortion and political opposition its higher cost growth will engender will be troublesome and will eventually require drastic reductions in the lavishness of “Medical Christmases” in which Congress adds new products. Medicare will, to some extent, become more similar to Medicaid than to Blue Cross. The key issues are whether supplementation will be made easy and whether the private sector will be allowed to lead.

1. See U.S. DEP’T OF HEALTH & HUMAN SERVS., 2002 ANNUAL REPORT OF THE BOARDS OF TRUSTEES OF THE FEDERAL HOSPITAL INSURANCE AND FEDERAL SUPPLEMENTARY MEDICAL INSURANCE TRUST FUNDS 18 (2002) [hereinafter MEDICARE TRUSTEES] (stating that there are forty million Medicare beneficiaries to 157 million workers in 2001, whereas there will be seventy-seven million beneficiaries to 182 million workers in 2030), available at http://www.cms.hhs.gov/publications/trusteesreport/2002/tr.pdf (on file with the Washington and Lee Law Review); U.S. DEP’T OF HEALTH & HUMAN SERVS., THE 2002 ANNUAL REPORT OF THE BOARD OF TRUSTEES OF THE FEDERAL OLD-AGE AND SURVIVORS INSURANCE AND DISABILITY INSURANCE TRUST FUNDS 80 (2002) [hereinafter OASDI TRUSTEES] (finding a ratio of thirty-five million beneficiaries to 168 million workers in 2001, with a predicted ratio of sixty-nine million beneficiaries to 195 million workers in 2030), available at http://www.ssa.gov/OACT /TR/TR02/tr02.pdf (on file with the Washington and Lee Law Review).

2. See MEDICARE TRUSTEES, supra note 1, at 18 (finding a ratio of 4.0 workers per beneficiary, but noting that there will be only 2.4 workers per beneficiary in 2030).

3. See id. at 39, 48 (showing that the $ 135,000 maximum tax base ceases to exist after 1993 and that total income has increased from $79 billion in 1990 to $171 billion in 2001).

4. See id. at 1, 6, 10, 29 (noting that Part B is funded through general revenue taxes, a base that is stronger than Part A).

5. See OASDI TRUSTEES, supra note 1, at 90 (finding an average annual percentage increase in real wage differential between 1.0 and 1.2 for years 2005 to 2080); MEDICARE TRUSTEES, supra note 1, at 6 (finding an intermediate assumption of 1.1% real wage differential).

6. See MEDICARE TRUSTEES, supra note 1, at 35 (finding that the Medicare population will grow from fifty-two million in 2015 to seventy-seven million in 2030); see also OASDI TRUSTEES, supra note 1, at 80 (finding that the working age population will grow from 191 million in 2015 to 195 million in 2030).

7. See Joseph P. Newhouse, Medical Care Costs: How Much Welfare Loss?, 6 J. ECON. PERSP. 3, 9-11 (1992) (explaining that medical expenses inflation was not a significant factor in increased medical costs).

8. See Christine Boccuti & Marilyn Moon, Comparing Medicare and Private Insurers: Growth Rates in Spending over Three Decades, HEALTH AFF., Mar./Apr., 2003, at 230, 231 (finding an average annual growth rate, since 1970, of 9.6% before adjusting for inflation.); Gail Wilensky & Joseph Newhouse, Medicare: What’s Right? What’s Wrong? What’s Next?, HEALTH AFF., Jan./Feb. 1999, at 92, 95 (“From 1970 to 1997 the real growth rate of Medicare spending was about 5.4% per beneficiary per year.”).

9. The lion’s share of this growth is almost surely new technology and not additional “waste” or increases in input prices. MEDICARE TRUSTEES, supra note 1, at 36.

10. See David Cutler & Mark McClellan, Is Technological Change in Medicine Worth It?, HEALTH AFF., Sept./Oct. 2001, at 11, 24 (finding that between 1950 and 1990, present value of per person medical spending has increased by $35,000, life expectancy has increased by seven years, and mortality rates have dropped among low-birth weight infants and sufferers of acute heart attacks).

11. See Boccuti & Moon, supra note 8, at 231, 234 (finding that, since 1970, Medicare has an average growth rate of 9.6%, which is lower than the growth rate of 11.1% for private health insurers, and that per enrollee payments for physicians and clinical services grew higher for private insurers than Medicare in the late 1980s and early 1990s).

12. See id. at 233 (noting that prescription drug spending is a “component of health care spending that has grown at an extremely rapid pace”).

13. See KANT PATEL & MARK E. RUSHEFSKY, HEALTH CARE POLITICS AND POLICY IN AMERICA 70 (2d ed. 1999) (laying out the implications of Medicaid spending rules).

14. See Mark V. Pauly, Should Medicare Be Less Generous to Higher-Income Beneficiaries?, in MEDICARE REFORM 65, 68 (Andrew J. Rettenmaier & Thomas R. Saving eds., 1999) (discussing the increase in family income among the elderly since Medicare’s inception and noting that 60% of elderly households now have incomes above 200% of the poverty line).

15. See Boccuti & Moon, supra note 8, at 234 (comparing spending growth between Medicare and private insurers for like services and noting the similarity).

16. See id. at 231 (explaining that cumulative analysis smoothes out annual fluctuations).

17. See id. at 235 (noting that Medicare’s cost containment strategy involved hospital payment reforms, whereas private insurers turned to managed care because they do not enjoy the same level of market clout).

18. Id.

19. See id. at 233-34 (stating that Medicare spending “is less affected by this component because it does not offer an outpatient drug benefit”).

20. See OASDI TRUSTEES, supra note 1, at 80 (demonstrating in a chart that, for the years 1995-2005, the growth rate increase in the elderly population is lower than in previous or subsequent years).

21. See Katherine Levit et al., Trends in U.S. Health Care Spending, 2001, HEALTH AFF., Jan./Feb. 2003, at 154, 161 (discussing recent trends in Medicare spending).

22. See Timothy Stoltzfus Jost & Sharon L. Davies, The Empire Strikes Back: A Critique of the Backlash against Fraud and Abuse Enforcement, 51 ALA. L. REV. 239, 257-58 (1999) (“Fraud and abuse enforcement officials claim that . . . providers are becoming more cautious in their billing because of concern about fraud and abuse enforcement and that this caution is having an effect on program expenditures.”).

23. See MEDICARE TRUSTEES, supra note 1, at 18 (explaining that under intermediate assumptions the Hospitalization Insurance Trust Fund assets will increase until the year 2015).

24. See id. (noting that under intermediate assumptions the hospital insurance fund assets would begin to decline relative to annual expenditures after the year 2015).

25. See id. at 26 (charting estimated future Supplementary Medical Insurance Trust Fund expenditures); Wilensky & Newhouse, supra note 8, at 93 (noting that the rapid rate of Part B spending growth is a factor motivating Medicare reform).

26. See Andrew J. Rettenmaier & Thomas R. Saving, Another Medicare Monster, WALL ST. J., June 24, 2003, at Al 6 (noting that, even without a new prescription drug benefit, Medicare’s claim on income taxes will double by 2020 causing tax increases unless other federal spending programs are cut back).

27. See 26 U.S.C. §§ 3101(b), 3111(b) (2000) (imposing a tax on employers and employees of 1.45% of wages). In competitive labor markets with a relatively inelastic supply of labor, the incidence of employer payroll taxes is primarily on money wages because pay must fall when taxes rise if equilibrium is to be reattained in the labor market. For application of this argument to the similar case of mandated employer-paid benefits, see MARK V. PAULY, HEALTH BENEFITS AT WORK 43 (1997) (explaining that “if the willingness of workers to work is largely unaffected by decreases in money wages accompanying the provision of employer payment for health insurance, the cost of a mandate will fall almost entirely on workers in the form of lower wages”).

28. See 42 U.S.C. § 1395(r), (j) (2000) (establishing that Part B is “to be financed from premium payments by enrollees together with contributions from funds appropriated by the Federal Government” and stating that the premiums shall be set at 50% of the monthly actuarial rate needed to cover one-half of the total benefits and administrative costs expected to be payable from the fund); Monthly Actuarial Rates and Monthly Supplementary Medical Insurance Premium Rate Beginning January 1, 2003, 67 Fed. Reg. 643 (Oct. 21, 2002) (setting the premium amount for 2003 at $58.70).

29. See MEDICARE TRUSTEES, supra note 1, at 46 (stating that coverage for certain home health services was transferred from the Hospitalization Insurance Trust Fund to the Supplementary Medical Insurance Trust Fund in 1998).

30. Part B expenses are approximately 60% of Part A expenses in 2001. For the source of the calculations, see CTRS. FOR MEDICARE & MEDICAID SERVS., U.S. DEP’T OF HEALTH & HUMAN SERVS., TABLE 10: EXPENDITURES FOR HEALTH SERVICES AND SUPPLIERS UNDER PUBLIC PROGRAMS, BY TYPE OF EXPENDITURE AND PROGRAM: CALENDAR YEAR 2002, at http://www.cms.hhs.gov/statistics/nhe/historical/tables.pdf (last visited Jan. 27, 2004) (on file with the Washington and Lee Law Review).

31. See MARTIN FELDSTEIN, PREFUNDING MEDICARE 1 (Nat’1 Bureau of Econ. Research, Working Paper No. 6917, 1999) (explaining that according to the 1998 projections of the Congressional budget office, government Medicare spending net of premiums individuals pay will rise from 2.5% of GDP now to 5.5% of GDP in 2030).

32. OASDI TRUSTEES, supra note 1, at 168-69.

33. See supra notes 27-28 and accompanying text (explaining that taxation finances all of Part A revenue and three quarters of Part B revenues).

34. See id. (explaining that taxation finances all of Part A revenue and three quarters of Part B revenues).

35. See MARK MCCLELLAN & JONATHAN SKINNER, THE INCIDENCE OF MEDICARE 2 (Nat’1 Bureau of Econ. Research, Working Paper No. 6013, 1997) (stating that “lifetime Medicare expenditures and hence the lifetime value of the program may be greater for higher-income beneficiaries, both because higher-income individuals incur more expenditures at a point in time and because they tend to live longer”).

36. See JOSEPH E. STIGLITZ, ECONOMICS OF THE PUBLIC SECTOR 539-40 (3d ed. 2000) (explaining that when the tax rate on upper income individuals is increased there will be little income effect from the tax, but there will be a large substitution effect causing a reduction in labor supply); FELDSTEIN, supra note 31, at 5 (explaining that the “deadweight loss associated with the rise in the marginal tax rate can be quite substantial”); see generally Dale W. Jorgenson & Kun-Young Yun, The Excess Burden of Taxation in the United States, J. OF ACCT., AUDITING & FIN., Fall 1991, at 487.

37. See STIGLITZ, supra note 36, at 111 (defining deadweight loss and excess burden as “the measure of the inefficiency of the tax” caused by individuals’ efforts to “avoid payments of the tax”).

38. See Jorgenson & Yun, supra note 36, at 487 (“Our most important conclusion is the excess burden imposed on the U.S. economy by the current tax system is very large.”).

39. See STIGLITZ, supra note 36, at 528 (“Deadweight loss increases with the square of the tax rate.”).

40. See Mark V. Pauly, Should We Be Worried About High Real Medical Spending Growth inthe United States?, HEALTH AFF. WEB EXCLUSIVE W3-15, W3-21 (Jan. 8, 2003), at http://content.healthaffairs.org/cgi/reprint/hlthaff.w3.15vl.pdf (noting that economists who have attempted to measure the cost of taxes argue that, depending on the type of tax used, it can be 30% or more of taxes collected) (on file with the Washington and Lee Law Review).

41. See Martin Feldstein, Social Security and Saving: New Time Series Evidence, NAT’L TAX J., June 1996, at 151, 162 (concluding that “each dollar of [Social Security wealth] reduces private saving by between two and three cents”).

42. See generally Patricia M. Danzon, Hidden Overhead Costs: Is Canada’s System Really Less Expensive?, HEALTH AFF., Spring 1992, at 21 (discussing the hidden costs in Canada’s social system).

43. See Bradley C. Strunk & Paul B. Ginsburg, Tracking Health Care Costs: Trends Stabilize But Remain High in 2002, HEALTH AFF. WEB EXCLUSIVE W3-266, W3-272 (June 11, 2003), at http://content.healthaffairs.org/cgi/reprint/hlthaff.w3.266vl.pdf (stating that “premiums for employment-based insurance continued to trend upward in 2003”) (last visited Oct. 23, 2003) (on file with the Washington and Lee Law Review).

44. U.S. DEP’T OF HEALTH & HUMAN SERVS., THE 2003 ANNUAL REPORT OF THE BOARD OF TRUSTEES OF THE FEDERAL OLD-AGE AND SURVIVORS INSURANCE AND DISABILITY INSURANCE TRUST FUNDS 168-69 (2003), available at http://www.ssa.gov/OACT/TR/TR03/tr03.pdf (on file with the Washington and Lee Law Review).

45. See infra Part VII (discussing the form of limitation on the use of new technology).

46. See Levit et al., supra note 21, at 161 (discussing the acceleration of private health insurance premium growth).

47. See Jon Gabel et al., Health Benefits in 2003: Premiums Reach Thirteen-Year High as Employers Adopt New Forms of Cost Sharing, HEALTH AFF., Sept./Oct, 2003, at 117, 117 (noting that premiums increased by 13.9%, the largest amount since 1990, from spring 2002 to spring 2003).

48. See Pauly, supra note 40, at W3-23 (discussing consumer dissatisfaction with the higher outpatient cost sharing that existed in the 1970s and 1980s and suggesting that cost sharing will not be an adequate solution to the current rising health care costs).

49. Id.

50. In 2001, out of pocket payments for hospital care (inpatient and outpatient combined) were only 3% of total spending. For figures used in the calculation, see CTRS. FOR MEDICARE & MEDICAID SERVS., U.S. DEP’T OF HEALTH & HUMAN SERVS., TABLE 6: HOSPITAL CARE EXPENDITURES AGGREGATE AND PER CAPITA AMOUNTS, PERCENT DISTRIBUTION AND AVERAGE ANNUAL PERCENTAGE CHANGE BY SOURCE OF FUNDS: SELECTED CALENDAR YEARS 1980-2012, at http://www.cms.hhs.gov/statistics/nhe/projections-2002/proj2002.pdf (last visited Jan. 28, 2004) (on file with the Washington and Lee Law Review). Further reductions in out-of-pocket spending can therefore be no more than this 3% of total hospital spending.

51. See Stephen Heffler et al., Health Spending Projections for 2002-2012, HEALTH AFF. WEB EXCLUSIVE W3-54 (Feb. 7, 2003), at http://content.healthaffairs.org/cgi/reprint /hlthaff.w3.54vl.pdf (stating that health insurance enrollment is expected to decline as a result of the weak economy and continued premium inflation) (on file with the Washington and Lee Law Review).

52. In fact private plans do assess technology for coverage policy, some rather more aggressively than Medicare. See generally Susan Gleeson, Blue Cross and Blue Shield Association Initiative in Technology Assessment, in ADOPTING NEW MEDICAL TECHNOLOGY 96 (Annetine C. Gelijns & Holly V. Dawkins eds., 1995); PENNY E. MOHR ET AL., PROJECT HOPE CENTER FOR HEALTH AFFAIRS, PAYING FOR NEW MEDICAL TECHNOLOGY: LESSONS FOR THE MEDICARE PROGRAM FROM OTHER LARGE HEALTH CARE PURCHASERS (2003); Alan M. Garber, Evidence-Based Coverage Policy, HEALTH AFF., Sept./Oct. 2001, at 62, 62.

53. See MOHR ET AL., supra note 52, at 8 (discussing the influence of Medicare on the pricing of new technologies).

54. See H.R. REP. No. 105-2, at 138 (1997) (comparing the administrative costs of Medicare to those of private insurers and finding Medicare’s costs to be lower); TIMOTHY STOLTZFUS JOST, DISENTITLEMENT? THE THREATS FACING OUR PUBLIC HEALTH-CARE PROGRAMS AND A RIGHTS-BASED RESPONSE 147 (2003) (stating that administrative costs of private plans are higher than those of Medicare because private plans have marketing and underwriting costs and must make a profit).

55. SHERRY GLIED, CHRONIC CONDITION: WHY HEALTH REFORM FAILS 47 (1997) (“The administration costs of health care are likely to continue increasing as long as the institutional structure of the health system accommodates technological change.”).

56. See JOST, supra note 54, at 147 (explaining that “Medicare holds such a large market share for so many goods and services that it is often able to demand one of the lowest prices in the market”).

57. See Uwe E. Reinhardt, Perspectives on the Pharmaceutical Industry, HEALTH AFF., Sept./Oct. 2001, at 136, 142 (noting that policymakers are concerned about how downward pressure on revenues caused by either the government or the private managed care sector will affect the research and development expenses of pharmaceutical companies); P.M. Sherer, The Link Between Gross Profitability and Pharmaceutical R&D Spending, HEALTH AFF. , Sept./Oct. 2001, at 216,220 (explaining that the pharmaceutical industry’s behavior of increasing research and development expenses as profit opportunities expand “has self-evident implications for policy interventions aimed at reducing industry prices and profits”).

58. See PATEL & RUSHEFSKY, supra note 13, at 70 (laying out the implications of Medicaid spending rules).

59. See generally ECONOMIC EVALUATION IN HEALTH CARE (Michael Drummond & Alistair McGuireeds., 2001).

60. See REASONABLE RATIONING: INTERNATIONAL EXPERIENCE OF PRIORITY SETTING IN HEALTH CARE 4 (Chris Ham & Glenn Roberts eds., 2003) (explaining that the increased cost of medical care has caused several governments to improve their health technology assessment in order to make more informed coverage decisions); Garber, supra note 52, at 64 (noting that “coverage policy usually consists of a set of procedures for deciding which items are covered”); Gleeson, supra note 52, at 97-98 (setting forth the “five technology assessment criteria used by the [Blue Cross and Blue Shield Association] to determine whether a technology improves health outcomes and can be recommended as eligible for coverage”).

61. See Gina Kolata, Newest Treatments Create a Quandary on Medicare Costs, N.Y. TIMES, Aug. 17, 2003, at A21 (noting that “cost has traditionally not been a consideration” in Medicare’s coverage decisions).

62. Id. (quoting Dr. Sean Tunis, Chief Medical Officer, Centers for Medicare and Medicaid Services).

63. See Susan Bartlett Foote, Why Medicare Cannot Promulgate a National Coverage Rule: a case o/Regula Mortis, 27 J. HEALTH, POL., POL’Y & LAW 707, 707 (2002) (explaining that “the medical device industry has prevented the implementation of a rule” by Medicare that would “clarify legal authority and describe specific criteria for evaluation of new technology in Medicare”).

64. See id. at 713 (discussing the Health Care Financing Administration’s draft coverage policy proposal that Congress designed to “provide beneficiaries with the best technology while also protecting the integrity of the Medicare Trust Fund”).

65. Id. at 711.

66. See MANCUR OLSON, JR. , THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS 29 (Revised ed. 1971) (“Where small groups with common interests are concerned, then, there is a systematic tendency for ‘exploitation’ of the great by the small!”).

67. 42 U.S.C. § 1395(a)(1)(A) (2000).

68. Medicare Program, Criteria and Procedures for Making Medical Services Coverage Decisions That Relate to Health Care Technology, 54 Fed Reg. 4302, 4317 (Jan. 30, 1989) (to be codified at 42 C.F.R. pt. 400, 405).

69. Medicare Program, Criteria for Making Coverage Decisions, 65 Fed. Reg. 31, 124, 27 (May 16, 2000) (to be codified at 42 C.F.R. pt. 405).

70. See generally Medicare Program, Criteria and Procedures for Making Medical Services Coverage Decisions That Relate to Health Care Technology, 54 Fed Reg. at 4302-18.

71. See generally Medicare Program, Criteria for Making Coverage Decisions, 65 Fed. Reg. at 31, 124-29.

72. See Susan Bartlett Foote, Focus On Locus: Evolution Of Medicare’s Local Coverage Policy, HEALTH AFF., July/Aug. 2003, at 137, 137 (“Medicare relies on private contractors-fiscal intermediaries (FIs, for Part A, Hospital Insurance) and carriers (for Part B, Supplementary Medical Insurance)-to process claims for payment. These local contractors must determine if the claim applies to a covered benefit and, if so, whether the item or service is ‘reasonable and necessary.'”).

73. See Foote, supra note 63, at 723 (“Congress itself could develop and pass into law detailed coverage criteria.”).

74. See id. at 724-25 (discussing the possibility of negotiated rulemaking as a way to force regulatory action when an interest group causes a regulatory stalemate).

75. See id. at 726 (suggesting that past coverage decisions could be viewed as precedent that creates a common law rule of coverage).

76. See James R. Knickman et al., Wealth Patterns Among Elderly Americans: implications for Health Affordability, HEALTH AFF., May/June 2003, at 168, 168 (stating that between 2000 and 2030 “fewer elderly will be in the lowest income bracket, and the percentage with real incomes exceeding $80,000 per year will almost double”).

Mark V. Pauly*

* Mark Pauly is Bendheim Professor and Chair of the Health Care Systems Department at the Wharton School of the University of Pennsylvania.

Copyright Washington & Lee University, School of Law Fall 2003

Provided by ProQuest Information and Learning Company. All rights Reserved