Health Affairs, 24, no. 6 (2005): 1415-1425 doi: 10.1377/hlthaff.24.6.1415 © 2005 by Project HOPE	New Online   Obama's Success & Next Steps   3 Health Policy Options for President Obama   A Hybrid Proposal   Bipartisan Compromise

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via ISI Web of Science (7) Citing Articles via Google Scholar Google Scholar Articles by Moran, D. W. Search for Related Content PubMed PubMed Citation Articles by Moran, D. W. Related Collections Health Reform Insurance Coverage Business Of Health Health Spending Politics Consumer Issues Insurance Market

Rethinking Reform

Whence And Whither Health Insurance? A Revisionist History

Abstract

 
Throughout the postwar era in federal health policy, policymakers have sought to expand both public and private insurance coverage, while wrestling with the cost consequences of the demand generated by the insurance-financing mechanisms thus created. This essay advances the view that the limits to insurance expansion have been reached and that public and private plan sponsors will henceforth continually "thin out" the coverage they offer. In this environment, policymakers seeking to mitigate access concerns may need to consider strategies that promote direct service delivery. This emerging regime, it is argued, will have important implications for the future of innovation in health care.

Since much of the U.S. health policy debate during the ensuing three decades can be seen as a contest between these two alternative ideologies of health care cost control, it is tempting to read "health care modern history" as a game of "King of the Hill" between proponents of regulatory cost-control strategies and those favoring "market-oriented solutions." Under this reading of history, changes in control of the political apparatus at either end of Pennsylvania Avenue result in the local ascendancy of one health cost management strategy over the other. When "free market" aficionados are at the wheel, we get increased emphasis on pluralistic alternatives to centrally financed fee-for-service (FFS) benefit programs, such as medical savings accounts (MSAs) and Medicare Advantage (MA) plans. When the regulatory mavens hold ascendancy, by contrast, we expand the scope of centralized fee schedules (such as "resource-based practice expense" reimbursement under Medicare) and press the investigative machinery of the executive branch into deeper scrutiny of for-profit service providers. Most press commentary reinforces the perception that health care financing policy is a seesaw battle back and forth between these two competing theories of control, waged over a fairly stable landscape of "access," "cost," and "quality" issues.

Although it is easy to become preoccupied with this "health policy as ideological struggle" model, doing so risks missing the important fact that the ground on which this battle is waged is not stable but has been shifting slowly but steadily over time. Although the labels used to characterize health policy issues have proved durable throughout the course of "modern history," the character of the issues themselves has changed subtly, but importantly, during this period. To see this, it is necessary to take a longer view of the trends driving U.S. health policy.

A Tale Of Two Financing Systems

Top A Tale Of Two... It’s All About Insurance Where Are We Headed? Implications NOTES

 
Readers of Health Affairs will not be surprised to learn that two main financing systems support the delivery of health care in the United States: employment-based health insurance and the constellation of large public insurance systems that support coverage for the main population groups (the poor and the elderly) least likely to be attached to employer-based coverage. In recent years, substantive distinctions between the forms of coverage offered in these two financing systems have blurred materially. If we look back far enough, however, we can identify some important differences in approaches that offer some hints about how U.S. health care financing might evolve if present trends play out unchecked.

As is well known, modern private health insurance began in the 1930s and quickly expanded as an employee fringe benefit when wage and price controls were implemented during the Second World War. By the 1960s, employer-based health insurance had became the dominant U.S. mode of health care financing. Even today, when workplace-based health insurance coverage is perceived as being under siege, nearly 90 percent of private-sector employees have access to some form of offer of health insurance financing from their employers.¹

Public-sector experience, 1946–2005. In the public sector, health care financing policy in the first twenty years after World War II was mostly about direct stimulus to the supply of available services. The Hill-Burton Act of 1946 represented a major commitment to the construction and renovation of hospitals.² The Health Professions Education Act of 1963 (P.L. 88-204) represented a comparable commitment to the expansion of the professional service supply. Although public-sector insurance financing was debated during this period, what the federal government actually did through most of this era was to consciously expand the health care delivery infrastructure.

Beginning in the 1960s, however, the federal government got into the insurance game in a major way. In addition to providing health insurance for federal employees (which began in the 1950s), the Kerr-Mills Act of 1960 formalized the prior practice of financing health care for welfare recipients via "vendor payments" into an explicit insurance mechanism. The adoption of Medicare and Medicaid in 1965 made the federal government the dominant force in U.S. health care finance. Almost immediately thereafter, Medicare outlays skyrocketed. After the expected initial ramping up during program implementation in 1967–68, the program unexpectedly continued to grow in inflation-adjusted terms, accelerating to real rates of increase in the 10–20 percent range for the balance of the 1970s. This growth came at an awkward time, during a period in which the United States was experiencing simultaneous increases in both inflation and unemployment. By the time "modern history" dawned, this state of affairs had shaken the Washington health policy establishment to its core.

The immediate policy response was a replay of the 1945–1965 policy agenda—in reverse. Having spent those two decades stimulating the expansion of health system infrastructure, Washington now kicked in with a heavy dose of supply constraint. In 1966 Congress had enacted legislation (the Comprehensive Health Planning and Services Act, P.L. 89-749) to formalize health planning to guide the local allocation of resources made available, under federal auspices, to promote health infrastructure building. By 1974 Congress had transformed that system into a national mandate to use "certificate-of-need" programs to damp down on infrastructure expansion (the National Health Planning and Resource Development Act of 1974, P.L. 93-641). Meanwhile, in the 1972 Social Security Amendments (P.L. 92-603), Congress created "Professional Standards Review Organizations" to scrutinize the growing volume of physician services with respect to their medical necessity. Although both the theory and the operational reality of these programs were contemporaneously controversial, it is clear in retrospect that they amounted to an effort to put the brakes on supply even as broad-based insurance coverage stepped on the demand accelerator. Whatever contribution supply constraints have made to rationalizing health system infrastructure, there is no evidence that they made any meaningful dent in the problem of rising costs.

Hence, when the Carter administration weighed in on the cost problem, it proposed explicit regulatory restraints on permissible top-line revenue growth in hospitals. The political taste for such a blunt approach to capping demand was lacking in 1977–79, but it foreshadowed the ultimately successful political effort in the 1980s and 1990s to eliminate retrospective cost reimbursement of institutional providers in favor of prospectively established administered pricing regimes.

In comparison to the futile effort to brake supply, these systems have inarguably demonstrated the federal government’s ability to dictate unit prices. With the extension of administered fee schedules to professional services in the 1990s, the public sector now has comprehensive unit price control. In 1998–99, in fact, total Medicare outlays actually declined in real terms for the first time since 1966, as the tight limits imposed by the Balanced Budget Act (BBA) of 1997 took hold.

Subsequent experience has shown, however, that unit price is only half of the overall demand equation. The current controversy over scheduled cuts in physician fees to offset rising volume demand is a clear illustration that unit price controls alone cannot effectively control aggregate costs. In fact, there is broad-based agreement in the analytical community that price constraints stimulate efforts by providers to crank up service volume to offset the income effects of price reductions.³ As the foregoing suggests, the cost management tools available to the public sector are simply not very effective. Unfortunately, the public sector’s success in using the sort of cost management tools available to the private sector has also been limited.

Private-sector experience, 1972–2005. During the period in which the federal establishment experimented with supply constraints before turning to administered price regimes, the private sector moved in other directions. Lacking the power to manage the health services marketplace through explicit regulation, private plan sponsors tried materially different strategies to control costs.

In the 1970s and into the 1980s, plan sponsors’ strategies were largely financial. The enactment of the Employee Retirement Income Security Act (ERISA) of 1974 (P.L. 93-406) provided a regulatory framework congenial to conversion of employer-sponsored plans to self-funded status. Analysis of the impact of ERISA on health benefits often focuses on the effects of ERISA’s exemption from state regulatory scrutiny. However, its major advantage in the first ten to fifteen years of its existence was the one-time—but sizable—financial impact of withdrawing the relatively healthy workforce of larger employers from the broader health care risk pool, to directly finance their lower costs in lieu of paying premiums. Over time, the health benefit management industry engineered products, such as so-called minimum premium plans, permitting ever-smaller companies to take advantage of self-funded options.

This trend sharply increased the risk profile of the small-group and individual insurance markets, which now faced materially higher premiums and uncertain access to the market. Ultimately, concerns about these impacts led public policy-makers to enact market reforms, culminating in the well-known Health Insurance Portability and Accountability Act (HIPAA) of 1996 (P.L. 104-191), to halt further erosion. These actions, however, could not reverse the aggregate financial effect of the departure of a large share of the workforce from the insured pool in the 1970s and 1980s. During 1994–2001, the number of people covered by commercial non-group policies declined from 31.9 million to 23.3 million.⁴

By the mid-1980s, most or all self-funded conversions that could take place had taken place, and health plan sponsors turned to alternative approaches to contain costs, which, while starting at a lower base, continued to rise. The mechanism selected was "managed care," which previously had been a specialized alternative product offered via organized delivery systems but which now became a broader selection of products exhibiting variations in benefit design and network structure.⁵ Within perhaps ten years, managed care, so broadly defined, became the dominant form of insurance available from both insured and self-funded sponsors. The favorable cost performance of such health plans in the 1990s—and the ensuing backlash against the most restrictive forms toward the end of that decade—have been well documented in these pages and need not be rehashed here.⁶ During roughly the same period, enrollment of Medicare and Medicaid beneficiaries in managed care grew rapidly before peaking at the turn of the millennium.

Where is the private market now headed? Although the present is always the most difficult period of time to characterize accurately, the current buzz centers on so-called consumer-directed benefit models, in which employees customize their own benefits. Implicit in this offering is the expectation that, left to their own devices, employees will elect more economically efficient (which is to say, less) coverage than they are now consuming and that this trend, over a sustained period, will provide a mechanism to back a fair amount of demand stimulus out of the system.

It’s All About Insurance

Top A Tale Of Two... It’s All About Insurance Where Are We Headed? Implications NOTES

 
As the foregoing history strongly suggests, the postwar history of U.S. health care finance has been, in an important sense, all about health insurance. The rise of insurance financing in the public and private sectors has been a major driver of increased access to services. In comparison to the 1945–1965 period, when public-sector efforts were focused primarily on increasing supply, the past forty years have empowered demand, by shielding the insured population from all but a fraction of the cost of the services it receives. Our failure to so endow a sizable share of our citizens continues to be held up as a major failure of health care policy.

Yet it also true that this demand, thus created, has been the major driver of the chronic health care inflation experienced since the 1960s. The demand for health care, priced at close to zero at the point of care, is virtually unlimited. A variety of factors, ranging from technological innovation to human avarice, have been offered up as explanations of our health cost problems. However, none of these potential contributors would "contribute" if they were not "pulled through" the health care marketplace by the demand engine of health insurance.

In such a marketplace, lack of insurance is clearly a catastrophic problem, since many products and services that are available in our health care system now bear prices that are very high relative to the income of a typical citizen. Yet the relentless growth in the cost of health benefits generated by this demand-pull inflation is the very reason why so many Americans are now uninsured.

The mechanism of employer-based health insurance, in an era of relentlessly rising costs, forces workers to consume an ever-increasing share of their compensation in the form of health insurance benefits. Since the dynamics of employee compensation occur in a relatively unconstrained free market for labor, both employers and employees are free to adjust the terms they offer and seek, to adjust to this growing wedge between compensation cost and realized net income.

Although some number of employers have dropped out of offering health insurance altogether, the more common response has been to "thin out" the coverage previously offered, shrinking the actuarial value of benefits through increased cost sharing and greater emphasis on catastrophic benefits. Seen in this light, the movement toward "consumer-directed" care can be understood more clearly as a mechanism to induce employees to voluntarily cooperate in the thinning out of their own health coverage in favor of other forms of compensation.

Although some employees have undoubtedly lost coverage involuntarily, voluntary labor-market decisions probably explain more of the story of how an increasing number of Americans are moving into the ranks of the uninsured. Definitive research on this question is curiously lacking.⁷ However, it seems likely that a substantial number of workers (particularly younger employees with lower self-perceived health risks) are voluntarily seeking employment with firms offering limited or no health benefits, which hence are able to attract workers with other forms of valued, less costly compensation. Meanwhile, we have clear evidence that retiree health benefits—an important source of coverage for the elderly and near-elderly—are thinning out rapidly and actually disappearing in many firms. During the 1997–2002 period alone, the Employee Benefit Research Institute (EBRI) reports that firms offering retiree coverage, particularly for early retirees, has shrunk markedly; EBRI’s evidence suggests that fewer than 10 percent of firms may offer such coverage before the decade is out.⁸

Where Are We Headed?

Top A Tale Of Two... It’s All About Insurance Where Are We Headed? Implications NOTES

 
In all, we have arrived at a point where we are beginning to see the truth of Herbert Stein’s famous dictum that "all unsustainable trends stop."⁹ The first-order effects are increasingly obvious. Employer-based insurance, despite the best efforts of policymakers of all persuasions, will continue to thin out and could amount to a "catastrophic only" insured benefit before we are too far into the next decade. We will know that a new equilibrium has been reached when the Democratic National Committee restricts its employee coverage offerings to MSAs. Meanwhile, public insurance systems will attempt to hold the line against this trend but will fight an increasingly uphill battle against an expanding population with "substandard" or no insurance. Although the ideological impetus toward universal coverage cannot be expected to dim in our lifetime, all of the forces in the health care marketplace appear to be moving solidly in the opposite direction.

In this sense, the recently enacted Medicare prescription drug benefit may come to be seen as the "last hurrah" of the Old Order. From here on out, public policymakers at all levels will struggle mightily to maintain existing coverage, let alone expand it. The means-testing of the Part B premium in the Medicare Prescription Drug, Improvement, and Modernization Act (MMA) of 2003, enacted in part to finance the drug coverage expansion, is a clear whiff of the future. This "thinning out" of the net actuarial value of Medicare coverage provides an important precedent for future cost management initiatives in the public sector.

The second-order effects of these changes are harder to predict, but the outlines of the choices we will need to make are beginning to emerge. Over the next twenty years, we should expect the seemingly unrelated topic of immigration policy to emerge as a major driver of the health care debate. From a social insurance perspective, the public sector’s ability to hang onto to Social Security and Medicare by midcentury will depend on an expansion of the payroll tax base well beyond anything now projected given present demographic trends. Absent accelerated immigration—or some other now-unforeseeable deus ex machina—the financial dynamics underpinning current public benefits commitments don’t look good.

Within the same time period, the public sector might shift course, expanding publicly financed direct care delivery systems as an alternative to increasingly unaffordable insurance systems. There is some evidence that larger private employers are already seeking to substitute direct care delivery for insurance-based finance.¹⁰ The current administration’s effort to markedly expand federally supported community health centers (CHCs) presents interesting evidence in this direction.¹¹ It is possible, by midcentury, that we will see public delivery systems emerge as an increasingly important mode of care delivery for the low- and middle-income population not otherwise publicly or privately insured. If this occurs, then the public sector will have come full circle in its health care financing policy in less than a century. The cost of building and operating such a system would dwarf the federal effort in the 1945–1965 period; however, doing so might prove much less costly than preserving the insurance systems erected in the 1960s. The data presented in Exhibit 1 offer an instructive comparison in that regard. It presents evidence on the comparative growth of real (inflation-adjusted) per beneficiary spending in Medicare, which is an insured benefit program, and the Veterans Affairs (VA) medical care system, which is an annually appropriated (and hence budgeted) direct care delivery system.¹²

View larger version (22K): [in this window] [in a new window]   EXHIBIT 1 Comparative Real Growth In Per Beneficiary Spending, Medicare Versus Veterans’ Medical Care, 1980–2002

In between the extremes of public insurance and public provision, proposals by the current administration and others to address the "thinning insurance" problem through subsidies for private insurance purchase (for example, tax credits) represent an interesting middle ground. Such subsidies would add net new financing to the game and hence would bridge at least part of the gap between the beneficiary’s current income and increasingly unaffordable comprehensive insurance products. Yet because these sorts of policies would (by conscious intent) leave beneficiaries exposed to variations in the market cost of such coverage, the most probable effect of such subsidies would be to cushion and facilitate the transition to a "catastrophic only" private insurance market structure over time.

Implications

Top A Tale Of Two... It’s All About Insurance Where Are We Headed? Implications NOTES

 
Politics of health policy. If this analysis is anywhere near accurate, then the politics of health policy will become increasingly interesting in the decades ahead. Advocates for expanding public insurance programs, already frustrated by the failure of public policy to come to grips with declining insurance, will be in for a very long century. Provider interests, whose business models are now inextricably tied to insured benefits regimes, will face excruciating revenue pressures as they confront a growing propensity to divert public resources from insured benefit programs to direct delivery models. Offerers of comprehensive insurance benefits, who may be best prepared to compete in an increasingly "budgeted" health benefits market, will nevertheless face challenges in adopting their market offerings to compete against "catastrophic only" models. The accommodations reached to balance the competing interests of these disparate actors, forged as they were in the "positive sum" insurance regime of the past forty years, will be torn asunder. It is, at present, hardly clear what will replace them.

Medical innovation. Although we will all undoubtedly become preoccupied with the purely political dimensions of all this, it is important to remember that the central question for policy is what this all will mean for the character of health care services that will be available to Americans under this New Order. Of particular importance is what a "thin insurance" marketplace will mean for the scope and pace of medical innovation—and for access to the benefits of whatever innovation will occur going forward. This question is of relevance because, since at least the 1980s, the pace of U.S. technological innovation in health care has been heavily influenced by the scope of available coverage. Most controversies regarding medical technology before then focused on the technology infrastructure of the system—that is, on important new diagnostic and treatment modalities such as computed tomography (CT) scanning, lithotripsy, and comparable "big ticket" additions to the system’s technological armamentarium. During that period, the question for policy was how to best discipline the rapid growth in capital spending for such technologies, the costs of which were then amortized across the system via retrospective cost reimbursement. Explicit supply constraints having proved ineffective, there was a broad consensus by the early 1980s to move to prospective payment methodologies, first for inpatient hospital services in Medicare and then for most institutional services throughout the system. By transferring the financial risk associated with financing the cost of new technology from the payers to the providers, prospective payment became an important check on the previously unconstrained growth of the medical technology infrastructure.

At about that time, however, the character of technological innovation changed in a subtle but important way. While new "infrastructure" technologies continued to emerge, the 1980s and 1990s saw the rapid growth of more "personal" technologies—new drugs and implantable medical devices that were personally consumed by the patient and hence were separately reimbursable under most public and private insurance schemes. Increasingly, such new technologies are being "pulled through" by insurance at ever-higher prices. In the absence of insurance, many of the highest-price technologies, which can carry price tags in the tens of thousands of dollars per patient, would simply be unaffordable for the vast majority of patients. Since the population frequency of these therapies is low, however, insurance systems can now pool the cost of such therapies at an affordable cost.

As private insurance thins out toward a more catastrophic form, however, this dynamic can be expected to change, in two ways. First, facing a much larger share of the first-dollar cost of expensive therapies, patients will be much more price-sensitive than they are now. Second, having transferred most of the first-dollar risks to the beneficiaries, health plan sponsors will focus their cost containment efforts more sharply on the big-ticket items that represent an increasing share of their costs. The combined effect of these two sets of pressures will push back strongly against manufacturers’ pricing flexibility. Increasingly, the question of whether a high-cost technology is covered at all by an insurance plan will become the most important determinant of product economics.

This economic regime change will have powerful effects on manufacturers’ incentives and will shape the character and direction of innovation. Although the course of technological innovation is inherently unpredictable, it seems likely that more attention in a price-sensitive world would be directed toward innovations that are demonstrably cost-reducing, at the expense of those that offer only incremental clinical benefits at a substantial cost.

Although this change in course, as stated, sounds positive relative to a common critique of the status quo ante, critics of market-based approaches to health policy problems will quite reasonably point out that decentralized decision making under intense financial pressure could result in decisions different than those that would be reached if judgments about products were reached in a more orderly fashion, involving disciplined application of the principles of "evidence-based medicine" and "cost-effectiveness analysis." Critics of "central planning" strategies for determining coverage policy would counter that these tools, while potentially useful in informing judgments about the value of new technologies, don’t in and of themselves offer a logic for making complex decisions in a centralized way. Although it is possible to transform the economic evidence regarding technologies to common standards such as cost per quality-adjusted life year (QALY) saved, such metrics are not self-implementing: They don’t tell us whether $30,000 per QALY is a bargain or whether $100,000 per QALY is exorbitant. This choice cannot be informed by analysis. At the end of the day, it is a purely political choice, which rational politicians will figure out how to duck as long as they can.

Over time, therefore, it seems safe to predict that emerging marketplace dynamics, rather than the expressed preferences of system reformers, will drive choices among emerging technologies in the same way that it now drives choices among amenities in health care. Confronted with a continuing decline in the actuarial value of private insurance, policymakers wishing to internalize the rationing mechanisms that will be applied to technology choice would need to up the ante on the public share of the insured system—which from today’s vantage point is close to inconceivable—or else up the public stake in direct delivery. Since initiatives in community-based primary care can address the problems of the ever more poorly insured without placing public decisionmakers in the position to have to make hard technology choices, it is very tempting to conclude that a new public/private détente could emerge across the boundary between "high tech" and "low tech" medicine, with federal funding focused on delivering the latter, and helping buy low-income and elderly people into a privately organized "catastrophic only" market that would become our primary conduit for pooling the cost of the former.

Will we like this? Given the potentially profound changes implied, during the decades to come, by the working of the forces just described, it is natural to consider whether the public will be sufficiently accepting of these changes to permit such a new financing structure to actually emerge. This may not matter: The forces at work here may well be so large as to dominate the impassioned efforts of partisans of all persuasions to achieve any alternative outcome, no matter how well supported by public opinion. For while the Tennessee Valley Authority stands as testimony to the U.S. public sector’s ability to extract useful work from water as it rushes downhill, this experience does not confer the capacity to make it flow back up again. Although we have demonstrated the capacity to stimulate demand for health care through insurance-financed systems, we clearly lack the capacity to sustain it, let alone redirect insurance finance in a northward direction.

For the next few generations of Washington health policymakers, therefore, "modern history" may come to seem a bit of a wild ride. For although it is unclear exactly where we are going, it is clear that we are making very good time.

Editor's Notes

 
Donald Moran (dwmoran{at}themorancompany.com) is president of the Moran Company in Arlington, Virginia.

NOTES

Top A Tale Of Two... It’s All About Insurance Where Are We Headed? Implications NOTES

 

1. M.W. Stanton, "Employer-Sponsored Health Insurance: Trends in Cost and Access," Research in Action, no. 17, September 2004, www.ahrq.gov/research/empspria/empspria.htm (2 September 2005).
2. The Hospital Survey and Construction Act of 1946 (P.L. 79-725) is codified in Title VI of the Public Health Service Act.
3. See N.X. Nguyen, "Physician Behavioral Response to Price Control," HCO Working Paper no. 46 (Washington: World Bank, 1995).
4. Data derived from analysis of U.S. Bureau of the Census, Statistical Abstract of the United States, 2000–2003 editions (Washington: U.S. Government Printing Office, various years).
5. D.W. Moran, "HMOs, Competition, and the Politics of Minimum Benefits," Milbank Quarterly 59, no. 2 (1981): 190–208.
6. D.W. Moran, "Federal Regulation of Managed Care: An Impulse in Search of a Theory?" Health Affairs 16, no. 6 (1997): 7–21.[Abstract]
7. I thank Jon Gabel of the Center for Studying Health System Change for his private assessment of the state of such research.
8. P. Fronstin, "The Impact of the Erosion of Retiree Health Benefits on Workers and Retirees," Issue Brief no. 279, March 2005, www.ebri.org/pdf/briefspdf/0305ib.pdf (2 September 2005).
9. Herbert Stein served as chairman of the President’s Council of Economic Advisers during the Nixon administration.
10. V. Fuhrmans, "One Cure for High Health Costs: In-House Clinics at Companies," Wall Street Journal, 11 February 2005.
11. G.W. Bush, Budget of the United States Government, Fiscal Year 2006 (Washington: U.S. Office of Management and Budget, 2005), 136.
12. The comparison shown actually overstates VA system growth somewhat, because the time series used to generate the graphic employs the population of veterans with service-connected disabilities as the denominator of the per beneficiary outlay calculation throughout the period. Since access to the VA system has been expanded, on a space-available basis, to a wider body of veterans during the past ten years, counting such "service-connected equivalents" would lower the measured growth rate materially.

                      What's this?