Distributive injustice(s) in American health care

Law and Contemporary Problems

Volumn 69, Issue 4, 2006, Pages 7-82

  Havighurst, Clark C ^a   Richmand, Barak D ^a  

^a DUKE UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 33947366476     PISSN: 00239186     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (484)

1. Although we focus on the private sector in this article, there are important distributive justice issues on the public side as well. To be sure, progressive redistribution appears in many programs of public subsidies and financing - the Medicaid program in particular. But many aspects of Medicare are not so progressive. Thus, less than a third of Medicare's funding in 2004 came from general revenues raised through progressive taxation. See Soc. Security & Medicare Bd. of Trs., Status of the Social Security and Medicare Programs: A Summary of the 2005 Annual Reports (2005) http://www.socialsecurity.gov/OACT/TRSUM/ trsummary.html (showing breakdown of 2004 income of the Hospital Insurance and Supplementary Medical Insurance trust funds).
2. A much larger share (fifty-four percent) was yielded by a 2.9% flat tax on the wages, salaries, and self-employment income of current workers - a tax that, because it applies only to earned income, is less equitable than a true flat tax. See also infra note 48.
3. Moreover, if one views Medicare only as a pay-as-you-go program, this tax appears to be an enormous intergenerational transfer, taking large amounts from today's workforce to provide health services for today's retirees. If Medicare is viewed as social insurance, however, the unfairness is less clear because, up to now at least, each generation has gotten substantially more out of the program than it has put in. See Mark McClellan & Jonathan Skinner, The Incidence of Medicare, 90 J. Pub. Econ. 257 (2006). Nevertheless, it is unlikely that this pattern will continue. It depends, after all, not only on continuing upward trends in both health care costs and life expectancy but also on the willingness of the next generation of workers to pay the taxes needed to support the elderly population in the same generous way. There are good reasons, it would seem, to question the faireness of Medicare's payroll tax to today's workers.
4. Medicare's intragenerational fairness can also be questioned - and is actively in dispute. It is at least an open question, it appears, whether, having made larger payments into the program, higher-income individuals may enjoy more than proportionately greater benefits from it. This could happen because higher-income individuals both live longer under the program and make more intensive use of its nominally equal entitlements. (We examine the latter matter at length infra notes 121-127 and accompanying text).
5. For the latest (conflicting) findings on whether Medicare, taken as a whole, is a regressive or progressive program, compare McClellan & Skinner, supra (using average incomes within zip codes as proxies for beneficiaries' incomes and finding net wealth transfers from lower- to higher-income beneficiaries),
6. with Jay Bhattacharya & Darius Lakdawalla, Does Medicare Benefit the Poor?, 90 J. Pub. Econ. 277 (2006) (using educational attainment as a proxy for income and finding net transfers from higher- to lower-income beneficiaries).
7. In view of this mixed record of distributive justice, one might wonder why Medicare is so rarely criticized by those concerned about the welfare of lower-income Americans. But see Jonathan Skinner & John E. Wennberg, Exceptionalism or Extravagance: What's Different About Health Care in South Florida?, 2003 Health Aff.(Web Exclusives) W3-372, W3-374 ("On equity grounds, we have problems with the idea of single working mothers in Nebraska (often themselves lacking health insurance) footing the bill for gold-plated health care provided to high-income Medicare enrollees in Miami.").
8. For a fuller discussion of the income elasticity of demand for health services and its significance for our thesis, see infra note 101 and text accompanying note 209.
9. A good example of this pervasive bias can be found in the stated primary mission of the $9 billion Robert Wood Johnson Foundation (RWJF): "To improve the health and health care of all Americans." RWJF Ann. Rep. 2004, at 1 (2004).
10. The Foundation's ambivalent attitude toward cost as a relevant consideration in its grant making has recently been documented - notably, in a publication sponsored by the Foundation itself. Carolyn Newbergh, The Robert Wood Johnson Foundation's Efforts to Contain Health Care Costs, in To Improve Health and Health Care, Volume VII: The Robert Wood Johnson Foundation Anthology 57-80 (2004).
11. Although the Foundation has not entirely ignored cost in its activities, it was only in 1998, twenty-six years after its founding, that it recognized cost considerations at all in its mission statement - as one of four subsidiary goals: "To assure that all Americans have access to basic health care at reasonable cost." Id. at 64.
12. In 2003, it substituted the word quality for basic, indicating that the objective of more and better health care still dominates. See id. at 80 n.2.
13. Not only is the notion that costs must be subjectively "reasonable" little more than lip service to an obvious public concern, but, whenever the Foundation has acknowledged high costs as a problem, it has mostly seen them only as an obstacle to achieving its main mission, not as a possibly unwarranted burden on those who pay them. See, e.g., Goals Update, in RWJF Ann. Rep. 1996, at 141 (1997) ("[C]ontrolling costs was clearly an essential prerequisite for our other goals...").
14. Newbergh describes how Steven Schroeder, M.D., RWJF's president from 1990 to 2002, downgraded interest in cost to a "half goal." Newbergh, supra, at 63.
15. Finally, it is significant that most of the Foundation's activities relating to cost containment that Newbergh describes involved sponsoring community, governmental, or industry-wide cooperative initiatives (for example, an RWJF program focused on Medicaid managed care, its "Community Programs for Affordable Health Care," and its "Physician-Directed Program to Improve Medical Care Services and Control Costs") rather than promoting ways of making costs and quality negotiable in the marketplace. See id. at 62.
16. Crucial to our claim that premium payers are overburdened is the proposition, stoutly defended by most economists, that, even though employers purchase most health coverage, its cost is ultimately borne virtually in full by employees, principally in the form of reduced wages. Logically, employers are primarily concerned about a worker's total compensation and are largely indifferent about the forms in which that compensation is paid. Empirical studies support the economists' prediction. See generally Jonathan Gruber, Health Insurance and the Labor Market 55 (Nat'l Bureau of Econ. Research, Working Paper No. 6762, 1998) (reviewing the empirical literature and concluding that "the results that attempt to control for worker selection, firm selection, or (ideally) both, have produced a fairly uniform result: the costs of health insurance are fully shifted to wages"). Individual studies have found that real wages fall after government-mandated increases in the cost of insurance and when health costs rise, especially for high-consumption groups.
17. See Jonathan Gruber & Alan B. Krueger, The Incidence of Mandated Employer-Provided Insurance: Lessons from Workers' Compensation Insurance 1 (Princeton Univ. Indus. Relations Section, Working Paper No. 279, 1990) (finding that areas and industries with high workers-compensation costs led to a corresponding reduction in wages);
18. Jonathan Gruber, The Incidence of Mandated Maternity Benefits, 84 Am. Econ. Rev. 622 (1994) (finding that the mandated expansion of health insurance coverage in the Pregnancy Discrimination Act of 1978 led to reduced wages);
19. Louise Sheiner, Health Care Costs, Wages, and Aging (Apr. 1999) (unpublished manuscript, on file with Fed. Reserve Bd. of Governors), available at http://www.federalreserve.gov/pubs/feds/1999/199919/ 199919pap.pdf (finding relatively lower wages for older workers and workers with family coverage in areas with high medical prices).
20. See, e.g., Holman W. Jenkins, Jr., Wanna Fix Health Care? Stop Hiding the Cost!, Wall St. J., Oct. 13, 2004, at A17.
21. The only extensive overall examination of distributional issues that we have found is a 1992 study, John Holahan & Sheila Zedlewski, Who Pays for Health Care in the United States? Implications for Health System Reform, 29 Inquiry 231 (1992).
22. Undertaken to inform efforts at major health reform, that study was mostly concerned with showing how the overall burden of paying for health care was then distributed among income classes in order to provide a baseline for evaluating any new plan or proposal. To be sure, that study found that "the distribution of the financing burden is regressive." Id. at 244. But, as Christopher Conover shows in this volume, this characterization is not helpful or accurate for present purposes because it depends upon an artificial and unrealistic definition of regressivity - in terms of the proportion of income that a given class spends on health care in general, including its own, rather than only the proportion that is diverted to essentially public purposes or to meeting regulatory requirements.
23. Christopher J. Conover, Distributional Considerations in the Overregulation of Health Professionals, Health Facilities, and Health Plans, 69 Law & Contemp. Probs. 181, 183 (Autumn 2006). Moreover, the method employed by Holahan and Zedlewski in identifying cost burdens - estimating the average burden borne by individuals in each income decile - did not reveal the particular burden borne by lower- and middle-income premium payers to which we call attention in this article. Indeed, because each higher decile almost certainly included a greater number of persons having health coverage, the average cost calculated for each decile effectively obscured differences in the burdens borne by insured and uninsured individuals, both within each decile and across the board. Thus, whereas Holahan and Zedlewski's data inadvertently made it appear that individuals in each higher income decile bore higher insurance costs than those with lower incomes, we hold it unlikely that insurance costs for those actually having insurance vary greatly according to income.
24. From the late 1980s to the late 1990s, the share of aggregate personal income received by the middle fifth of the population fell from 17.2% to 16.2%, while the income of the top fifth increased from 42.1% to 45.4% Jared Bernstein et al., Pulling Apart: A State-By-State Analysis of Income Trends xi (2000). The poorest quintile likewise fell further behind the high earners. Thus, incomes at the twentieth percentile were 16.8% of incomes at the ninetieth percentile in 1989 and 15.9% of such incomes ($18,556 versus $116,472) ten years later.
25. Carmen DeNavas-Walt et al., U.S. Census Bureau, Income in The United States: 2002 26, available at http://www.census.gov/prod/2003pubs/ p60-221.pdf.
26. See generally Organization for Economic Cooperation and Development (OECD) Health Data, chart 4-5 (June 5. 2005), http://www.oecd.org/ dataoecd/35/13/34966969.pdf. In 2003, when health spending in the United States represented 15% of GDP, no other nation spent more than (Switzerland's) 11.5% of its GDP on health care in that year, and the median level of expenditure among all thirty OECD nations was 8.6% of GDP. If Switzerland's 11.5% of GDP devoted to health care is converted into "purchasing power parity international dollars" (PPP$), the OECD data can be read to show that Switzerland in fact spent only 67% as much per capita on health care in 2003 as the United States ($3,781 versus $5,635).
27. See Uwe E. Reinhardt et al., U.S. Health Care Spending in an International Context Health Aff., May-June 2004, at 10 (using this methodology to derive similar estimates). Treating as "excessive" all U.S. health spending over this percentage provides a basis of sorts for the eye-catching number suggested in the text as a possible measure of the magnitude of U.S. overspending. The extreme outlier status of the United States also appears from a comparison of its 2003 health spending of 5,635 PPP$ per capita with median per capita spending of 2,161 PPP$ among all OECD nations.
28. See id. To be sure, the income-elasticity of demand for health care is such that richer nations naturally spend higher percentages of their national income on it.
29. See infra note 101. But this article shows not only that the level of spendin in the U.S. results from a unique combiantion of dysfunctional markets and misguided public policies but also that it is working-class Americans, and not the affluent, whose money is being spent excessively on health care - without anything resembling their informed consent.
30. In addition, working Americans bear much of the cost of the largely pay-as-you-go Medicare program through a payroll tax of questionable fairness. See supra note 1.
31. Christine Borger et al., Health Spending Projections Through 2015: Changes on the Horizon, 2006 Health Aff. (Web Exclusives) W61, W62 exh. 1.
32. Kaiser Family Foundation & Health Research and Educational Trust, Employer Health Benefits: 2005 Annual Survey 16 (2005) (estimating the average annual premium for family health insurance coverage in 2005 as $10,880);
33. U.S. Dep't Hous. & Urban Dev., Notice PDR-2005-01 (2005) (estimating median family income in 2005 as $58,000).
34. Wasteful spending might also take the form of excessive administrative costs, which many think they see in the U.S. private sector. See, e.g., David U. Himmelstein, Steffie Woolhandler & Sidney M. Woolf, Administrative Waste in the U.S. Health Care System in 2003: The Cost to the Nation, the States, and the District of Columbia, with State-Specific Estimates of Potential Savings, 34 Int'l J. Health Services 79 (2004). Although a decentralized, competitive industry will inevitably incur certain costs that could be avoided in a system run by government, those added costs are usually justified by a competitive market's better incentives for improved performance and greater ability to satisfy differing consumer needs and preferences. To the extent that the nominally decentralized U.S. health care system does not deliver these benefits (as we later suggest it largely fails to do), any higher costs it entails may be counted as unjustified burdens imposed on consumers for the benefit of the insurance industry and other interests. Rather than centralizing administration in government's hands, however, the better way to address inefficiency of this kind might be to enable, and encourage, health insurers to realize the usual benefits of competition.
35. See Clark C. Havighurst, Why Preserve Private Health Care Financing?, in American Health Policy: Critical Issues for Reform 87 (1993) (arguing that private health plans are hard to defend unless they are allowed to offer, and do offer, consumers a full range of health care options, including economizing opportunities).
36. Reinhardt et al., supra note 8;
37. Gerard F. Anderson et al., It's the Prices, Stupid: Why the United States Is So Different from Other Countries, Health Aff., May-June 2003, at 89. Comparisons of care in the United States with care in other OECD nations suggest that American health care consumers do not receive generally better quality but are simply paying more for comparable goods and services.
38. See, e.g., Peter S. Hussey et al., How Does the Quality of Care Compare in Five Countries?, 23 Health Aff. 89(2004).
39. The Medicare Payment Advisory Commission (MedPAC) reports that although hospital payment-to-cost ratios for private payers declined fairly steadily through the 1990s, dropping from approximately 130% to 115%, comparable ratios for Medicare and Medicaid were substantially lower. MedPAC, Medicare Payment Policy 62 fig.2B-1 (2002).
40. For physician services, Medicare payments were only eighty-three percent of private payments in 2004. MedPAC, Medicare Payment Policy 81 (2006).
41. See infra note 34.
42. See generally Jean Tirole, The Theory of Indutrial Organization 65-78 (1988) (outlining the distortions and inefficiencies associated with monopoly in economic theory). For reasons why economists are chary of basing policy prescriptions on claims that their adoption will increase allocative efficiency, see note 212 infra.
43. Thus, health insurance eliminates most of the "deadweight loss triangle" that appears under the demand curve in textbook illustrations of monopoly's misallocative effect. See Arti K. Rai, The Information Revolution Reaches Pharmaceuticals: Balancing Innovation Incentives, Cost, and Access in the Post-Genomics Era, 2001 U. Ill. L. Rev. 173, 202-09 (2001) (observing that "providing access to insurance is a low-cost mechanism for reducing deadweight loss");
44. Martin Gaynor, et al., Are Invisible Hands Good Hands? Moral Hazard, Competition, and the Second-Best in Health Care Markets, 108 J. Pol. Econ. 992 (2000) (considering offsetting allocative effects of insurance-induced moral hazard and supracompetitive prices for health services).
45. For a graphic presentation of deadweight loss, see Tirole, supra note 16, at 67.
46. Most economists take no professional position regarding the distribution or redistribution of income because it has no effect on aggregate welfare (that is, efficiency) unless one makes certain assumptions about the marginal utility of income to different individuals. See generally Daniel A. Farber, What (If Anything) Can Economics Say About Equity?, 101 Mich. L. Rev. 1791 (2003)
47. (reviewing Louis Kaplow & Steven Shavell, Fairness versus Welfare (2002), extensively examining arguments in law and economics). But monopoly's redistributive effects can be a potent political issue - as we suggest they should be in the current state of U.S. health care.
48. In the normal monopoly case, there is a trade-off between allocative inefficiency and redistribution. Thus, while more elastic demand makes high prices more likely to discourage desirable consumption, it also means that there is less consumer surplus (see infra note 54) available for a monopolist to capture; likewise, to whatever extent a monopolist is able to practice price discrimination, consumption is facilitated and deadweight loss is reduced - but the monopolist earns greater profits at consumers' expense. Strikingly, this reciprocal relationship between monopoly's misallocative and redistributive effects does not generally hold in the U.S. health care market. As the text in this and the next section explains, once U.S.-style health insurance and a few other factors (specifically, the peculiar incentives and conduct of nonprofit firms) are added to the mix, monopoly may generate excessive rather than suboptimal consumption. And this misallocation of resources (if it occurs as we hypothesize), instead of being offset in some sense by a reduction in monopoly's redistributive effect, is simply an additional burden on the economy.
49. The differences among different forms of cost sharing are important in this context. In health plans with so-called tiered benefits, payment of a fixed co-payment usually entitles the insured to full coverage of the remaining cost of "medically necessary" care and medications. In other plans, stop-loss provisions entirely eliminate co-insurance at some point. In both cases, once the initial deductible or other cost-sharing requirement is met, the sky may be almost literally the only limit on the monopolist's pricing freedom. To be sure, the amount of pricing freedom enjoyed by sellers of patented, therapeutically unique prescription drugs has not been quite so great because, until fairly recently, a large proportion of the population lacked extensive coverage for these products, and Medicare generally did not cover them (even though its beneficiaries were particularly heavy users). But, according to one survey, seventy-three percent of non-institutionalized Medicare beneficiaries aged sixty-five or older had some form of public or private (Medigap or retiree) coverage of prescription drugs in 2003. Dana G. Safran et al., Prescription Drug Coverage and Seniors: Findings from a 2003 National Survey, 2005 Health Aff. (Web Exclusives) W5-152.
50. Also, coverage for the rest of the population expanded rapidly from the late 1980s. See Patricia M. Danzon & Mark V. Pauly, Health Insurance and the Growth in Pharmaceutical Expenditures, 45 J. Law & Econ. 587 (2002).
51. The finding by Danzon and Pauly that the rapidly rising share of the national health dollar claimed by prescription drugs during the period in question reflected improved insurance coverage is consistent with the hypothesis that monopolistic sellers found it increasingly profitable to price their products in the range where demand was especially inelastic due to health insurance. In any event, our observations about how monopoly and health insurance interact appear to have important (though as yet unrecognized) implications for the implementation of the new Medicare prescription drug benefit. See generally Jennifer Bowman et al., Access to Cancer Drugs in Medicare Part D: Formulary Placement and Beneficiary Cost Sharing in 2006, 25 Health Aff. 1240 (2006).
52. A typical monopolist seeking maximum profit sets its unit price so that "marginal cost" (the cost of producing one additional unit) does not exceed the additional ("marginal") revenue generated by that price across all units. Because higher prices generally reduce the number of units that can be sold, there is normally a finite point beyond which a higher price per unit will net the monopolist a lower total profit, not a higher one. With health insurance helping consumers pay the monopolist's price, however, marginal cost and marginal revenue will equate at a higher level. In other words, with insurance in the picture, the monopolist's profit-maximizing price will be significantly higher, and the redistribution of income it causes will be significantly greater.
53. See generally Roger D. Blair & Jill Boylston Herndon, Physician Cooperative Bargaining Ventures: An Economic Analysis, 71 Antitrust L.J. 989 (2004).
54. For truly stunning examples of the price-increasing and profit-generating effects of combining U.S.-style health insurance and monopoly, see Geeta Anand, The Most Expensive Drugs (pts. 1-4), Wall St. J., Nov. 15, 2005, at A1 (quoting one drug company executive as saying, "I never dreamed we could charge that much");
55. Wall St. J., Nov. 16, 2005, at A1;
56. Wall St. J., Dec. 1, 2005, at A1;
57. Wall St. J., Dec. 28, 2005 at A1.
58. This series of articles reports that insurers are paying amounts up to, and even in excess of, $600,000 per patient per year for drugs needed to treat a relatively small number of individuals with rare chronic conditions. Purveyors of these drugs, which generally cost relatively little to produce, enjoy exceptionally strong protection against competition for their drugs and biologics under the Orphan Drug Act of 1983, Pub. L. No. 97-414, 96 Stat. 2049 (1983). Although other health-sector monopolists, even patent monopolists, may not possess the same market dominance as firms protected by the Orphan Drug Act, the insurers with which they deal have few tools with which to resist paying quite high prices for often small or debatable incremental benefits. See infra notes 24 & 27.
59. See Rai, supra note 17, at 208 ("The extent to which plans can engage in price/quality competition based on true cost-benefit tradeoffs... is still limited by the structure of contract and tort law. Limitations on quality that emerge from limitations on beneficial coverage will be difficult to implement in our current system.").
60. One of the instant authors has consumed much ink in trying to make the world safe for such economizing contracts. See, e.g., Clark C. Havighurst, Health Care Choices: Private Contract as Instruments of Health Reform (1995) [hereinafter Havighurst, Health Care Choice].
61. The legal system and the market have been largely unwilling to embrace them, however. See generally Mark V. Pauly, Competition and New Technology, 24 Health Aff. 1523 (2005) (recognizing lack of contractual freedom as a reason why costs of new technology are uncontrolled);
62. Clark C. Havighurst, How the Health Care Revolution Fell Short, 65 Law & Contemp. Probs. 55, 67-74 (Autumn 2002) [hereinafter Havighurst, How the Revolution Fell Short].
63. Indeed, managed health care ran into a political firestorm in the 1990s when it began to appear that health plans might invoke their contracts to ration arguably beneficial care. See generally id. at 64-100;
64. Mark A. Hall, The Death of Managed Care. A Regulatory Autopsy, 30 J. Health Pol. Pol'y & L. 427 (2005) (reviewing and explaining the dramatic decline in health plans' efforts to actively manage health care costs);
65. see also Michael E. Chernew et al., Barriers to Constraining Health Care Cost Growth, Health Aff., Nov.-Dec. 2004, at 122 (reporting post-backlash interviews with health plan administrators and finding unwillingness to constrain technology-driven cost increases).
66. Evidence of insurers' lack of either the will or the means to limit their insureds' access to costly treatments on the basis of well-calibrated benefit-cost comparisons appears in their difficulty in resisting paying even the staggering prices demanded by firms with monopolies under the Orphan Drug Act. See Anand, supra note 23, pt. 1 at A1 (reporting, however, that, in the face of Orphan Drug Act monopolies, "employers and insurers are now pushing back" but only by "excluding coverage of certain orphan drugs [or] requiring employees to pay as much as half the cost of the pricey medicines").
67. "[S]ince 1981, the Commission and DOJ have challenged relatively few hospital mergers, in some instances seeking relief only for part of the transaction.... From 1994 through 2000[,]... when there were approximately 900 hospital mergers, the Agencies and state antitrust enforcers lost all seven cases they litigated." Fed. Trade Comm'n & U.S. Dep't of Justice, Improving Health Care: A Dose of Competition ch. 4, at 1 (July 2004).
68. See generally id. at 1-33;
69. Cory S. Capps et al., Antitrust Policy and Hospital Mergers: Recommendations for a New Approach, 47 Antitrust Bull. 677 (2002).
70. On the price-increasing effects of mergers and consolidations, see David Dranove, The Economic Evolution of American Health Care 122 (2000) ("I have asked many providers why they wanted to merge. Although publicly they all invoked the synergies mantra, virtually everyone stated privately that the main reason for merging was to avoid competition and/ or obtain market power.");
71. Jack Zwanziger & Cathleen Mooney, Has Price Competition Changed Hospital Revenues and Expenses in New York?, 42 Inquiry 183 (2005) (finding that mergers undermined price- and cost-reducing effects of hospital competition following deregulation);
72. Cory Capps & David Dranove, Hospital Consolidation and Negotiated PPO Prices, Health Aff., Mar.-Apr. 2004, at 175;
73. Martin Gaynor & William B. Vogt, Competition Among Hospitals, 34 RAND J. Econ. 764, 764 (2003) ("During the second half of the 1990s, a dramatic wave of hospital consolidation occurred in the United States... [M]any local markets, including quite a few large cities such as Boston, Minneapolis, and San Francisco, have come to be dominated by two or three large hospital systems. Not surprisingly, many health plans have complained about rising prices as a result of this consolidation.").
74. On the courts' special tolerance for anticompetitive mergers of nonprofit hospitals, see infra note 41.
75. On enforcement policy toward the formation of large medical groups and physician networks, see Fed. Trade Comm'n & U.S. Dep't of Justice, Statements of Enforcement Policy in Health Care 61-105 (1996), available at http://www.ftc.gov/reports/hlth3s.pdf.
76. See also Fed. Trade Comm'n & U.S. Dep't of Justice, supra note 25, ch. 2, at 34-41.
77. The agencies' enforcement policy guidelines take the view that physician joint ventures must involve a substantial degree of "integration" - either financial risk-sharing or close cooperation in clinical matters - before the agencies will deem them anything other than naked price-fixing if they facilitate collective bargaining with payers over prices. See, e.g., letter from Jeffrey W. Brennan, Bureau of Competition, Fed. Trade Comm'n, to John J. Miles (Feb. 19, 2002) (MedSouth advisory opinion), available at http://www.ftc.gov/bc/adops/medsouth.htm (giving tentative approval, subject to re-examination in light of actual experience, to a potentially large physician joint venture in Denver that promised some clinical efficiencies in coordinating care provided by independent medical practices). In the MedSouth matter, the FTC staff neglected to consider that the venture, if truly procompetitive, should not need price-fixing to succeed because the promised efficiencies and improvements, if realized, should make participating physicians especially attractive to payers and thus, presumably, able to command higher compensation in individual, rather than collective, negotiations; also, the FTC has reported no follow-up examination of the MedSouth joint venture to verify that its consequences were not anticompetitive. More generally, one commentator has observed that the federal enforcement agencies have been slow to challenge physician or other provider networks.... Generally, they have targeted only near-monopolies and outright cartels. Further, the agencies' advisory opinions in many cases have generously extended the safe-harbor limits contained in their own policy statements. Consequently, many private attorneys advise clients that it is a relatively low risk proposition to form networks that encompass large segments of a market. In sum, agencies' failure to back up their advisory opinions with enforcement actions may have undermined the prophylactic potential of their advisories.
78. Thomas L. Greaney, Whither Antitrust? The Uncertain Future of Competition Law in Health Care, Health Aff., Mar.-Apr. 2002, at 185, 190. Even when enforcement occurs, the sanctions are generally prospective cease-and-desist orders, the threat of which creates little deterrent to future violations by other physician groups.
79. Monopoly and market power, defined as the ability to charge prices higher than marginal cost, are always matters of degree, of course. Thus, many pharmaceutical products, including those with patent protection, have reasonably close substitutes, limiting their sellers' pricing freedom. Yet price competition in markets for prescription drugs is often less than robust even after generic substitutes enter the market. See Atanu Saha et al., Generic Competition in the U.S. Pharmaceutical Industry (May 2005) (unpublished manuscript), available at http://www.analysisgroup.com/AnalysisGroup/uploadedFiles/Publishing/ Articles/Generic%20Competition%20in20the%20U.S.%20Pharmaceutical%20 Industry.pdf (finding that "each additional [generic] entrant on average is associated with a 0.2% decline in brand price. Nevertheless, unless the number of generic competitors is large, brand prices continue to rise in absolute terms.");
80. see also Alex Berenson, A Cancer Drug's Big Price Rise Disturbs Doctors and Patients, N.Y. Times, Mar. 12, 2006, at A1 (reporting large price increases of unpatented, single-source drugs for small market segments and pressure on insurers to cover such drugs). For present purposes, it suffices to understand that the pricing freedom of firms with market power is frequently greatly enhanced by health insurance even though in some circumstances insurers strengthen price competition.
81. See infra text accompanying notes 90-92.
82. See supra notes 25 & 26;
83. see also MedPac, Medicare Payment Policy 57 (2005) (noting that insurers' use of selective contracting "has been limited by both hospital consolidation and consumers' reluctance to accept limitations on their choice of providers");
84. Hall, supra note 24 (describing how the backlash against managed care weakened health plans' discretion in dealing with powerful providers). As a practical matter, hospitals' monopoly power does not appear to be exercised so much by raising the price of individual services for which there is no close substitute (geographically or otherwise) as by resisting insurer demands for steeper discounts from arbitrarily set list prices for its full range of services; thus, for example, a hospital may be able to charge an insurer $5 for an aspirin tablet because it has a powerful position in the market for open-heart surgery.
85. For insights on how hospitals usually negotiate their prices, not service-by-service but by agreeing to across-the-board discounts for a bundle of services, see Christopher P. Tompkins, Stuart H. Altman & Efrat Eilat, The Precarious Pricing System for Hospital Services, 25 Health Aff. 45 (2006).
86. Unfortunately, it is difficult to quantify hospitals' market power except by noting that hospitals' revenues from private sources tend to exceed allocations of fully distributed costs (not necessarily the true costs of treating patients insured by such private sources) by substantially greater margins than their revenues from public sources. See id. at 47. Although hard data on hospitals' monopoly profits are lacking, we see no reason to doubt our conceptual hypothesis that, although insurers keep prices well down for many things, they still enable true monopolists to charge exorbitant prices that premium payers ultimately pay.
87. See Bradley C. Strunk, Paul B. Ginsburg & John P. Cookson, Tracking Health Care Costs: Declining Growth Trend Pauses in 2004, 2005 Health Aff. (Web Exclusives) W5-286 (finding a 7.2% increase in prescription drugs spending from 2003 to 2004, accounting for 21% of the overall increase in health care costs in 2004);
88. see also Katharine Levit et al., Health Spending Rebound Continues in 2002, Health Aff., Jan.-Feb. 2004, at 147 (noting that spending on prescription drugs increased to 11% of total health care expenditures in 2002 from 7% in 1997 and that drugs accounted for 16% of the total increase in health expenditures in 2002);
89. Danzon & Pauly, supra note 20.
90. A study by Price WaterhouseCoopers found that drugs and medical devices together accounted for 22% of premium increases from 2001 to 2002. PriceWaterhouseCoopers, The Factors Fueling Rising Healthcare Costs, at tbl.1 (April 2002).
91. Richard A. Posner, Antitrust Law 13-18 (2d ed. 2001).
92. See infra notes 45 & 59.
93. Besides hospitals, private payers and providers of other kinds sometimes cross-subsidize services for which individuals pay either nothing at all or less than the cost of the services or coverage they receive. See, e.g., New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U.S. 645 (1995) (describing, and upholding against a claim of federal preemption, a complex New York statutory scheme, since repealed, designed to preserve the ability of Blue Cross, as well as some private hospitals, to finance care for individuals who would otherwise be uninsured or uninsurable);
94. see also IHC Health Plans, Inc. v. Comm'r, 325 F.3d 1188 (10th Cir. 2003) (denying federal tax exemption to health maintenance organization in part because it failed, unlike some other HMOs, to cross-subsidize indigent care, education, or research);
95. see also In re Health Care Admin. Bd., 415 A.2d 1147 (N.J. 1980) (upholding a regulation requiring nursing homes to reserve a reasonable number of their beds for indigent persons as a condition of licensure).
96. To be sure, a responsible government monopsonist would pay suppliers and providers enough to ensure continuing supplies of needed goods and services of appropriate quality - which would be in jeopardy if potential new entrants could not expect investment returns at least comparable to what they could earn elsewhere. Nevertheless, it appears that government does in fact sometimes fail to compensate providers, particularly hospitals, generously enough for them to break even in caring for their government-financed patients. See Stuart H. Altman, David Schactman & Efrat Eilat, Could U.S. Hospitals Go the Way of U.S. Airlines?, 25 Health Aff. 11, 14 (2006) ("[I]n 2003[,] Medicaid reimbursed hospitals at 92 percent of costs, and Medicare, at 95 percent.").
97. Medicaid appears to be the worst offender in this regard, since state Medicaid budgets are generally tight and legislatures often find it easier to resist provider lobbies and welfare advocates than to raise taxes. See generally Jason S. Lee et al., Medicare Payment Policy: Does Cost Shifting Matter?, 2003 Health Aff. (Web Exclusives) W3-480, W3-485 (quoting view of long-time observer that "the big cost shifter is Medicaid" and presenting graphic evidence of significant cost shifting by both Medicare and Medicaid from 1987-1992).
98. Since 1992, hospitals have, on average, had positive margins on their Medicare inpatients but have lost money on outpatients. MedPAC, Medicare Payment Policy 78-79, fig.3A-7 and tbl.3A2 (2004) (showing that inpatient margins followed a bell-shaped trajectory from 1% in 1993, to a peak of approximately 16% in 1997, to 4.7% in 2002, while outpatient margins from 2000 to 2002 were consistently below zero, fluctuating between -12% and -6%).
99. The prime example of such an "unfunded mandate" is the Emergency Medical Treatment and Active Labor Act, 42 U.S.C. § 1395dd (2003) [hereinafter EMTALA], which requires that any hospital that both maintains an emergency department and takes Medicare money must screen every patient (not just Medicare patients) coming to the emergency room and must stabilize any emergency medical condition found, all without regard to the patient's ability to pay.
100. Although a hospital is free to close its emergency room and escape this costly obligation (and a significant number have done so), most hospitals believe that their mission requires them to maintain money-losing services of this and other kinds. See infra note 40.
101. Other examples of care that is mandated but not paid for by government include laws prohibiting "discrimination" against the disabled and handicapped. See, e.g., In re Baby "K", 832 F. Supp. 1022 (E.D. Va. 1993), aff'd on other grounds, 16 F.3d 590 (4th Cir. 1994) (finding several statutory duties of hospital to treat hopeless case);
102. see also E. Haavi Morreim, Futilitarianism, Exoticare, and Coerced Altruism: The ADA Meets Its Limits, 25 Seton Hall L. Rev. 883 (1995).
103. Another example is the "free-care" obligation imposed at one time on hospitals pursuant to the Hill-Burton Act. See Am. Hosp. Ass'n v. Schweiker, 721 F.2d 170 (7th Cir. 1983).
104. See also Methodist Med. Ctr. v. Ingram, 413 N.E.2d 402 (Ill. 1980) (holding the state's police power "sufficient to justify, in proper circumstances, uncompensated deprivation of... property").
105. For an estimate that health-profession education costs hospitals $20-25 billion annually, see Bruce C. Vladeck, Paying for Hospitals' Community Service, 25 Health Aff. 34, 38 (2006).
106. Medicare allowances to hospitals frequently contain upward adjustments for "indirect medical education" and "direct medical education" (totaling $3.7 billion and $2.2 billion, respectively, in 1999). Lee et al., supra note 34 at 485.
107. See also Sean Nicholson, Medicare Hospital Subsidies: Money in Search of A Purpose 7-24, 55 tbl.3 (2002). In considering the issue of distributive justice in U.S. health care, it is appropriate to question the practice of forcing taxpayers or premium payers, or both, to subsidize the training of individuals for lucrative careers as physicians or other health professionals.
108. Vladeck, supra note 36, at 37, exh. 1.
109. Although hospitals are not generally compelled to create money-losing services, a hospital that chooses to maintain an unprofitable service, such as a burn, shock-trauma, or neonatal intensive care unit, or to become a regional referral center, is legally obligated to serve all comers without regard to ability to pay. EMTALA, 42 U.S.C. § 1395dd(g) (2000).
110. );
111. see also Nicholson, supra note 36, at 25-35.
112. Medicare is also especially generous to certain kinds of rural hospitals, and allowances paid to teaching hospitals in recognition of their educational functions, see supra note 36, are also viewed as supporting the safety net, as well as the training of health professionals.
113. It is relevant for present purposes that Medicare payments to hospitals are financed almost entirely by a flat tax on wages and salaries, not from general revenues. See infra text accompanying notes 47 & 48.
114. For an estimate that that for 2001 the net cost to hospitals for uncompensated care was $1.5-3 billion, see Jack Hadley & John Holohan, How Much Medical Care Do the Uninsured Use, and Who Pays for It?, 2003 Health Aff. (Web Exclusives) W3-66, W3-76.
115. This relatively low estimate of hospital uncompensated-care costs, derived by upwardly adjusting reports from uninsured patients themselves, was seemingly provided to suggest that the new public cost of covering the uninsured would be small. The hospital industry understandably reports a much higher burden from bad debt and uncompensated care. See First Hearing in a Series on Tax Exemption: Pricing Practices of Hospitals, Before the Subcommittee on Oversight of the House Committee on Ways and Means, 108th Cong. (2004)
116. (statement of David Bernd, Chairman of the Board of the American Hospital Association) (reporting that hospitals incurred $22 billion in uncompensated care costs in 2002).
117. See also Altman et al., supra note 34, at 14 ("[G]eneral hospitals provide a sizable amount of uncompensated care - an average of 5.5 percent of total general hospital costs, or about $25 billion, in 2003."). The rapid increase in the number of the uninsured is putting increased pressure on hospital budgets.
118. See Vladeck, supra note 36, at 37 (estimating hospitals' total community service costs at $80-95 billion in 2003).
119. This is not to say there is an actual "cost-shift," at least in the strict sense that, say, a cutback in government payments or an increase in the hospital's uncompensated-care burden translates directly into a compensating increase in prices charged to private payers. Indeed, economists are quick to point out that a firm possessing market power in any market would normally exercise that power to the fullest, whatever its other costs or obligations in other markets might be. See Michael A. Morrisey, Cost Shifting: New Myths, Old Confusion, and Enduring Reality, 2003 Health Aff. (Web Exclusives) W3-489.
120. In the real world, however, there are influences that appear to keep some hospitals, particularly nonprofit ones, from maximizing profits. See generally Allen Dobson, Joan Da Vanzo & Namrata Sen, The Cost-Shift Payment "Hydraulic": Foundation, History, and Implications, 25 Health Aff. 22 (2006);
121. Paul B. Ginsburg, Can Hospitals and Physicians Shift the Effects of Cuts in Medicare Reimbursement to Private Payers?, 2003 Health Aff. (Web Exclusives) W3-472 (suggesting that some hospital boards might deem their mission to include some obligation to charge fair prices).
122. On the other hand, managers of nonprofits have a strong interest in expanding the firm's size and maximizing its output as a way of enhancing their own authority, prestige, job satisfaction, and perquisites. The principal way they can do this is by earning excess profits whenever possible and either reinvesting them in bricks and mortar or using them to cross-subsidize activities that the market will not support. Price discrimination (that is, charging different prices to different buyers based on differences in ability or willingness to pay) is therefore common in nonprofit firms, even to the extent of charging below-cost (even zero) prices for some activities - a practice that would normally be irrational for a for-profit monopolist. Strictly speaking, such price discrimination is not cost shifting, because the hospital is rationally pursuing its own objectives, rather than being forced into involuntary spending. See Morrisey, supra. However, the burden on premium payers is the same.
123. The belief that nonprofit monopolies are largely benign appears to account for the antitrust agencies' lack of success in challenging mergers bestowing market power on nonprofit hospitals. See Barak D. Richman, The Corrosive Combination of Nonprofit Monopolies and U.S.-Style Health Insurance. Implications for Antitrust and Merger Policy, 69 Law & Contemp. Probs. 139, 142-43 (Autumn 2006) (observing the significance for antitrust policy of this article's emphasis on the extraordinary pricing freedom enjoyed by monopolists, nonprofit as well as for-profit, selling services covered by U.S.-style health insurance).
124. See also infra note 45.
125. It is well established that nonprofit hospitals are more likely than forprofits to offer unprofitable services. See Jill R. Horwitz, Does Corporate Ownership Matter? Service Provision in the Hospital Industry 2 (Nat'l Bureau of Econ. Research, Working Paper No. 11376, 2005) (also observing that nonprofits are more likely than government hospitals to offer highly profitable services). But, even though many believe that this fact alone establishes that nonprofits are in all respects socially beneficial, there remains the possibility that nonprofits frequently exercise market power and invest the resulting profits in activities to which the public would ascribe relatively little value.
126. Nonprofit corporations can be exempt from various federal, state, and local taxes if (1) their earnings accrue only to the firm and not to any private interest (which might reinvest them outside the health sector) and (2) their resources, surpluses, and other assets are used exclusively for charitable purposes, variously defined. Nonprofit hospitals, which generate supracompetitive returns in many lines of business even when the whole institution seems to struggle, have been treated with special favor under federal law, which accepts "the promotion of health" in a community as a purpose charitable enough for tax exemption without regard to the amount of indigent care provided. Rev. Rul. 69-545, 1969-2 C.B. 117.
127. Such attention as these firms receive nearly always focuses only on similarities and differences in the behavior of nonprofit and for-profit firms and on the appropriateness of tax exemptions for the former. See, e.g., Mark Schlesinger & Bradford Gray, How Nonprofits Matter in American Medicine, and What to Do About It, 2006 Health Aff. (Web Exclusives) W6-287 (reviewing empirical literature comparing the behavior of nonprofits and for-profits in the health sector);
128. David A. Hyman & William M. Sage, Subsidizing Health Care Providers Through the Tax Code: Status or Conduct?, 2006 Health Aff. (Web Exclusives) W6-312 (suggesting that nonprofits' tax subsidies be tied to quantifiable measures of performance).
129. This article emphasizes significantly larger issues, particularly the unfairness of systematically financing costly public services by undue impositions on premium payers rather than out of public funds. See Clark C. Havighurst, The Debate over Health Care Cost-Containment Regulation: The Issues and the Interests, in Incentives vs. Controls in Health Policy: Broadening the Debate 9 (Jack A. Meyer ed., 1985) (opining that, at least in the mid-1980s, "cross-subsidies in the health care industry... may constitute the most entrenched, most extravagant, and least closely-supervised government-tolerated use of private monopoly to generate revenues for public purposes anywhere in the U.S. economy").
130. Many hospitals enjoy much of their market power solely because of regulatory protection (under so-called certificate-of-need laws) against competition that would undermine their ability to generate revenues needed for seemingly worthy purposes. Regulation that confers monopoly power on private interests as a quid pro quo for providing publicly approved services has been characterized as "taxation by regulation." Richard A. Posner, Taxation by Regulation, 2 Bell J. Econ. & Mgmt. Sci. 22 (1971).
131. But see infra note 212.
132. To be sure, monopoly's extraordinary profitability in the health sector can also induce allocative inefficiency in the form of less-good things, including wasteful spending in pursuit and defense of market dominance. See supra text accompanying notes 31 & 32.
133. Examples of such rent-seeking behavior in the hospital industry include advertising, see Robert J. Town & Imran Currim, Hospital Advertising in California, 1991-1997, 39 Inquiry 298 (2002);
134. Although cross-subsidization is sometimes analogized to a system of taxation, its unfair regressivity as a kind of head tax is generally not observed. E.g., Dobson et al., supra note 40, at 30-31 (likening the cost shift to a premium tax without observing its regressivity);
135. Holahan & Zedlewski, supra note 6, at 236 (treating uncompensated care "as a kind of premium surtax for families with private insurance that is transferred to those who receive uncompensated care").
136. Such subsidies represent roughly twelve percent of total Medicare payments for hospital inpatient care. Nicholson, supra note 36.
137. See supra note 1. An additional inequity results to the extent that higher-income employees contribute higher percentages of their earnings to tax-favored retirement plans, thereby escaping payroll taxes that, unlike deferred income taxes, are not recaptured when plan accumulations are distributed.
138. See Tompkins et al., supra note 29;
139. Uwe E. Reinhardt, The Pricing of U.S. Hospital Services: Chaos Behind a Veil of Secrecy, 25 Health Aff. 57 (2006).
140. For a review of legal and other issues raised by such discriminatory pricing by hospitals, see Mark A. Hall, Paying for What You Get and Getting What You Pay for: Legal Responses to Consumer-Driven Health Care, 69 Law & Contemp. Probs. 159, 161-65 (Autumn 2006).
141. See Melissa B. Jacoby, Teresa Sullivan & Elizabeth Warren, Rethinking the Debates over Health Care Financing: Evidence from the Bankruptcy Courts, 76 N.Y.U. L. Rev. 375 (2001);
142. David U. Himmelstein et al., Illness and Injury As Contributors to Bankruptcy, 2005 Health Aff. (Web Exclusives) W5-63.
143. But see David Dranove & Michael L. Millenson, Medical Bankruptcy: Myth versus Fact, 2006 Health Aff. (Web Exlusives) W74 (finding earlier estimates of frequency of bankruptcies caused by health costs to be exaggerated).
144. See also Hall, supra note 50, at 163 (reporting that hospitals are rapidly reducing their efforts to "tax" at least the lower-income uninsured)
145. On class-action litigation challenging such discriminatory pricing and other alleged neglect of nonprofit hospitals' charitable mission, see id. at 162-63.
146. In economic theory, the usual monopolist's potential rewards are limited to the "consumer surplus" (measured by the maximum prices that individual consumers would pay, as reflected in the demand curve) that consumers would enjoy if they could purchase the service or product at a competitive price. In reality, given the difficulty of price discrimination (that is, charging different prices to different buyers, depending on where they appear on the demand curve), ordinary monopolists cannot hope to capture all consumer surplus, let alone the value of any positive externalities the product may yield. Incentives to pursue valuable innovations are therefore suboptimal in the absence of insurance. The innovation literature leaves no doubt that suboptimal investments are made in basic scientific research, e.g., Richard R. Nelson, The Simple Economics of Basic Scientific Research, 67 J. Pol. Econ. 297 (1959), although there is some disagreement over whether investments in applied research and development of specific goods and services are also suboptimal.
147. See generally Antitrust, Innovation, and Competitiveness (Thomas M. Jorde & David J. Teece eds., 1992).
148. Darius Lakdawalla & Neeraj Sood, Insurance and Innovation in Health Care Markets (Nat'l Bureau of Econ. Research, Working Paper No. 11602, 2005).
149. For a theoretical demonstration, using different assumptions about pricing than those used by Lackdawalla and Sood, that universal health insurance, even with optimal copayments, may yield excessive incentives for innovation, see Alan M. Garber, Charles I. Jones & Paul M. Romer, Insurance and Incentives for Medical Innovation (Nat'l Bureau of Econ. Research, Working Paper 12080, 2006).
150. A series of studies by such researchers as David Cutler and Frank Lichtenberg show large social gains from investing in new medical and pharmaceutical technology. See infra note 68. Such studies appear to justify all spending on technology development because the consumer surplus generated by life-saving and health-improving technologies is so large, easily exceeding the social costs incurred.
151. Indeed, the authors' seeming lack of concern about marginal trade-offs in the adoption of new technologies, see infra note 68, appears to be attributable to their sense that, in general, innovation incentives are seriously suboptimal. Certainly, the case for valuing R&D very highly is strong, particularly if one takes an international perspective.
152. See, e.g., Frank R. Lichtenberg, The Impact of New Drug Launches on Longevity: Evidence from Longitudinal Disease-Level Data from 52 Countries. 1982-2001 (Nat'l Bureau of Econ. Research, Working Paper No. 9754, 2003) (attributing forty percent of world-wide decline in mortality in the period studied to the introduction of innovative drugs). (Many health problems in third-world nations would yield to new technologies, yet those nations are too poor to offer innovators much incentive to produce them, and strict price controls on many health-related technologies in other developed nations would seem to unduly diminish incentives for R&D, making the U.S. market the principal source of such incentives.) The potentially large social benefits of health-related R&D notwithstanding, however, it is far from clear to us that those benefits should be given full weight in assessing U.S. health policy and used to justify imposing their heavy costs on lower-and middle-income premium payers in U.S. health plans.
153. See infra note 154 (discussing and questioning so-called Kaldor-Hicks efficiency, which permits gains to winners to justify imposing costs on others without regard to distributional consequences).
154. See Garber et al., supra note 56, at 14 (illustrating how, even with well-designed insurance, "profits earned by [a] pharmaceutical company can exceed the consumer surplus associated with the drug treatment")
155. As in the analogous case of hospital cross-subsidies, we are reluctant to accept the argument that large social benefits justify maintaining innovation incentives, without regard to who bears the cost burden, in large part because of Richard Posner's observation that the prospect of large monopoly profits can induce wasteful as well as productive behavior. See supra text accompanying notes 31 & 32.
156. For example, there is a significant literature on the wasteful expenditures that the patent system can induce - even without the prospect of extraordinary monopoly profits of the kind that U.S.-style health insurance makes possible. See, e.g., Mark F. Grady & Jay I. Alexander, Patent Law and Rent Dissipation, 78 Va. L. Rev. 305, 308 (1992) ("The defect of the system is that if multiple inventors expend resources in competition for the patent monopoly, the benefit to society of having the invention will be dissipated by the cost of numerous, redundant, development efforts.").
157. See also Yoram Barzel, Optimal Timing of Innovations, 50 Rev. Econ. Statistics 348, 349 (1968) ("[C]ompetition among potential innovators may deprive innovations of all their special economic value." In addition to the wastefulness of so-called "patent races," the social value of innovations may be further dissipated in promoting, attacking, defending, and inventing around valuable patents. In the health care sector, for example, one sees extensive efforts to create and heavily promote relatively modest product improvements and to differentiate brand-name products from nearly equivalent generics by heavy investment in direct-to-consumer advertising. In particular, there is evidence that the decreased price elasticity of demand for brand-name drugs following the expansion of insurance coverage for prescription drugs in the 1990s increased manufacturers' expected returns from promotional efforts, many of which are socially unproductive.
158. Danzon & Pauly, supra note 20.
159. See infra Part IV.C.
160. The importance of the practical inability of consumers to purchase health insurance providing only selective coverage of costly technology has recently been recognized by economist Mark Pauly. Pauly, supra note 24.
161. See also Donald W. Moran, Whence and Whither Health Insurance? A Revisionist History, 24 Health Aff. 1415, 1423 (2005) (predicting that, with the advent of so-called consumer-directed health care, see infra text accompanying notes 93-98, consumer cost-consciousness and competitive pressure on health plans to control big-ticket items "will push back strongly against manufactures' pricing flexibility. Increasingly, the question of whether a high-cost technology is covered at all by an isurance plan will become the most important determinant of product economics.")
162. United States v. Aluminum Co., 148 F.2d 416, 430 (2d Cir. 1945) (famous antitrust opinion by Learned Hand, J., acknowledging that even a powerful monopoly might be lawful if gained and maintained by "superior skill, foresight, and industry" and not by unlawful "monopolizing" behavior).
163. See supra text accompanying notes 25-27 on the wide variety of factors, many of them fortuitous, that may explain why a seller is free to set unduly profitable prices in health care markets.
164. See supra text accompanying notes 42-45.
165. Cf. Posner, supra note 44.
166. One justification often given for exempting certain private entities from public taxation is that they assume burdens that the public would otherwise have to bear in educating children or caring for the uninsured sick. See, e.g., Utah County v. Intermountain Health Care, Inc., 709 P.2d 265, 278 (Utah 1985) (denying state tax exemption to hospitals in the absence of "the essential element of gift to the community, either through nonreciprocal provision of services or through the alleviation of a government burden"). This "burden theory" is usually invoked, however, without any determination that, but for the private entity's efforts, the public sector would in fact incur costs equivalent to the tax relief granted. In other words, a judge or public official inclined to favor government generosity in general or public support for the particular activity in question (e.g., health care) might confer a tax exemption under the burden theory even if, had it come to a vote, the legislature would not have been so generous. Reasonable minds differ over whether, in an ostensible democracy, such spending choices should be made exclusively by elected officials.
167. This apparent (though unproven) inequity is discussed at length in Part III.
168. We have not found in economics literature anyone expressly hypothesizing, as we do, that U.S.-style health insurance, by substantially increasing actual and anticipated returns to both nonprofit and for-profit monopolies in the health sector, not only adds to monopoly's redistributive effects but also induces allocative inefficiencies both different from and more troublesome than those usually associated with monopoly. But see Garber et al., supra note 56 (2006 working paper demonstrating that health insurance "creates incentives for a monopoly provider of a pharmaceutical to charge far more for its product than it otherwise would").
169. Reasons why economists may be less concerned than we are about monopoly's effects in the health care sector include their sense that deadweight loss is not a serious problem, see supra note 212,
170. and their professional agnosticism about the welfare implications of redistributing income, supra note 18.
171. In addition, they may assume that unlawful monopoly is both rare and a matter for antitrust enforcement when it does occur and that lawful monopoly is either a natural phenomenon or a transient reward for entrepreneurial endeavor. In any event, our concern is not about monopoly as such but about how it and health insurance interact. Here the problem is that economists do not always appreciate the significance (see Part IV.C.) of de facto and de jure limits on insurers' willingness and ability to optimally counteract moral hazard. See, e.g., Gaynor, et al., supra note 17 (considering whether, in theory, imperfect competition in medical markets might, by raising prices, offset the misallocative tendencies of moral hazard, but answering the question only on the assumption that, contrary to our observation, the insurance industry "is competitive [and] chooses insurance policies...that maximize consumer welfare").
172. But see Pauly, supra note 24.
173. The articles we have found that are at all close to our theme include Danzon and Pauly's demonstration of how broader insurance coverage and new technologies reciprocally induce each other. Danzon & Pauly, supra note 20 (observing that insurance increases inelasticity of demand and may thus "may affect... the launch price of new drugs").
174. See also Garber et al., supra note 56 (demonstrating that health insurance facilitates pricing of pharmaceuticals in ways that, in theory, may create excessive incentives for innovation).
175. See, e.g., 2004 Economic Report of the President, H.R. Doc. 108-145 (2004), 24, 189 ("Over-reliance on health insurance as a payment mechanism leads to an inefficient use of resources in providing and utilizing health care.").
176. See generally Charles E. Phelps, Health Economics 325 (2003) (describing moral hazard as the "induced demand due to the health insurance coverage").
177. Recent work by Amy Finkelstein estimates that the spread of health insurance from 1950 to 1990 (including the implementation of Medicare and Medicaid) accounted for at least forty percent of the dramatic increase in per capita health spending during that period. Amy Finkelstein, The Aggregate Effects of Health Insurance. Evidence from the Introduction of Medicare (Nat'l Bureau of Econ. Research, Working Paper No. 11619, 2005).
178. Although public and private insurance provided both valuable financial security and subsidized access to essential health services, the moral-hazard effect detected by Finkelstein is substantially greater than economists had previously detected in studies of individual behavior under various insurance arrangements (for example, the RAND Health Insurance Experiment). In contrast to the earlier studies, Finkelstein's long time horizon enables her to detect long-term market-wide effects induced by the substantially steeper demand curves that sellers increasingly faced as health insurance spread. These effects include greatly altered styles of medical practice and strong incentives to create and use technologies that would not pass most people's benefit-cost test. Finkelstein and others have also shown that Medicare, which provided only very limited coverage for prescription drugs, did not appreciably stimulate pharmaceutical innovation. Daron Acemoglu et al., Did Medicare Induce Pharmaceutical Innovation? 2 (Nat'l Bureau of Econ. Research, Working Paper No. 11949, 2006).
179. For looks at the other side of the same coin, observing that markets with large HMO market shares featured slower diffusion of new technologies and correspondingly lower health care expenditures, see Laurence C. Baker, Managed Care and Technology Adoption in Health Care:
180. see also Laurence C. Baker et al., The Relationship Between Technology Availability and Health Care Spending, 2003 Health Aff. (Web Exclusives) W3-537, W3-547-48 (acknowledging slowing the rate of availability of technology may slow the spending growth rate).
181. Precisely because health care generates such large amounts of consumer surplus compared to other industries (because the value of its benefits to individuals - when it yields benefits - can be so great, perhaps the difference between life and death), it is difficult to think about health care at the margin, where costs are apt to be large and benefits small, especially in probabilistic terms. Occasional studies seek to justify large and increasing expenditures on health care by noting the huge benefits obtained in recent years from improving life expectancy for, say, victims of heart disease or afflicted neonates. E.g., David M. Cutler, Your Money or Your Life: Strong Medicine for America's Health Care System 63 (2004) ("On the basis of low-birth-weight-infant- and cardiovascular-disease-care alone, therefore, the benefits of medical care are about equal to its costs.");
182. David M. Cutler & Mark McClellan, Is Technological Change in Medicine Worth It?, Health Aff., Sept.-Oct. 2001, at 11 (examining five new technologies and concluding from these examples that "medical spending as a whole is clearly worth the cost" (emphasis added));
183. Lichtenberg, supra note 57;
184. Frank R. Lichtenberg, Are the Benefits of Newer Drugs Worth Their Cost? Evidence from the 1996 MEPS, Health Aff., Sept.-Oct. 2001, at 241;
185. Frank R. Lichtenberg, Pharmaceutical Innovation, Mortality Reduction, and Economic Growth (Nat'l Bureau of Econ. Research, Working Paper No. 6569, 1998) (concluding that "a one-time R&D expenditure of about $15 billion subsequently saves 1.6 million life-years per year, whose annual value is about $27 billion").
186. Such studies divert attention from the all-important margin, however. See Jonathan S. Skinner, Douglas O. Staiger & Elliott S. Fisher, Is Technological Change In Medicine Always Worth It? The Case of Acute Myocardial Infarction, 25 Health Aff. 34 (2006) (questioning Cutler and McClellan, supra, on cost-justifications for increased spending on cardiovascular disease after 1996; also noting, from regional data, lack of correlation between spending increases and improvements in survival rates);
187. see also Havighurst, How the Revolution Fell Short, supra note 24, at 80 n.89 (criticizing Cutler & McClellan, supra).
188. Many investments in R&D produce little, if any, useful knowledge, and many new technologies represent only marginal improvements, at best, over earlier, cheaper treatments. The crucial fact remains that the system lacks accepted mechanisms for comparing marginal benefits and costs, especially in cases in which a patient hoping for relief naturally demands heavy spending on his own behalf. See supra note 24. Indeed, because the public resists any recognition of tradeoffs, it is politically dangerous even to suggest in a public forum that people might be better off with less rather than more health care. Moreover, any policymaker concerned that health care may be claiming too large a piece of the economic pie must also reckon with the political strength of the health care industry. And, finally, it is hard to argue with the industry's consistent ability to attract capital and create jobs in an otherwise up-and-down economy or with its seemingly miraculous technical accomplishments. Largely for these reasons, it is only rising health care costs that ever trigger political concern. Whatever its magnitude, inefficiency that is already embedded in the economy is simply never going to be viewed as a problem by the political class.
189. See Havighurst, Health Care Choices, supra note 24, at 89-92 (putting high and rising health care costs in political and economic perspective).
190. To the extent that health insurance reduces the apparent price of the insured service below its marginal cost, it introduces a welfare-loss triangle comparable to the deadweight-loss triangle it helps to eliminate in the case of monopoly. See supra note 17.
191. This triangle lies above the service's demand curve, however, and represents not underconsumption, but arguable overconsumption, of services - specifically, services that consumers would not have found worth purchasing at marginal cost (the competitive price). But see infra note 212 (expressing caveat about relying too mindlessly on demand curves as indicators of welfare). The case of patented pharmaceuticals is anomalous because their marginal cost of production is usually very low. Even a co-insurance payment of twenty percent might therefore exceed it, creating at least a modest deadweight loss of the usual kind. As a substitute for insured (and therefore underpriced) hospital and other medical care, it seems likely that drugs are used less in therapy than they would be under an efficient pricing system - that is, if everything were priced at marginal cost.
192. See infra text accompanying note 209.
193. The article we have found that comes closest to expressing insights on the latter issue that are similar to our own is Daniel Shapiro, Why Even Egalitarians Should Favor Market Health Insurance, 15 Soc. Phil. & Pol'y 84 (1998).
194. By the same token, it is inefficient, strictly speaking, for health insurance to cover a service that easily passes a benefit-cost test in a particular case if covering it generally would generate moral-hazard or administrative costs reversing the benefit-cost calculation. See infra note 75.
195. Part IV of this Article focuses more directly on the legal system's responsibility for the regrettable state of U.S. health care, including its effects in raising the costs of resisting moral hazard.
196. On the regressive consequences of cost sharing as a method of counteracting moral hazard, see infra notes 105-108 and accompanying text.
197. The nation's failure to provide health coverage for the uninsured is in large measure a consequence of the high and (currently) uncontrollable moral-hazard costs that such coverage would certainly entail. Even so, it would be hard (in both senses of the word) to argue that present policy toward the uninsured is, on this basis, actually "efficient." See supra note 72. Our point instead is that the insurance gap exists and widens over time largely because U.S. health insurers are unduly limited by custom, law, and regulation in what they may or can do to contain moral hazard. Indeed, our main argument is that these de facto and de jure restraints on health insurers' ability to administer their premium pools are part of a larger pattern of health policies that, whether intentionally or not, burden middle- and lower-income Americans in order to serve other interests. That they also contribute heavily to the plight of the uninsured, while in some respects a separate matter, should certainly be of equivalent concern. For a study showing the various ways in which rising health insurance premiums are adversely affecting lower-wage workers, see Katherine Baicker & Amitabh Chandra, The Labor Market Effects of Rising Health Insurance Premiums (Nat'l Bureau of Econ. Research, Working Paper No. 11160, 2005) (estimating that a 10% increase in health insurance premiums results in a 1.6% reduction in the aggregate probability of being employed, an increase of 1.9% in the likelihood that a worker will be employed only part-time, and 2.3% decrease in wages for those who remain employed with employer-sponsored health insurance; for hourly workers, such a premium increase will reduce hours worked by 1% and reduce the probability of being offered health insurance by 3.8 %; also finding it likely that workers covered by employer insurance will "bear the full incidence of increases in health insurance premiums").
198. See also Michael Chernew et al., Increasing Health Insurance Costs and the Decline in Insurance Coverage, 40 Health Servs. Res. 1021, 1034 (2005) (finding that rising health insurance premiums accounted for over half of the decline in health insurance coverage during the 1990s).
199. See generally Timothy P. Blanchard, "Medical Necessity" Determinations - A Continuing Healthcare Policy Problem, 37 J. Health L. 599 (2003);
200. William Sage, Managed Care's Crimea: Medical Necessity, Therapeutic Benefit, and the Goals of Administrative Process in Health Insurance, 53 Duke L.J. 597 (2003);
201. Einer Elhauge, The Limited Regulatory Potential of Medical Technology Assessment, 82 Va. L. Rev. 1525 (1996).
202. The no-man's-land metaphor is explained graphically in Clark C. Havighurst & James F. Blumstein, Coping with Quality/Cost Trade-offs in Medical Care: The Role of PSROs, 70 Nw. U.L. Rev. 6,15-20 (1975).
203. See also Havighurst, Health Care Choices, supra note 24, at 93-96.
204. Although not labeled as such in earlier text, the extra pricing freedom that health insurance confers on monopolists is itself a manifestation of moral hazard, flowing as it does from the removal of consumers' price-consciousness in purchasing insured services.
205. Anderson et al., supra note 13;
206. Reinhardt et al., supra note 8.
207. See infra note 197.
208. See Jonathan Skinner, Elliott Fisher & John Wennberg, The Efficiency of Medicare, in Analyses in the Economics of Aging (D. Wise, ed., 2005);
209. Elliot S. Fisher et al., The Implications of Regional Variations in Medicare Spending. Part 1: The Content, Quality, and Accessibility of Care, 138 Annals of Internal Med. 273 (2003);
210. Elliot S. Fisher et al., The Implications of Regional Variations in Medicare Spending. Part 2: Health Outcomes and Satisfaction with Care, 138 Annals of Internal Med. 288 (2003);
211. Katherine Baicker & Amitabh Chandra, Medicare Spending, the Physician Workforce, and the Beneficiaries' Quality of Care, 2004 Health Aff. (Web Exclusives) W4-184.
212. Studies to show inappropriate utilization were common in the 1980s and early 1990s, many of them conducted under the auspices of the RAND Corporation. See, e.g., Mark R. Chassin et al., Indications for Selected Medical and Surgical Procedures: A Literature Review and Ratings of Appropriateness: Coronary Artery Bypass Graft Surgery (1986);
213. Robert H. Brook et al., Predicting the Appropriate Use of Carotid Endarterectomy, Upper Gastrointestinal Endoscopy, and Coronary Angiography, 323 New Eng. J. Med. 1173 (1990);
214. Lucian L. Leape et al., The Appropriateness of Use of Coronary Artery Bypass Graft Surgery in New York State, 269 J. Am. Med. Ass'N 753 (1993);
215. Lee H. Hilborne et al., The Appropriateness of Use of Percutaneous Transluminal Coronary Angioplasty in New York State, 269 J. Am. Med. Ass'n 761 (1993);
216. Steven J. Bernstein et al., The Appropriateness of Use of Coronary Angiography in New York State, 269 J. Am. Med. Ass'n 766 (1993).
217. In order to ensure acceptance by the medical profession, the earlier RAND studies relied heavily on professional opinion without asking physicians to compare benefits and costs, yet still found numerous surgical and other procedures to be significantly overused. See, e.g., Rolla Edward Park et al., Physician Ratings of Appropriate Indications for Six Medical and Surgical Procedures 6 (1986).
218. In later studies, in order to identify spending of truly marginal benefit, the researchers asked physicians to focus, not just on medical necessity or appropriateness, but on "cruciality," producing an even more troublesome picture. See, e.g., Paul P. Lee et al., Cataract Surgery: A Literature Review and Ratings of Appropriateness and Cruciality 48-50, 163-274 (1993).
219. These studies and their findings are discussed in some detail in Havighurst, Health Care Choices, supra note 24, at 254-62.
220. Phelps, Supra note 66, at 356-57 (1997) (estimating, based on empirical estimates of demand for insurance, that "employer-group health insurance premiums would be only about 55 percent as large today if the tax subsidy were not in effect"; "it seems possible that the health sector would be at least 10 to 20 percent smaller without the tax subsidy for health insurance").
221. See generally Mark V. Pauly, Taxation, Health Insurance, and Market Failure in the Medical Economy, 24 J. Econ. Lit. 629 (1986). A complementary subsidy allows deductions for health insurance premiums paid by self-employed individuals. The total 2004 cost, in revenue forgone, of the various federal and state "tax expenditures" subsidizing privately purchased health insurance has been estimated at $209.9 billion.
222. John Sheils & Randall Haught, The Cost of Tax-Exempt Health Benefits in 2004, 2004 Health Aff. (Web Exclusives) W4-106. Other tax subsidies with possible implications for fairness include the deduction allowed (for itemizers) for out-of-pocket health expenses above 7.5% of adjusted gross income and the favorable tax treatment of health spending through flexible spending accounts. On new tax breaks for health savings accounts, see infra notes 93-98 and accompanying text.
223. For employees in the highest federal and state tax brackets, the saving achievable may exceed fifty percent.
224. E.g., Madison Powers & Ruth Faden, Social Justice: The Moral Foundations of Public Health and Health Policy 132-33 (2006). It is estimated that 26.7% of the economic benefit they confer accrues to the 14% of taxpayers with annual incomes of at least $100,000; likewise, "only 28.4 percent of all [these] tax expenditures will go to families with incomes below $50,000, even though this group contains 57.5 percent of all U.S. families."
225. Sheils & Haught, supra note 84, at 110. But for this substantial regressivity, characterizing the exclusion from taxable income as a "subsidy" might not be appropriate at all, since a taxing authority might simply find it fairer to tax individuals' income only after certain basic necessities were provided for. But the subsidy here is decidedly not limited to encouraging the purchase of only basic coverage.
226. See Lawrence Zelenak, Of Head Taxes, Income Taxes, and Distributive Justice in American Health Care, 69 Law & Contemp. Probs. 103,113 (Autumn 2006).
227. For a study showing awareness of this point, see Holahan & Zedlewski, supra note 6, at 235, 240 tb1.6 (in estimating total health care cost burdens borne by various deciles of consumer-taxpayers, "we calculate income and payroll taxes that are required to finance the employer-paid health benefit tax exclusion"). Many believe that tax expenditures - exclusions and deductions designed to achieve substantive policy objectives - are generally unwise as a matter of tax policy because they shrink the tax base, thus requiring higher marginal tax rates on other income to produce the same amount of revenue. These higher rates, it is believed, adversely affect overall productivity. In any event, it is unclear that higher-income Americans need any subsidy at all to encourage their purchase of essential health coverage.
228. See Havighurst, How the Revolution Fell Short, supra note 24, at 70 ("[Because] only a generous plan sends workers the message most employers want to convey about their concern for worker welfare... and also because health benefits are useful in attracting and keeping workers with the best chances of being hired by someone else, employee health plans are likely to be costlier than even the average worker would demand (even with tax subsidies enhancing his purchasing power).").
229. See infra text accompanying note 100.
230. See supra note 4. Employers only rarely offer their workers more than one health care option with the employee required to pay the full additional cost above the cost of the lowest-price plan.
231. See Jonathan Gruber & Alan B. Krueger, The Incidence of Mandated Employer-Provided Insurance: Lessons from Workers' Compensation Insurance 1 (Princeton Univ. Indus. Relations Section, Working Paper No. 279, 1990) (finding that areas and industries with high workers-compensation costs led to a corresponding reduction in wages);
232. Jonathan Gruber, The Incidence of Mandated Maternity Benefits, 84 Am. Econ. Rev. 622 (1994) (finding that the mandated expansion of health insurance coverage in the Pregnancy Discrimination Act of 1978 led to reduced wages);
233. Louise Sheiner, Health Care Costs, Wages, and Aging (Apr. 1999) (unpublished manuscript, on file with Fed. Reserve Bd. of Governors), available at http://www.federalreserve.gov/pubs/feds/1999/199919/ 199919pap.pdf (finding relatively lower wages for older workers and workers with family coverage in areas with high medical prices).
234. See Alain C. Enthoven, The Fortune 500 Model for Health Care: Is Now the Time to Change?, 27 J. Health Pol. Pol'y & L. 37 (2002);
235. see also infra note 119.
236. see Enthoven, supra note 90. Regressivity would not be a significant problem under Enthoven's prescription - assuming that incremental costs are calculated actuarially, with appropriate awareness of different groups' propensity to use their coverage.
237. The only time that consumers may actually see the true cost of health coverage is in choosing between jobs that do and do not carry health benefits. See Mark V. Pauly, Health Benefits at Work: An Economic and Political Analysis of Employment-Based Health Insurance (1997);
238. see also supra note 4. Although wage differentials between otherwise similar jobs have been noted, such all-or-nothing choices, with health care as only one of many considerations being weighed, are not clearly reliable expressions of consumer preferences with respect to marginally beneficial care.
239. See Jonathan Gruber & Alan B. Krueger, The Incidence of Mandated Employer-Provided Insurance: Lessons from Workers' Compensation Insurance 1 (Princeton Univ. Indus. Relations Section, Working Paper No. 279, 1990) (finding that areas and industries with high workers-compensation costs led to a corresponding reduction in wages);
240. Jonathan Gruber, The Incidence of Mandated Maternity Benefits, 84 Am. Econ. Rev. 622 (1994) (finding that the mandated expansion of health insurance coverage in the Pregnancy Discrimination Act of 1978 led to reduced wages);
241. Louise Sheiner, Health Care Costs, Wages, and Aging (Apr. 1999) (unpublished manuscript, on file with Fed. Reserve Bd. of Governors), available at http://www.federalreserve.gov/pubs/feds/1999/199919/ 199919pap.pdf (finding relatively lower wages for older workers and workers with family coverage in areas with high medical prices).
242. The political consequences of the tax subsidy are rarely recognized, even by scholars. But see Havighurst, How the Revolution Fell Short, supra note 24, at 78-86.
243. Pub. L. No. 108-173, § 101, 117 Stat. 2066 (2003).
244. The HDHP must have an annual deductible anywhere from $1000 to $5000 for single coverage or from $2000 to $10,000 for families, in which event untaxed dollars may be contributed to the individual's HSA each year in amounts up to the lesser of $2250 for an individual ($4500 for a family) or the HDHP's deductible. (Because the stated amounts are indexed for inflation, they are understated here). HSAs, which may be maintained with either a financial institution or an insurer, can then accumulate investment earnings tax-free. Account owners may draw on these funds at any time, without tax on the distribution, to pay a broad range of "qualified medical expenses," broadly defined. See infra note 96.
245. For arguments for expanding the scope of HSAs even further, see Michael F. Cannon, Health Savings Accounts: Do the Critics Have a Point?, Pol'y Analysis (Cato Inst., Washington, D.C.) May 30,2006, at 1, available at http://www.cato.org/pub_display.php?pub_id=6395.
246. See Pauly, supra note 84; see also infra note 217 (suggesting limited, refundable tax credits as a more practical and equitable way to subsidize and universalize basic coverage).
247. It is relevant to our thesis in this article that the strategy chosen creates, whether by accident or design, an important new tax shelter for the well-to-do. Lower-bracket taxpayers will not, it seems certain, be in a position to take equivalent advantage of this new tax break because, in addition to deriving smaller tax savings than those with higher incomes, they will find it harder to deposit the full amounts allowed. Wealthier taxpayers, on the other hand, not only can afford to deposit more untaxed money in their HSAs but also can elect to spend after-tax dollars on their medical care, leaving their HSA funds to accumulate tax-free, as in a tax-favored retirement plan. (After an HSA owner attains age sixty-five, any funds remaining in the HSA may be either withdrawn as taxable income or rolled over into a tax-favored retirement account for further tax deferral.) The effect is to give higher-bracket taxpayers yet another means of deferring taxes on present income and of escaping payroll taxes altogether. See Eric Dash, Wall Street Senses Opportunities in Health Care Savings Accounts, N.Y. Times, Jan. 27, 2006, at A1, A16 ("Not since the creation of the individual retirement account in the mid-1970's has such a potentially huge mountain of money landed in the lap of the financial services industry."). A very strong tax-equity argument can be made for making earnings on HSAs taxable with individuals' other income.
248. The need to manage health benefits and moral hazard will be essentially unaffected by the CDHC reforms because the great majority of health care spending occurs in cases that would quickly consume the contemplated deductibles. See generally Mark A. Hall & Clark C. Havighurst, Reviving Managed Health Care with Health Savings Accounts, 24 Health Aff. 1490, 1490-91 (2005) (arguing that "combining managed care with HSAs can help to re-legitimize managed care in the public eye by clarifying the respective decision-making responsibilities of health plans and patients.... Specifically, the availability of HSAs should make it clearer to most people that plans' denials of coverage are not meant to ration health care itself but only to limit the availability of third-party financing.").
249. See also infra notes 148-50 and accompanying text.
250. See Uwe E. Reinhardt, Resource Allocation in Health Care: The Allocation of Lifestyles to Providers, 65 Milbank Q. 153 (1987) (emphasizing the extent to which consumer savings from enhanced price competition would come at the direct expense of industry insiders). Although many people admire the health care industry for providing numerous secure, well-paying jobs and for being relatively recession-proof, these features are a direct consequence of a financing system that denies consumers easy opportunities to economize when family budgets are squeezed. Moreover, the reciprocal effect of the health sector's relative stability is greater instability in the rest of the economy, including lessened job security for those who work outside the health sector but pay tribute to it through their health insurance premiums. This is simply one more unrecognized way in which U.S. health policy hurts working Americans while benefiting stakeholders in the health sector.
251. To be sure, generous health benefits are also found in plans that some employers maintain only for lower- and middle-income, usually unionized, workers. Labor unions' persistent demands for especially generous benefits do not disprove, however, our contention that rich health plans are contrary to the true interests of middle-income workers. Instead, they prove only that, like politicians, union leaders (another elite) know how to exploit, in their own interest, workers' mistaken belief that the employer alone bears the cost of their health benefits.
252. For recent journalism implying both surprise and concern that health services often vary according to the patient's income, see Janny Scott, Life at the Top in America Isn't Just Better, It's Longer, N.Y. Times, May 16, 2005, at A1.
253. But demand for much health care is income-elastic, meaning that people naturally spend more on it as their incomes increase. See also Renee Mentnech et al., An Analysis of Utilization and Access from the NHIS: 1984-92, Health Care Fin. Rev. Winter 1995, at 51, 55-56 (concluding from National Health Interview Survey data that among patients in relatively good health, higher income increases the probability of a physician visit);
254. see generally Phelps, supra note 66, at 148-49. Studies of individuals under full insurance indicate that income elasticity is positive but rather small.
255. See Emmett B. Keeler et al., The Demand for Episodes of Medical Treatment in the Health Insurance Experiment (1988) (data from RAND experiment showing relatively small income effects, elasticities of about 0.2).
256. Income elasticities estimated from time series data, which capture the effects of new medial technologies, are higher and approach unity. See Martin Feldstein, Hospital Cost Inflation: A Study of Nonprofit Price Dynamics, 60 Am. Econ. Rev. 853-72 (1971);
257. Catherine McLaughlin, HMO Growth and Hospital Expenses and Use: A Simultaneous-Equations Approach, 22 Health Services Res. 183-202 (1987) (finding an income elasticity of 0.7 using data from 1972 to 1982).
258. Although proper measurement of income elasticity would focus on permanent income, most correlation studies use data reflecting transitory income, thus diluting the apparent effect of income on health expenditures if households encountering sudden sickness both earn below their normal income and increase their consumption of health care. In any event, studies show that, for many health services, income elasticity exceeds unity, which means percentage increases in income translate into even greater percentage increases in spending on those services. See, e.g.. Ronald Andersen & Lee Benham, Factors Affecting the Relationship Between Family Income and Medical Care Consumption, in H.E. Klarman, Empirical Studies in Health Economics (1970) (finding income elasticity greater than unity for dental care);
259. Fred Goldman & Michael Grossman, The Demand for Pediatric Care: An Hedonic Approach, 86 J. Pol. Econ. 259 (1978);
260. Marian E. Gornick, Disparities in Medicare Services: Potential Causes, Plausible Explanations, and Recommendations, Health Care Fin. Rev. Summer 2000, at 23 (citing data on physician visits and medical procedures);
261. Carol L. Jenkins, Resource Effects on Access to Long-Term Care for Frail Older People, 13 J. Aging & Soc. Pol'y 35 (2001) (presenting data on nursing home care).
262. In our view, the income elasticity of demand for health care makes much of the recent concern over disparities in the care received by different racial, ethnic, and income groups, see infra note 111, unrealistic, reflecting a too-casual assumption that consumption of health care should be equal for all groups and classes. Such equality, in addition to being unrealistic as a practical matter - which is not to say that it should not be actively promoted in specific contexts - would be achievable at reasonable cost only by leveling down, thus radically denying people the freedom to spend more on health care than others choose or are able to spend. Cf. Chaoulli v. Quebec, 2005 S.C.C. 35, 29272, [2005] S.C.J. No. 33 QUICKLAW (June 9, 2005) (Canadian Supreme Court's invalidation of Quebec law prohibiting purchase of private health insurance for services covered by Canada's national health program, on the ground that adverse health consequences resulting from national program's waiting lists made prohibition unconstitutional). Moreover, while equality has great symbolic value for many, maintaining it as a goal hampers efforts to ameliorate the specific inequities we identify in this article.
263. In general, society treats health care as a merit good, not to be rationed solely by ability and willingness to pay. At the margin, however, spending on health care provides only limited value for money spent - perhaps only hope or a slightly reduced probability or imminence of a bad outcome - making it (presumably) socially acceptable for persons with higher incomes to receive services that others cannot afford. Such persons are in a position to put a higher valuation on their well-being, ascribing higher costs to sick days and higher value to healthy ones, and to substitute away from time-intensive investments in health, such as bed rest, in favor of paying for medical interventions.
264. This tendency was dramatized for one of the authors when, in the 1970s, his employer, in a seemingly progressive move, combined two similar health plans it maintained for hourly-paid and higher-paid salaried workers, respectively, with the result that the former's premium contributions rose while the latter's declined.
265. See generally James P. Smith, Healthy Bodies and Thick Wallets: The Dual Relation Between Health and Economic Status, 13 J. Econ. Persp. 145 (1999) (illustrating that socioeconomic status tends to positively correlate with health status).
266. Even if it should appear that a health plan's expenditures were the same per capita for both high- and lower-income members, it could be argued that the poorer health status of the latter group should entitle them to additional, not just equal, spending; in any event, a finding of such equality would belie the usual assumption that, in employee health plans, the healthy wealthy substantially subsidize the low-wage sick. Although we suspect that it may be the other way around, the picture is more complicated to the extent the health status of lower-income workers is worse. For a fuller exposition of the possibility that the predominant redistribution is as generally assumed, see Hall, supra note 50, at 165-68. See also infra note 116.
267. The classic study was the Rand Health Insurance Experiment. See Joseph Newhouse et al., Free for All? Lessons From the Rand Health Insurance Experiment (1993).
268. See also Julie Hudman & Molly O'Malley, Kaiser Commission on Medicaid and the Uninsured, Health Insurance Premiums and Cost Sharing: Findings from the Research of Low Income Populations (2003);
269. Mitchell D. Wong et al., Effects of Cost Sharing on Care Seeking and Health Status: Results from the Medical Outcomes Study, 91 Am. J. Pub. Health 1889 (2001);
270. Avi Dor & William Encinosa, Does Cost Sharing Affect Compliance? The Case of Prescription Drugs (Nat'l Bureau of Econ. Research, Working Paper No. 10738, 2004).
271. The apprehended unfairness would not occur if, as we believe is only rarely the case, the plan charged patients who choose the more expensive option its full incremental cost. Cf. infra notes 119 & 120. It is also notable - although our point here is somewhat different - that using more or costlier services, out-of-network providers, or non-formulary drugs may frequently produce better health outcomes. Thus, those who are most discouraged by additional charges may get poorer service while still seemingly subsidizing higher-quality care for those whose consumption choices are less affected. See, e.g., Dor & Encinosa, supra note 105 (showing that co-insurance and, to a lesser extent, fixed co-payments unduly discourage patients from taking efficacious preventive medicine).
272. This would occur if, as seems likely, lower-income individuals are more reluctant than higher-income participants in the same HDHPs to spend their HSA funds to satisfy deductibles and other cost-sharing prerequisites for tapping insurance funds. Indeed, the raison detre of the CDHC reform strategy is to encourage use of heavier cost sharing to counteract moral hazard, displacing the arguably more even-handed rationing methods employed by managed-care organizations. See Hall & Havighurst, supra note 97, at 1492 ("[T]he strategy of causing consumers to set aside assets for spending on their own health care should inspire at least some economizing behavior of the sort that has been systematically missing with comprehensive first-dollar coverage."). The empirical issue, not yet studied (as far as we are aware), is whether and how much HSAs will affect the income-elasticity of demand for health services. See supra note 101
273. Intuition suggests that such effects will occur under any insurance plan providing ostensibly equal benefits for both high- and low-income enrollees. Although we have found no studies attempting to detect actual regressive effects in individual employment groups, data from the RAND Health Insurance Experiment clearly showed that, in a controlled setting, cost sharing had noticeably greater effects on middle-income consumers than on higher-income ones. Newhouse, supra note 105, at 46. This is just the effect we believe causes systematically regressive effects in employee health plans.
274. M. Gregg Bloche, Race and Discretion in American Medicine, 1 Yale J. Health Pol'y L. & Ethics 95, 108 (2001).
275. A suggestive example of how some patients can "work the system" appears in a notable recent case: Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355 (2002) (finding no obstacle in federal law to enforcing Illinois statute requiring so-called external review of HMO denials of coverage). The plaintiff, having been clever enough to find an out-of-state surgeon offering an especially aggressive treatment for her neurological condition, was able to persuade her HMO doctor (though he disclaimed any expertise) to opine that the surgery was medically necessary - contrary to several other medical opinions supporting the
276. Researchers are currently paying a great deal of attention to disparities that correlate with patients' racial and ethnic characteristics. See, e.g., Institute of Medicine, Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare (2003);
277. Symposium, Racial & Ethnic Disparities, 24 Health Aff. 316 (2005).
278. Some of these disparities may be attributable in whole or in part to considerations of the kind mentioned in the text. Although most studies of the matter have focused only on inequality as such, several can be cited in support of our hypothesis that physicians resonate to income-correlated patient expectations. E.g., Michelle van Ryn & Jane Burke, The Effect of Patient Race and Socio-Economic Status on Physicians' Perceptions of Patients, 50 Soc. Sci. & Med. 813 (2000) (examining physicians' attitudes toward patients and revealing that lower-SES patients are viewed as less independent, less rational, less responsible, and less intelligent than their wealthier counterparts);
279. S. Willems et al., Socio-Economic Status of the Patient and Doctor-Patient Communication: Does It Make a Difference?, 56 Patient Educ. & Counseling 139 (2005) (reviewing literature finding that patients from lower social classes receive less-effective communication and exert less control over their consultations with their physician). Evidence from countries in which health insurance status is equal for all patients also suggests that low-SES patients tend to receive fewer services at health-plan expense.
280. See infra note 121 & 122.
281. See supra note 4.
282. See Jonathan Gruber & Alan B. Krueger, The Incidence of Mandated Employer-Provided Insurance: Lessons from Workers' Compensation Insurance 1 (Princeton Univ. Indus. Relations Section, Working Paper No. 279, 1990) (finding that areas and industries with high workers-compensation costs led to a corresponding reduction in wages);
283. Jonathan Gruber, The Incidence of Mandated Maternity Benefits, 84 Am. Econ. Rev. 622 (1994) (finding that the mandated expansion of health insurance coverage in the Pregnancy Discrimination Act of 1978 led to reduced wages);
284. Louise Sheiner, Health Care Costs, Wages, and Aging (Apr. 1999) (unpublished manuscript, on file with Fed. Reserve Bd. of Governors), available at http://www.federalreserve.gov/pubs/feds/1999/199919/ 199919pap.pdf (finding relatively lower wages for older workers and workers with family coverage in areas with high medical prices).
285. Under this hypothesis, the employer referred to in supra note 103 would be expected to adjust wages (upward) and salaries (downward) in subsequent years to eliminate the seeming inequity. But proving that such adjustments actually occurred would be difficult if, rather than being made explicitly, they took the form of unconscious responses to labor market conditions.
286. One apparent exception - that nevertheless may only help to prove the general rule - is the lower wages paid by employers to obese workers, which have been convincingly attributed in part to such employees' greater demands on the employer's health insurance. Jay Bhattacharya & M. Kate Bundorf, The Incidence of the Healthcare Costs of Obesity (Nat'l Bureau of Econ. Research, Working Paper No. 11303, 2005). As to employees in general, however, unless an employer is self-insured, it is unlikely to have good information on how heavily various employee subgroups use their health benefits; indeed, because the researchers in the cited study did not distinguish between self-insured and other employers or show that employers knew what wage adjustments to make, they could not rule out the possibility that the reduced wages of obese workers reflected only assumptions or general prejudice. In any event, it does not follow from the treatment of obese employees that the salaries of high-utilizing, high-income employees would be similarly reduced. On the other hand, if the market does adjust wages downward for high utilizers, it should also, presumably, adjust them favorably for employees using fewer services. But few would expect to find that employees on the wrong side of widely noted racial disparities are receiving compensatorily higher wages.
287. This suggestion that insurance pools should be subdivided flies in the face of the conventional view that large, heterogeneous pools are socially desirable because they seem to cause healthier, wealthier insureds to subsidize the care of less healthy, lower-paid enrollees. See supra note 104. Although not finally proven, the hypothesis we offer in the text suggests not only that such progressive redistribution is not necessarily the rule but also that most employment groups provide, not social insurance, but what might be called "anti-social" insurance, with subsidies actually running in regressive directions. It is probable, to be sure, that large insurance groups organized in employment settings do benefit incidentally those whose recurrent health problems and serious chronic conditions might make them uninsurable, or insurable only at prohibitive cost, in an insurance market in which individuals seek to pool their health risks only with others facing similar risks. In any event, analyzing these matters is obviously complicated by the possibility, discussed in the text, that the labor market ultimately causes wages and salaries to reflect the actual burdens that various income classes impose on the employer's health plan.
288. In theory, optimal cost sharing is a function of several factors, including the insured's ability to bear financial risk and the slope of the group's demand curve for the service in question (which reveals whether it is viewed as a discretionary service or a vital one in most cases). Indeed, but for the high cost of administering finely tuned coverage, coinsurance rates would be set separately for each discrete service rather than, as is the general practice, applied across the board to all services (a notable exception being the generally higher coinsurance rates charged for outpatient psychiatric services, imposed in the belief that such services are especially discretionary). The problem might be addressed in part, of course, by offering different plans with different cost sharing so that lower-income persons could purchase coverage more suitable for their situations.
289. See Shapiro, supra note 71.
290. On the case of employers dealing with labor unions, see supra note 100. The usual explanations given for large risk pools are that they can command lower premiums for actuarial reasons and can drive harder bargains with providers. Although such considerations introduce tradeoffs that might tilt the balance in some cases, the alternative explanation we offer suggests that some such benefits, even if real, may come at the expense of distributional fairness.
291. Kaiser Family Foundation & Health Research and Educational Trust, Employer Health Benefits: 2004 Annual Survey 63, exh. 4.7 (2004) (reporting that only nineteen percent of employers offering employees a choice of plan - and the same percentage of the largest employers as well - make the same dollar contribution regardless of plan chosen);
292. see Enthoven, supra note 90. Regressivity would not be a significant problem under Enthoven's prescription - assuming that incremental costs are calculated actuarially, with appropriate awareness of different groups' propensity to use their coverage.
293. Indeed, we speculate that employers pool all their nonunionized employees for purposes of health coverage in part because the unwitting contributions of lower-income workers make it cheaper for them to provide the benefits that high-income employees particularly desire. See supra text accompanying notes 88 & 89. Under the hypothesis that employers prefer arrangements under which the rank-and-file subsidize the coverage of more affluent employees, one would expect most employers to resist - as they have - Enthoven's proposal to equalize contributions for each employee. Also, employers could be expected to be slow - as they have been - to adopt the CDHC approach, under which HSAs must be funded equally for all employees.
294. Gary Claxton et al., What High-Deductible Plans Look Like: Findings from a National Survey of Employers, 2005 Health Aff. (Web Exclusives) W5-434 (finding less than four percent of employers offered HSA/HDHP coverage in 2005).
295. David A. Alter et al., Socioeconomic Status, Service Patterns, and Perceptions of Care Among Survivors of Acute Myocardial Infarction in Canada, 291 J.A.M.A. 1100, 1103-04 (2004);
296. see also Norman Frohlich et al., Health Service Use in the Winnipeg Regional Health Authority: Variations Across Areas in Relation to Health and Socioeconomic Status, Health Mgt. Forum, Supplement (Winter 2002) 9-14 (presenting additional Canadian data).
297. M. Sutton et al., Allocation of Resources to English Areas: Individual and Small Area Determinants of Morbidity and Use of Health Care Resources, Report to the Department of Health (Edinburgh: Information and Services Division, 2002);
298. see also Max Exworthy et al., Evidence into Policy and Practice? Measuring the Progress of U.S. and U.K. Policies to Tackle Disparities and Inequalities in U.S. and U.K. Health and Health Care, 84 Milbank Q. 75, 79 tbl.1 (2006).
299. For data from Australia, see Anthony Scott et al., Is General Practitioner Decision Making Associated with Patient Socio-economic Status?, 42 Soc. Sci. Med. 35 (1996) (finding patients of high socioeconomic status more likely to be tested for illnesses and less likely to receive a prescription, suggesting the visit was unnecessary).
300. See Karen Davis & Roger Reynolds, Medicare and the Utilization of Health Care Services by the Elderly, 10 J. Hum. Resources 36 (1975) (finding significant income effects on the number of physician visits by Medicare beneficiaries with a similar health status);
301. Marian E. Gornick et al., Effects of Race and Income on Mortality and Use of Services Among Medicare Beneficiaries, 335 New Eng. J. Med. 791 (1996) (showing a positive correlation between Medicare beneficiaries' income and their utilization of certain physician services, including screenings);
302. Charles R. Link et al., Equity and the Utilization of Health Care Services by the Medicare Elderly, 17 J. Hum. Resources 195 (1982) (concluding that income affects the number of physician visits by those on Medicare when controlling for health status);
303. Gornick, supra note 101.
304. On the regressivity of Medicare's financing - that is, the question of who pays as opposed to who benefits (the subject of discussion here) - see supra notes 1 & 48.
305. On the larger question of Medicare's net distributional effects, see supra note 1.
306. Their more recent data suggests that lower-income households began to account for higher levels of spending in the mid-1990s. See McClellan & Skinner, supra note 1;
307. Julie Lee, Mark McClellan & Jonathan Skinner, The Distributional Effects of Medicare Expenditures (Nat'l Bureau of Econ. Research, Working Paper No. 6910, 1999);
308. Jonathan Skinner & Weiping Zhou, The Measurement and Evolution of Health Inequality: Evidence from the U.S. Medicare Population (Nat'l Bureau of Econ. Research Working Paper No. 10842, 2004).
309. But see Bhattacharya & Lakdawalla, supra note 1. Query whether similar distributional effects may occur under the new Medicare prescription-drug benefit once it is finally implemented. See supra note 20. This benefit, enacted in 2003, includes a so-called "doughnut hole" - that is, a lack of coverage for any annual expenditures between $2250 and $3600 that applies only after an initial government outlay of seventy-five percent of all expenditures up to $2250 (after a $250 deductible) and that a beneficiary must cover himself before new catastrophic drug coverage kicks in. The up-front benefit, covering easily budgetable expenditures, was obviously adopted so that the (Republican) proponents could represent to voters that the bill, while generally confusing, would immediately benefit all beneficiaries, not just the minority with very high drug costs. But this representation hid the doughnut hole, which could present a problem for many lower-income beneficiaries, with the result that higher-income users of the system are likely to chain a disproportionate share of total benefits.
310. Lisa R. Shugarman et al., Differences in Medicare Expenditures During the Last 3 Years of Life, 19 J. Gen. Internal Med. 127 (2004). Note that focusing on consumption in years prior to death, rather than on consumption by age cohorts, may be a good way to ensure comparisons of consumption by individuals with a similar health status.
311. Mark McClellan & Jonathan Skinner, Medicare Reform: Who Pays and Who Benefits?, Health Aff., Jan.-Feb. 1999, at 48, 59.
312. The authors observe, however, that Medicare's overall unfairness was mitigated by legislation in 1994 removing the earlier ceiling on the amount of salaries and wages subject to the Medicare tax. Another study, adding some irony to consumption disparities, finds that much of the higher utilization is explained by the more impatient-based and specialist-oriented patterns of care that are typical of high-spending regions and that this additional consumption does not appear to improve quality of care, access to care, satisfaction with care, or health outcomes. Fisher et al., supra note 81.
313. See Rezaul K. Khandker & Lauren A. McCormack, Medicare Spending by Beneficiaries with Various Types of Supplemental Insurance, 56 Med. Care Res. Rev. 137 (1999) (finding that Medicare beneficiaries who enjoy Medigap or employer-sponsored supplemental coverage consume more Medicare dollars than beneficiaries covered by Medicare alone;
314. Renee Mentnech et al., An Analysis of Utilization and Access from NHIS: 1984-92, Health Care Fin. Rev., Winter 1995, at 51 (similar finding with respect to a Medicare beneficiary's likelihood of visiting a physician);
315. Nadereh Pourat et al., Socioeconomic Differences in Medicare Supplemental Coverage, Health Aff. Sept.-Oct. 2000, at 186 (finding that lower-income individuals are less likely to enjoy supplemental coverage).
316. On the macroeconomic consequences, indirectly adverse to Americans working outside the health sector, of limiting consumers' opportunities to economize on health care, see supra note 99.
317. Both federal and state laws require hospitals with emergency departments to stabilize emergent conditions without regard to commercial considerations. See, e.g., supra note 35 and accompanying text; Thompson v. Sun City Cmty. Hosp., 688 P.2d 605, 610-11 (Ariz. 1984) (enforcing the statutory duty to treat an indigent's emergency at county expense).
318. Although these laws stop short of imposing a duty to provide free extended care or other services, federal, state, and local governments also provide subventions, public hospitals, and public clinics as additional safety nets for those with serious health needs and no health insurance. See supra note 38. Also, in the same spirit as laws requiring emergency care, common-law courts do not permit a health care provider having an established relationship with a patient to terminate it at will if doing so would place the patient in peril. E.g., Surgical Consultants, P.C. v. Ball, 447 N.W.2d 676, 682 (Iowa Ct. App. 1989) (holding that although her doctor's office had told the plaintiff she was no longer a patient because of an unpaid bill, a valid claim for "abandonment" could be proved by "evidence that the physician has terminated the relationship at a critical stage of the patient's treatment, that the termination was done without reason or sufficient notice to enable the patient to procure another physician, and that the patient [was] injured as a result thereof.").
319. In addition, insurance law provides special protections against such dangerous insurer practices as bad-faith denials of claims for benefits. See generally William M. Shernoff et al., Insurance Bad Faith Litigation (1984 & Supp. 2006);
320. Alan O. Sykes, "Bad Faith" Breach of Contract by First-Party Insurers, 25 J. Legal Stud. 405 (1996).
321. See supra note 102.
322. The Clinton administration's proposed Health Security Act of 1994 was first presented to the public with the claim that it "empowers consumers to make more cost-conscious choices by choosing among health plans on the basis of price and quality. Consumers reap the savings from enrolling in a health plan that delivers the guaranteed benefits for a lower premium." Press Release, The White House, Health Security Preliminary Plan Summary (Sept. 22, 1993), available at http://www.clintonfoundation.org/legacy/ 092293-press-release-on-health-care-security-plan.htm. But although this and other rhetoric implied that competition and consumer choice would primarily drive the projected system, the plan's details, by specifying seemingly generous "guaranteed benefits," greatly limited consumers' options and the role of private contracts in particularizing rights and obligations.
323. In fact, the proposal's egalitarianism quotient (and thus its probable cost) was quite high. See Havighurst, Health Care Choices, supra note 24, at 29-89, for a thorough review of the Clinton proposal in this light.
324. As a thought experiment, it is instructive to ask why Americans cannot buy Canadian-style or U.K.-style health care, both substantially cheaper than the U.S. variety though in some respects (waiting time, for example) probably less good. Cf. Chaoulli v. Quebec, 2005 S.C.C. 35, 29272, [2005] S.C.J. No. 33 QUICKLAW (June 9, 2005). It is our sense that such innovative offerings, even if not actually barred by law, would face enough legal uncertainty to make them untenable despite their potential attractiveness to many consumers. No single firm could afford to defend its innovation against the inevitable legal challenges if its competitors could follow at no cost any trails it succeeded in blazing. It would be hard to argue, though, that what is good enough to satisfy a political majority of Canadians or Britons is not good enough for many Americans.
325. As an interest group becomes larger and more diverse, the practical problems of organizing for political or other collective action become greater. See generally Mancur Olson, The Logic of Collective Action (rev. ed. 1971);
326. Mark A. Peterson, From Trust to Political Power: Interest Groups, Public Choice, and Health Care, 26 J. Health Pol. Pol'y & L. 1145, 1154-57 (2001) (describing the declining political effectiveness of the American Medical Association and the rise of competing interest groups in the health care field).
327. perennial problem in policy analysis is the quickness with which some analysts endorse a larger role for government than for markets on the ground that consumers, on whose choices markets depend, are ignorant, powerless, and easily manipulated. See, e.g., Russell Korobkin, The Efficiency of Managed Care "Patient Protection" Laws: Incomplete Contracts, Bounded Rationality, and Market Failure, 85 Cornell L. Rev. 1 (1999). Yet those same consumers, precisely because they have so little hope of protecting themselves successfully in the political process, are apt to be even more ignorant, individually powerless, and manipulable in their capacity as voters - to whom democratic government is supposed to be accountable. Moreover, some consumer-voters are better equipped to participate in and influence the political process than others, compounding the potential for harm to the most ignorant and least politically effective. On the other hand, an attractively democratic feature of free markets is the absence of majority dictation to the minority and the incentives they therefore create for participants to cater specifically to minority tastes and needs. Whatever need there may be for law and regulation to protect against real abuses, it too often serves as an excuse for disenfranchising consumers and for empowering government and those whom it serves best.
328. See, e.g., Symposium, Getting Beyond Cynicism: New Theories of the Regulatory State, 87 Cornell L. Rev. 267 (2002).
329. For an example of special-interest influence over seemingly benign legislation in the health care field, see Sallyanne Payton & Rhoda Powsner, Regulation Through the Looking Glass: Hospitals, Blue Cross, and Certificate-of-Need, 79 Mich. L. Rev. 203 (1980), which shows how hospital and insurance interests supported early certificate-of-need laws that were promoted as cost-containment measures but in fact served to curb competition and to legitimize, rather than prevent, cost increases. Although our overview cannot do full justice to all the specific regulatory programs and legal rules touched upon, our survey adds weight to our claim that U.S. health care is systematically arranged and governed to take unfair advantage of the lower- and middle-income majority. We have already suggested in Part II the breathtaking magnitude of the injustice that concerns us.
330. See supra text accompanying notes 84-92.
331. Many of the most questionable regulatory policies, moreover, are maintained well out of public view and are even more rarely seen as controversial. Although federal regulators are now required to provide benefit-cost comparisons when adopting new regulations, similar requirements rarely govern health-sector regulation at the state level. See generally Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. Chi. L. Rev. 1 (1995) (discussing use of regulatory impact analyses by federal agencies).
332. For a discussion and comparison of regulatory impact statements by state agencies, see Robert W. Hahn, State and Federal Regulatory Reform: A Comparative Analysis, 29 J. Legal Stud. 873 (2000).
333. In a classic 1963 article on medical economics, economist Kenneth Arrow speculated that consumers' fears about purchasing in ignorance alone were a sufficient explanation for health-sector regulation: The general uncertainty about the prospects of medical treatment is socially handled by rigid entry [i.e., physician licensing] requirements. These are designed to reduce the uncertainty in the mind of the consumer as to the quality of product insofar as this is possible. I think this explanation, which is perhaps the naive one, is much more tenable than any idea of a [medical] monopoly seeking to increase incomes. No doubt restriction on entry is desirable from the point of view of the existing physicians, but the public pressure needed to achieve the restriction must come from deeper causes. Kenneth J. Arrow, Uncertainty and the Welfare Economics of Medical Care, 53 Am. Econ. Rev. 941, 966 (1963). Writing before Medicare and Medicaid and at a time when the public was both less insulated from and less concerned about the cost of health services, Arrow had no reason to be concerned about the possibility of overregulation.
334. Even here, state governments, which are responsible for some of the most restrictive regulatory controls, see none of the costs of Medicare and only a fraction of their own Medicaid costs, which are paid in substantial measure with matching federal funds. See James F. Blumstein & Frank A. Sloan, Health Care Reform Through Medicaid Managed Care: Tennessee (TennCare) as a Case Study and a Paradigm, 53 Vand. L. Rev. 125, 136-49 (2000).
335. Although proposals to "cap" the tax exclusion have been offered from time to time, they have always fallen on deaf political ears. See Havighurst, Health Care Choices, supra note 24, at 103 ("[C]apping the tax subsidy is a notion that only a policy wonk could love, a meritorious policy idea with no natural political constituency.").
336. See generally Marie Gottschalk, The Shadow Welfare State: Labor, Business, and The Politics of Health Care in the United States 42-44 (2000) (observing how, after World War II and the introduction of the tax subsidy, the labor movement divided its efforts between bargaining for health benefits and advocacy of national health insurance, but failing to highlight the tax subsidy as a key explanation for the movement's priorities).
337. For the view that "[c]onsumers have sought the kind of health insurance they have... precisely because they don't wish to be forced to make rational trade-offs when they are confronted with medical care consumption decisions," see Bruce C. Vladeck, The Market v. Regulation: The Case for Regulation, 59 Milbank Memorial Fund Q. 209, 211 (1981). Although it is true that people buy coverage to avoid difficult choices, Vladeck's statement begs the question whether people have in fact been offered either market or political choices requiring them to face the high marginal cost of the extra levels of protection he says they demand.
338. For recent survey data demonstrating the increasing salience of costs in consumer attitudes toward health care, see Employee Benefit Research Institute, 2006 Health Confidence Survey: Dissatisfaction with Health Care Sytem Doubles Since 1998 (2006), available at http://www.ebri.org/ publications/notes/index.cfm?fa=notesPrint&content_id=3758.
339. For fuller elaboration of the "shell game" metaphor, see Havighurst, How the Revolution Fell Short, supra note 24, at 78-80.
340. Jonathan Gruber & Robin McKnight, Why Did Employee Health Insurance Contributions Rise?, 22 J. Health Econ. 1085, 1085 (2003) (observing that between 1982 and 1998 the percentage of employees whose employers paid the full cost of their coverage fell from forty-four percent to twenty-eight percent).
341. I.R.C. § 125 (LEXIS 2005) (providing for so-called cafeteria plans).
342. Helen Levy, Who Pays for Health Insurance? Employee Contributions to Health Insurance Premiums 9-10 (Princeton Univ. Indus. Relations Section, Working Paper No. 398, 1998);
343. see also Gruber & McKnight, supra note 145, at 1088-89.
344. Although the latter source suggests that this strategy may cause employees simply to seek coverage available under a spouse's plan or to rely on Medicaid to provide for their dependents, the decline in the rate of employee "take-up" of employer-offered coverage has also contributed to the growth of the uninsured population. See Michael A. Morrisey, Price Sensitivity in Health Care: Implications for Health Care Policy 36-37 (2d ed. 2005) (summarizing evidence that decline in take-up rates, rather than in the number of employers offering coverage, is principally responsible for the increased number of uninsured).
345. See supra notes 96 & 108 on distributional implications of the CDHC reforms.
346. Some purport to fear that high-deductible coverage will cause healthier, often wealthier employees to contribute less than they currently do to the support of coworkers with more or chronic health problems. See generally Beth Fuchs & Julia A. James, Nat'l Health Pol'y Forum, Health Savings Accounts: The Fundamentals (2005), available at http://www.nhpf.org/pdfs_bp/BP_HSAs_04-11-05.pdf (reviewing arguments for and against HSAs). Yet, as noted previously in Part III, our conjecture is that, when people with disparate risks and preferences are arbitrarily pooled in the same health plan, subsidies may actually run the other way, with higher-income insureds, as a class, getting more out of the plan than their lower-income coworkers. See supra note 107 and accompanying text.
347. See Hall & Havighurst, supra note 97 (arguing that HSAs and other innovations could make the public more comfortable with health plans' predetermination of benefits and other methods of rationing coverage).
348. For an overview of disruptive innovation and the view that competition to create radically different, often lower-cost, alternative products has been crucial in destabilizing and ultimately improving the performance of many industries, see Clayton Christensen, The Innovator's Dilemma (1997).
349. In our view, many potential innovations stymied by overregulation in the health care sector would be helpfully destructive of conventional paradigms and beneficial from the standpoint of efficiency and consumer welfare. For an application of disruptive innovation theorizing to the health care sector and to the thesis of this article, see Lesley H. Curtis & Kevin A. Schulman, Overregulation of Health Care: Mustings on Disruptive Innovation Theory, 69 Law & Contemp. Probs. 195, 197-206 (Autumn 2006).
350. See generally Fed. Trade Comm'n & U.S. Dep't of Justice, supra note 25, ch.2, at 25-28;
351. Morris M. Kleiner, Occupational Licensing, 14 J. Econ. Persp. 189 (2000) (overview of economic literature on licensure, with emphasis on the need for empirical studies of cost and other impacts);
352. Morris M. Kleiner & Robert T. Kudrle, Does Regulation Affect Economic Outcomes?: The Case of Dentistry, 43 J.L. & Econ. 547 (2000) (estimating that state licensure causes thirty percent increase in dentists' fees).
353. See supra note 139.
354. An alternative to exclusionary licensure is public certification, sometimes called title licensure. This intermediate form of regulation, which addresses the information problem by limiting not who may practice in a field but who may use certain titles, was famously advocated by economist Milton Friedman. Milton Friedman, Capitalism and Freedom 144-49 (2d ed. 1982).
355. In economic theory, regulation is not deemed inefficient as long as the gains to the winners exceed the losses to the losers, whether or not the losers are compensated in some way. See Richard A. Posner, Economic Analysis of Law 13-15 (5th ed. 1998) (discussing this so-called Kaldor-Hicks definition of efficiency).
356. Even if regulation passes this test, however, social justice is obviously implicated if lower-income persons are regularly on the losing end. See Uwe E. Reinhardt, Can Efficiency Be Left to the Market?, 26 J. Health Pol. Pol'y & L. 967, 984 (2001) (admonishing that "Kaldor's expedient normative dictum is never an excuse for setting aside moral thought"). Moreover, even in the event that lower-income wage earners are not net losers once all social programs (which serve mostly the poor) are taken into account, there is no good reason why the health care system should be organized and regulated to their disadvantage.
357. See, e.g., Sidney L. Carroll & Robert J. Gaston, Occupational Restrictions and the Quality of Service Received, 47 So. Econ. J. 959 (1981) (stressing that social costs of regulation, which must be compared to benefits, include harms resulting from consumers' forgoing of valuable services because of regulation-induced higher prices).
358. For further discussion and a graphic bell curve illustrating how majority rule is likely to yield overregulation, see Havighurst, How the Revolution Fell Short, supra note 24, at 82-86.
359. See id (illustrating how majority rule under conditions of systematic consumer-voter ignorance about the incidence of health care costs might produce "hyper-regulation").
360. See Clark C. Havighurst, Contesting Anticompetitive Actions Taken in the Name of the State: State Action Immunity and Health Care Markets, 31 J. Health Pol. Pol'y & Law 585 (2006) (discussing the limited extent to which federal antitrust law limits state authority to immunize anticompetitive actions by state licensing and regulatory boards).
361. See generally Symposium, Private Accreditation in the Regulatory State, 57 Law & Contemp. Probs. (Autumn 1994).
362. See generally Clark C. Havighurst et al., Health Care Law and Policy 709-52 (2d ed. 1998).
363. See Clark C. Havighurst, Deregulating the Health Care Industry: Planning for Competition (1982);
364. see also supra note 136.
365. See, e.g., Sujit Choudhry et al., Specialty versus Community Hospitals: What Role for the Law?, 2005 Health Aff. (Web Exclusives) W5-361 (proposing certificate-of-need regulation to protect community hospitals against opportunistic competition from specialized providers);
366. Clark C. Havighurst, Monopoly Is Not the Answer, 2005 Health Aff. (Web Exclusives) W5-373 (criticizing Choudhry et al. proposal).
367. See Employee Retirement Income Security Act of 1974, 29 U.S.C. §§ 1001-03 (LEXIS 2005) (preempting state laws relating to employee health benefits);
368. Aetna Health Inc. v. Davila, 542 U.S. 200 (2004) (finding certain health plans immunized by ERISA from state-imposed liability for wrongful denials of health benefits).
369. See generally Gail A. Jensen, Regulating the Content of Health Plans, in American Health Policy: Critical Issues for Reform 167 (Robert B. Helms ed., 1993).
370. Ironically, only small employers and their workers are affected by such cost-increasing benefit mandates because employers large enough to self-insure their employees' health benefits are exempted from such state requirements by federal legislation. See supra note 163 and accompanying text. A recent federal legislative proposal to exempt health plans offered to pools of small businesses from various state coverage mandates, S. 1955, 109th Cong. (2006), was opposed by a broad-based coalition of health care providers, patient-advocacy groups, insurance regulators, and state attorneys general.
371. See From the CQ Newsroom: Industry Group Backs Health Association Bill as States Push for Their Rights, http://www.cmwf.org/healthpolicyweek/ healthpolicyweek_show.htm?doc_id=370240#doc370242 (last visited Nov. 10, 2006).
372. See generally Mark A. Hall, The Death of Managed Care: A Regulatory Autopsy, 30 J. Health Pol. Pol'y & L. 427 (2005) (concluding that health plans scaled back earlier efforts to manage health care costs less because of new regulation than for other reasons);
373. Havighurst, How the Revolution Fell Short, supra note 24 (extensive
374. See Frank A. Sloan & Mark A. Hall, Market Failures and the Evolution of State Regulation of Managed Care, 65 Law & Contemp. Probs. 169 (Fall 2002) (reviewing so-called patient-protection legislation).
375. See Havighurst, How the Revolution Fell Short, supra note 24, at 86 (extending bell-curve illustration,
376. see supra notes 156-57, to the account for the "hyper-regulation" of managed care).
377. Cynthia Smith et al., National Health Spending in 2004. Recent Slowdown Led by Prescription Drug Spending, 25 Health Aff. 186, 187 exh. 1 (2006) (showing health care expenditures as stable around 13.8% of GDP from 1993-2000, when managed care was dominant, but rising to 16% of GDP in 2004).
378. Cf. Bipartisan Patient Protection Act, S. 1052, H.R. 2563, § 104, 107th Cong. (2001) (proposal to mandate external professional review of any coverage decision based on "medical necessity or appropriateness" would have freed external reviewers from honoring contractual limits unless the applicable exclusion is categorically or numerically "exact");
379. see also supra note 110.
380. See Havighurst, Health Care Choices, supra note 24, at 222-64.
381. See generally id.
382. Economists and other policy analysts rarely recognize the unavailability in the market of economizing choices and the legal risks that inhibit the offering of such choices as the serious policy problems they are. See supra note 65.
383. But see Pauly, supra note 24, at 1528 (recognizing that, for legal and other reasons, "[p]eople cannot generally choose knowledgeably among a variety of plans characterized by explicitly different policies toward new technology.").
384. Although Pauly's discussion focuses principally on encouraging private rationing of "new" technology, it concludes, wisely, that "some type of legal safe harbor has to be created for insurers that implement well-designed plans for limiting technology [in general], and [that] the 'community or standard practice' and 'medical necessity' concepts need to be jettisoned." Id. at 1534.
385. See Michael F. Cannon & Michael D. Tanner, Healthy Competition: What's Holding Back Health Care and How to Free It 69, 116 (2005) (advocating still greater tax incentives for HSAs and suggesting that federal law widen consumer choice by enabling consumers to purchase health coverage from insurers regulated by states other than their own).
386. Although difficult to define, detect, and distinguish from other consequences of moral hazard, defensive medicine is widely (and plausibly) believed to raise the overall cost of American medical care. See Daniel Kessler & Mark McClellan, Do Doctors Practice Defensive Medicine?, 111 Q.J. Econ. 353 (1996) (discussing defensive medicine, comparing expenditures and outcomes in treating elderly heart patients in states with and without recent malpractice reforms, and estimating that greater intensity of care associated with higher malpractice risks, but not with better outcomes, raised costs five percent to nine percent).
387. But see Michelle M. Mello & Troyen A. Brennan, Deterrence of Medical Errors: Theory and Evidence for Malpractice Reform, 80 Tex. L. Rev. 1595, 1607 (2002) (noting paucity of clear evidence that defensive practice is a problem).
388. Indeed, the standard of care applied in any given lawsuit is rarely apparent even after the case is decided, but is instead merely implicit in a general jury verdict based on conflicting expert testimony about what the standard should be. Although a movement to create so-called clinical practice guidelines began in the 1980s and has evolved into widespread insistence that physicians practice so-called evidence-based medicine, practice guidelines, however well grounded they may be in evidence of efficacy and appropriateness (cost is still largely left out of the calculus), do not generally have official status but are only evidence of prevailing practice standards that juries may consider without being bound by them. See generally Troyen A. Brennan, Practice Guidelines and Malpractice Litigation: Collision or Cohesion?, 16 J. Health Pol. Pol'y & L. 67 (1991);
389. Mark Hall, The Defensive Effect of Medical Practice Policies in Malpractice Litigation, 54 Law & Contemp. Probs. 119 (Spring 1991);
390. Clark C. Havighurst, Practice Guidelines as Legal Standards Governing Physician Liability, 54 Law & Contemp. Probs. 87 (Spring 1991);
391. Andrew L. Hyams et al., Practice Guidelines and Malpractice Litigation: A Two-Way Street, 122 Annals of Int. Med. 450 (1995) (reporting survey of guideline use in actual cases);
392. Michelle M. Mello, Of Swords and Shields: The Role of Clinical Practice Guidelines in Medical Malpractice Litigation, 149 U. Pa. L. Rev. 645 (2001);
393. Arnold J. Rosoff, Evidence-based Medicine and the Law: The Courts Confront Clinical Practice Guidelines, 26 J. Health Pol. Pol'y & L. 327 (2001).
394. Although the licensure system might seem to provide protection for consumers against negligent physicians, discipline is only rarely administered because of simple incompetence. Gary L. Gaumer, Regulating Health Professionals: A Review of the Empirical Literature, 62 Milbank Memorial Fund Q. 380, 407 (1984) ("[R]esearch on credentialing shows that contemporary credentialing procedures may not be reliably screening actual practice competence."). On the other hand, private credentialing and other measures in well-run hospitals provide some assurance of quality. Significantly, however, neither these efforts nor the tort system itself - the high cost of which seems justifiable only if it effectively deters numerous patient injuries - has worked well enough in preventing or deterring patient injuries to preclude some strong, authoritative adverse criticism of the overall quality of U.S. health care.
395. Eg., Institute of Medicine, Crossing the Quality Chasm: A New Health Care System for the 21st Century (2001);
396. Institute of Medicine, To Err is Human: Building a Safer Health System (1999).
397. Recent industry efforts to respond to this embarrassing criticism have yet, it appears, to yield substantial improvements. See Lucian L. Leape & Donald M. Berwick, Five Years After To Err is Human: What Have We Learned?, 293 J.A.M.A. 2384, 2384 (2005) ("[L]ittle evidence exists from any source that systematic improvements in safety are widely available.").
398. See Havighurst, Health Care Choices, supra note 24, at 284-302 (suggesting various strategies for modifying overly demanding standards of care, including dispensing altogether with fault as the basis for liability).
399. For the suggestion that selected clinical practice guidelines could be incorporated in health care contracts to establish the standard of care, see supra text accompanying notes 171-72;
400. Havighurst, supra note 175.
401. See Havighurst, Health Care Choices, supra note 24, at 271-84 (suggesting various modifications of remedies for negligence and malpractice).
402. Virtually the only contractual modifications of traditional tort rights that courts have shown much willingness to enforce are arbitration clauses. E.g., Buraczynski v. Eyring, 919 S.W.2d 314 (Tenn. 1996) (enforcing agreements with two different patients undergoing knee reconstructions and finding that although the agreements were contracts of adhesion because presented by the physician on a take-it-or-leave-it basis, they were not unconscionable or oppressive);
403. Hawkins v. Superior Court, 152 Cal. Rptr. 491 (Cal. Ct. App. 1979) (holding arbitration agreement signed by husband as part of comprehensive contract with Kaiser plan was binding on his enrolled spouse).
404. But see Obstetrics & Gynecologists v. Pepper, 693 P.2d 1259 (Nev. 1985) (holding arbitration agreement offered by a clinic on take-it-or-leave-it basis was an adhesion contract, unenforceable in absence of evidence of plaintiffs knowing assent). This exception to the usual hostility to contractual reforms helps to prove the general rule, however, since an arbitration clause changes no more than the forum in which a claim is heard, not the substance of the claim nor the remedy available.
405. See, e.g., Madden v. Kaiser Foundation Hospitals, 552 P.2d 1178, 1186 (Cal. 1976) (enforcing arbitration clause in HMO contract because it "does not detract from Kaiser's duty to use reasonable care in treating patients, nor limit its liability for breach of this duty, but merely substitutes one forum for another"). Interestingly, arbitration clauses have been uncommon in health-plan contracts because, by lowering the cost of suing, they tend to increase the number of suits - hardly a result a health plan would desire.
406. Cf. Engalla v. Permanente Med. Group, Inc., 938 P.2d 903 (Cal. 1997) (finding Kaiser arbitration system cumbersome and costly, contrary to how it was represented to consumers, suggesting that Kaiser did not want arbitration to make its doctors any easier to sue than in the civil courts). On the other hand, HMOs might offer arbitration clauses more widely - for the mutual benefit of both the plan and its subscribers - if such clauses could be accompanied by other contract terms limiting substantive rights and obligations.
407. For scholarly discussions of the merits of letting consumers, with basic protections against fraud and overreaching, choose alternative liability regimes, see Symposium, Medical Malpractice: Can the Private Sector Find Relief?, 49 Law & Contemp. Probs. 143-320 (Spring 1986);
408. see also Richard A. Epstein, Medical Malpractice: The Case for Contract, 1976 Am. B. Found. Res. J. 87 (1976).
409. Of course, the ostensible justification for the legal system's skeptical attitude toward contracts limiting consumers' rights is the well-documented inability of consumers to make well-informed, rational choices. See supra note 134. It has never been clear, however, why consumers' difficulty in wisely choosing for themselves entitles privileged elites, with values, preferences, and economic interests of their own, to choose for them, thus forcing them to accept a legal regime in which they have virtually no voice at all. In our view, the often-impressive findings of psychology and behavioral economics should not be used only to discredit law-and-economics theorizing but should instead be seen as enriching it.
410. See, e.g., Christine Jolls, Cass Sunstein & Richard Thaler, A Behavioral Approach to Law and Economics, 50 Stan. L. Rev. 1471 (1998) (providing an overview of ways experimental psychology can inform thinking about legal rules). For example, findings about the limitations of consumers as decisionmakers can be seen to strengthen the case for letting consumers select and rely upon more disinterested agents and for reforming legal procedures to assist consumers in making consequential choices by simplifying and clarifying options, "framing" issues, responsibly shaping perceptions, forcing reconsiderations, and otherwise.
411. See generally Christine Jolls & Cass R. Sunstein, Debiasing through Law, 35 J. Leg. Stud. 199 (2006).
412. On the phenomenon and pervasiveness of the framing effect, see Norbert Schwarz, Self-Reports: How the Questions Shape the Answers, 54 Am. Psychologist 93 (1999).
413. Note the strong similarity between such a belief on the part of the legal profession in its own benignity and the medical paradigm under which physicians deem themselves to be, as ethical professionals, superior makers of spending decisions.
414. Many state courts have been so protective of their authority over the tort system that, using various somewhat strained constructions of the judiciary's powers under their state constitution, they have invalidated numerous efforts by state legislatures to reform malpractice law. E.g., Ferdon v. Wisconsin Patients Compensation Fund, 701 N.W.2d 440 (Wis. 2005) (finding a cap on non-economic damages unconstitutional and reviewing cases from other jurisdictions);
415. see also Havighurst et al., supra note 160, at 941-89 (2d ed. 1998) (reviewing case law on state malpractice reforms);
416. Carly N. Kelly & Michelle M. Mello, Are Medical Malpractice Damages Caps Constitutional? An Overview of State Litigation, 33 J.L. Med. & Ethics 515 (2005).
417. One might expect contractual reforms, being consensual, to fare better than legislative prescriptions, but they have not. A possible justification for maintaining the inalienability of tort rights is to preserve the integrity of the tort system as a public good, a vital deterrent to future malpractice. But, whereas one might be concerned that those opting out of the tort system are free-riding on its continuing beneficial effects on the quality of care, the evidence that the system serves that useful purpose is not strong. See supra note 176. Although the licensure system might seem to provide protection for consumers against negligent physicians, discipline is only rarely administered because of simple incompetence. Gary L. Gaumer, Regulating Health Professionals: A Review of the Empirical Literature, 62 Milbank Memorial Fund Q. 380, 407 (1984) ("[R]esearch on credentialing shows that contemporary credentialing procedures may not be reliably screening actual practice competence."). On the other hand, private credentialing and other measures in well-run hospitals provide some assurance of quality. Significantly, however, neither these efforts nor the tort system itself - the high cost of which seems justifiable only if it effectively deters numerous patient injuries - has worked well enough in preventing or deterring patient injuries to preclude some strong, authoritative adverse criticism of the overall quality of U.S. health care.
418. Eg., Institute of Medicine, Crossing the Quality Chasm: A New Health Care System for the 21st Century (2001);
419. Institute of Medicine, To Err is Human: Building a Safer Health System (1999).
420. Recent industry efforts to respond to this embarrassing criticism have yet, it appears, to yield substantial improvements. See Lucian L. Leape & Donald M. Berwick, Five Years After To Err is Human: What Have We Learned?, 293 J.A.M.A. 2384, 2384 (2005) ("[L]ittle evidence exists from any source that systematic improvements in safety are widely available.").
421. Paul C. Weiler et al., A Measure of Malpractice: Medical Injury, Malpractice Litigation, and Patient Compensation 17 (1993) ("[E]ven when one leaves aside the cost of securing and investing insurance funds and focuses simply on the process of claims administration and distribution, only about 40 percent of the total amount expended in the claims process actually reaches injured patients as compensation for their injuries.").
422. But see Mello & Brennan, supra note 174 (mildly defending current system while suggesting modest reforms).
423. For suggested ways of making the tort system pull more helpfully in the direction of quality assurance, see Jennifer Arlen & W. Bentley MacLeod, Malpractice Liability for Physicians and Managed Care Organizations, 78 N.Y.U. L. Rev. 1929, 1996-98 (2003) (arguing that "entity-level liability" for physician torts, rather than traditional liability rules, would come closer to creating optimal incentives for quality);
424. Clark C. Havighurst, Vicarious Liability: Relocating Responsibility for the Quality of Care, 26 Am. J.L. & Med. 7 (2000) (arguing that making organized health plans vicariously liable for the torts of their participating providers would both inspire integrated efforts to improve quality and restore needed legitimacy to managed health care by making plans responsible for the quality, as well as the cost, of care);
425. Randall Bovbjerg & Lawrence R. Tancredi, Liability Reform Should Make Patients Safer: "Avoidable Classes of Events" Are a Key Improvement, 33 J.L. Med. & Ethics 478 (2005) (expanding earlier proposals to impose automatic, i.e., strict, liability for certain, normally preventable adverse outcomes as a way both to ensure fair compensation at low administrative cost and to strengthen incentives to avoid such compensable events);
426. Clark C. Havighurst & Lawrence R. Tancredi, "Medical Adversity Insurance" - A No-Fault Approach to Medical Malpractice and Quality Assurance, 51 Milbank Q. 125 (1974) (earlier proposal to the same effect).
427. To be sure, reforms in roughly half the states now allow juries to reduce damages for income losses to the extent that the plaintiff has been compensated for such losses already by various so-called "collateral sources" (public or private income-replacement programs, for example). See generally Nat'l Ass'n of Mutual Ins. Companies, Collateral Source Rule Reform, http://www.namic.org/reports/ tortReform/CollateralSourceRule.asp (last visited Feb. 19, 2006) (summarizing collateral-source reforms by state). But, whereas some such reforms appear to diminish regressivity, many reforms focus only on denying double recovery for medical expenses, thereby increasing the relative significance of lost income as an element of tort damages. Under some new rules for calculating damages, therefore, higher-income persons stand to profit even more from the malpractice system than lower-income premium payers.
428. There is unfortunately only limited empirical evidence supporting our perception here. Nevertheless, some research finds disparities in malpractice awards correlated with race, gender, and age, factors that may also correlate with income. See Martha Chamallas, Questioning the Use of Race-Specific and Gender-Specific Economic Data in Tort Litigation: A Constitutional Argument, 63 Fordham L. Rev. 73 (1994) (finding evidence that the tort system's emphasis on economic over non-economic damages enables white men to recover more in tort awards than women and minorities);
429. Lucinda M. Finley, The Hidden Victims of Tort Reform: Women, Children, and the Elderly, 53 Emory L.J. 1263 (2004) (concluding that elderly plaintiffs and young children rely disproportionately on non-economic damages and would be adversely affected by certain tort reforms).
430. See text accompanying supra notes 109-11.
431. No court, it appears, has ever been asked to consider whether a statutory cap on non-economic damages denies equal protection - or its equivalent under a state constitution - to lower-income citizens. See Kelly & Mello, supra note 183, at 521-23 (outlining arguments accepted and rejected in state litigation). The usual equal-protection objection to damage caps - that they particularly disadvantage the most seriously injured victims - is hardly credible, one would think, when, ex ante, all consumer-patients are similarly situated. On similarly close analysis, however, there would seem to be merit in the argument that caps disproportionately affect lower-income persons and are particularly unfair to those who, having purchased standard health coverage, have reason to expect that a legislature will not deny them equal value for their money.
432. But see John Rawls, A Theory of Justice (1971) (purporting to prove logically, by reasoning behind a "veil of ignorance," that economic inequality can be justified only as a necessary side effect of an incentive system beneficial to the least advantaged).
433. As noted supra note 18, economists' work on distributional effects is generally descriptive rather than judgmental.
434. (reviewing Louis Kaplow & Steven Shavell, Fairness Versus Welfare (2002), extensively examining arguments in law and economics). But monopol's redistributive effects can be a potent political issue as we suggest they should be in the current state of U.S. health care.
435. See Barak D. Richman, Behavioral Economics and Health Policy: Understanding Medicaid's Failure, 90 Cornell L. Rev. 705, 715 (2005) (concluding that "for [certain] individuals, insurance status is a choice variable much more than an unavoidable consequence of poverty").
436. See supra note 8.
437. Anderson et al., supra note 13, at 90.
438. To be sure, these authors compare U.S. prices only with prices that are depressed in some measure by government-sponsored monopsony power, not with competitive prices equal to marginal cost. Moreover, some of the higher charges in the U.S. may be borne - not altogether fairly, see supra note 1 - by taxpayers rather than by premium payers as such. Nevertheless, these authors' observation of substantially higher prices in the U.S. system strongly supports our concern about the redistributive effects of U.S.-style health insurance.
439. Vladeck, supra note 36, at 41.
440. See supra notes 36-38 & 47-48 and accompanying text.
441. For data on the significant extent to which U.S. prices for brand-name pharmaceuticals exceed prices in other OECD nations (except Japan), see Patricia M. Danzon & Michael F. Furukawa, Prices and Availability of Pharmaceuticals: Evidence from Nine Countries, 2003 Health Aff. (Web Exclusives) W3-521 (also noting that generic drug competition, when patents do not preclude it, is generally stronger in the U.S. than in other countries).
442. See also supra note 27.
443. see also Alex Berenson, A Cancer Drug's Big Price Rise Disturbs Doctors and Patients, N.Y. Times, Mar. 12, 2006, at A1 (reporting large price increases of unpatented, single-source drugs for small market segments and pressure on insurers to cover such drugs). For present purposes, it suffices to understand that the pricing freedom of firms with market power is frequently greatly enhanced by health insurance even though in some circumstances insurers strengthen price competition.
444. Anderson et al., supra note 13, provide evidence that, even as the United States spends far more than other nations on health care, the rates at which Americans consume many important health services are generally no higher than in several other countries. Although these authors suggest that their data shows that overutilization is not a serious problem in the United States, evidence on national averages is not helpful in answering the question whether some Americans whose private insurance enables them to consume more and better than average health care are in fact buying more of it than is good for them in welfare terms.
445. See supra text accompanying notes 79 & 80.
446. Antitrust law treats so-called tying arrangements as unlawful when used by a seller of a unique and valuable product to force consumers desiring it to purchase an additional, perhaps unwanted good. See, e.g., Jefferson Parish Hosp. Dist. No. 2. v. Hyde, 466 U.S. 2, 12 (1984) (deeming a hospital's maternity and anesthesia services to be separate products that, if bundled, might be subject to antitrust law's prohibition on tying and stating that "the essential characteristic of an invalid tying arrangement lies in the seller's exploitation of its control over the tying product to force the buyer into the purchase of a tied product that the buyer either did not want at all, or might have preferred to purchase elsewhere on different terms"). Antitrust doctrine thus supports our view that consumers wanting basic health coverage should not also have to pay for unnecessary bells and costly whistles.
447. See supra note 11 and accompanying text.
448. Occasional experiments with "barebones" coverage have found few takers for it. See Mark A. Hall, Reforming Private Health Insurance 57 (1994);
449. Families USA, No Sale: The Failure of Barebones Insurance (1993). The apparent explanations are several. One of these is the difficulty that employers would face in offering potentially controversial coverage to workers who do not see the trade-off with their take-home pay. Second, low-cost options must compete with the safety net, which, though deficient in many respects, carries no up-front price tag at all. Third, "barebones" offerings, such as they were, were designed with most of the usual legal and conventional constraints intact. Thus, even when a state waived some mandates with respect to benefits, the effect was only to allow some additional categorical exclusions; "medical necessity" continued to govern basic coverage, and providers, partly because of legal compulsions, continued to practice in their usual costly ways.
450. See Havighurst, How the Revolution Fell Short, supra note 24, at 71 ("Unless and until a critical mass of employers offer benefits in forms that invite real economizing at the core of clinical practice, consumers whose welfare would be enhanced by purchasing revolutionary low-cost coverage will find no health plans offering it.").
451. See supra note 75.
452. See supra note 99 and accompanying text.
453. See, e.g., David M. Eddy, Evidence-Based Medicine: A Unified Approach, 24 Health Aff. 9 (2005) (summarizing industry reactions over time to revelations of widespread, unexplained variations in medical practice and of shortcomings in the way clinical policies are developed and implemented).
454. The cited article appears in a symposium showing the field's current efforts and mixed success in improving medical decisionmaking. Symposium, Putting Evidence into Practice, 24 Health Aff. 7 (2005).
455. See Symposium, The Quality Conundrum, Health Aff., July-Aug. 2002, at 12.
456. See Symposium, supra note 111.
457. Although the various activities reviewed in the cited symposia, supra notes 203-205, illustrate how the medical profession and health care industry as a whole regularly respond to revelations of inadequate performance, such efforts to do better, however sincere they may be, are also well calculated to maintain the profession's and the industry's elite status and to head off radical, exogenous reforms that would introduce real accountability to either government or consumers in the marketplace.
458. For a recent example of how even thoughtful commentators focus single-mindedly on how fairly health services are distributed while neglecting to consider where the cost burden falls, see Powers & Faden, supra note 86.
459. To be sure, consumers are generally ill-informed and cognitively challenged, and their preferences, as expressed in either market or political choices, are often incoherent. See supra note 181. But these circumstances do not automatically qualify elites, with interests of their own as well as different values, to make choices on their behalf.
460. See supra note 101.
461. See Todd Gilmer & Richard Kronick, It's the Premiums, Stupid: Projections of the Uninsured through 2013, 2005 Health Aff. (Web Exclusives) W5-143 (predicting very substantial increases in number of uninsured as cost increases continue to outpace increases in personal income). Once they become uninsured, of course, patients are much less likely to pay providers' bills, as the increasing number of bankruptcies caused by health-care-related liabilities reveals.
462. See supra note 51.
463. See Altman et al., supra note 34 (predicting economic instability for hospitals as demands for uncompensated care grow and revenue sources dry up). Other articles in the same symposium make similar predictions.
464. E.g., Dobson et al., supra note 40, at 30 ("This secular trend in combination with technology-driven health care cost increases has the potential to destabilize the U.S. health care financing system to the extent that low-wage (if not median-wage) workers will no longer be able to afford health care coverage.");
465. Vladeck, supra note 36 (noting potential for a "death spiral").
466. There are other reasons, to be sure, why economists are slow to assert that resources are being under- or over-allocated to a particular sector of the economy. For one thing, demand curves, while useful in theory, are a poor indicator of social welfare in markets for merit goods - which, by definition, should not be distributed solely on the basis of ability and willingness to pay. See Reinhardt, supra note 154, at 978-90 (noting the questionable social implications of relying exclusively on willingness-to-pay criteria, which demand curves incorporate, and of employing so-called Kaldor-Hicks cost-benefit criteria to evaluate welfare effects in health care markets). Another reason why allocative efficiency is problematic for economists is the so-called "problem of second best."
467. See generally F.M. Scherer & David Ross, Industrial Market Structure and Economic Performance 38 (3d ed. 1990) ("[O]ne might conclude that the whole question of allocative efficiency is so confused and uncertain, once second-best considerations are introduced, that policy-makers would be well advised to give up trying to achieve the best possible allocation of resources.");
468. Havighurst, How the Revolution Fell Short, supra note 24, at 80-81 (suggesting that, despite its force in weakening confidence that competition is always allocatively efficient, "second-best theorizing" should magnify, not diminish, concern that the nation is allocating excessive resources to underpriced health care). Despite the reasonableness of economists' hesitancy about relying on economic theory, however, this Article has shown that health insurance, especially as we know it in the United States, creates a situation with serious misallocative tendencies.
469. But see supra note 68 ("Whatever its magnitude, inefficiency that is already embedded in the economy is simply never going to be viewed as a problem by the political class.").
470. E.g., Michael E. Chernew, Increased Spending on Health Care: How Much Can the United States Afford?, Health Aff., July-Aug. 2003, at 15;
471. Henry J. Aaron, Should Public Policy Seek to Control the Growth of Health Care Spending?, 2003 Health Aff. (Web Exclusives) W3-28;
472. Mark V. Pauly, Should We Be Worried About High Real Medical Spending Growth in the United States?, 2003 Health Aff. (Web Exclusives) W3-15. The Pauly article, however, does address concerns similar to ours.
473. See also Pauly, supra note 24.
474. On the importance of focusing on marginal, rather than the more politically potent aggregate, benefits of health care spending in appraising allocative efficiency, see supra note 68.
475. For reasons why any such hope may be unrealistic, see David A. Hyman, Getting the Haves to Come out Behind: Fixing the Distributive Injustices of American Health Care, 69 Law & Contemp. Probs. 265, 273-82 (Autumn 2006).
476. Some health economists, for example, like to emphasize the limitations of both economic theory and markets in general and to criticize advocates of market-oriented policies for having simplistic, theory-driven views. E.g., Thomas Rice, The Economics of Health Reconsidered (2d ed. 2003);
477. Reinhardt, supra note 154. Yet to argue, as so many non-economists also do, that markets should not be trusted simply because they do not satisfy the rigorous conditions necessary to achieve Pareto optimality is also to use the unrealistic textbook model as the benchmark for a policy prescription. In fact, real-word markets do many things quite well despite their limitations, particularly in comparison with real-world government. Moreover, government can improve the market's performance by such measures as providing information, strictly policing fraud, and enforcing private contracts and the antitrust laws. Ideally, health policy debates focus on such practical, rather than ideological, considerations.
478. See supra note 181.
479. This model, which features menus of health-plan options maintained and explained to consumers by employers or other "sponsors," has been described most helpfully by its principal architect, Alain
480. Alain C. Enthoven & Richard Kronick, A Consumer-Choice Health Plan for the 1990s: Universal Health Insurance in a System Designed to Promote Quality and Economy, 320 New Eng. J. Med. pt. 1, at 29 & pt. 2, at 94 (1989).
481. See supra note 131, recalling the Clinton administration's proposed Health Security Act, which was based, ostensibly at least, on the managed-competition model and might have had a better chance of enactment had it been truer to that model. An easy way to ensure near-universal participation might be to offer a substantial, refundable tax credit to those who could demonstrate that they (or their employer) had spent at least that much on acceptable health coverage; then, anyone going uninsured would, by forgoing the tax credit, be effectively financing his share of the safety net (perhaps a residual public financing program) rather than free-riding on the health system's charitable impulses and capacity.
482. Perhaps what we have witnessed in the thirty years, and what is in store for the indefinite future, is a continouos playing out of the cycle that Guido Calabresi and Philip Bobbitt observed some years ago in public policies affecting so-called "tragic choices." See Guido Calabresi & Philip Bobbitt, Tragic Choices (1978). Their insight was that there are some situations (tragic choices) in which, even though economic efficiency may clearly dictate that unlucky individuals should bear some serious hardships rather than having them prevented or alleviated by public action, our political and legal institutions cannot, and will not, indefinitely accept such apparently avoidable tragedies. Instead, they predicted, public policy in such cases is destined to evolve endlessly in cycles, emphasizing at each stage some value - efficiency, compassion, fairness, or openness, for example - that previous policy had neglected. Under the Calabresi-Bobbitt hypothesis, it may be that seriously consequential choices about health care can never be permanently removed from the public agenda and placed finally (even with public subsidies) in private hands. On the other hand, there are other factors, including the tax subsidy, special-interest politics, and the peculiar division of policymaking responsibilities among federal and state legislatures and courts, that may account for the nation's failure, over many years, to put a coherent health policy permanently in place. Our hope is that this long-standing gridlock can be broken by publicizing the serious unfairnesses of the present system.
483. For the observation that a system based on private financing may not justify its high administrative and other costs unless it enables consumers to make consequential, welfare-enhancing choices, see Havighurst, supra note 12. Because the costs of a private system include those associated with legal and political uncertainties, it may be simply impossible for the American polity to adopt successfully and for all time the policy that strikes us not only as the ideal one but also as the one best suited for the diverse American people.
484. Given our emphasis on distributional issues, it is relevant to ask how a new public financing program, whatever its form, would be financed. The options include increasing progressive rates under the income tax, a flat tax on payrolls or total income, and a consumption tax. Although any of these would be fairer than the methods by which the health care industry currently finances the production of many public goods, a recent proposal by Victor Fuchs and Ezekial Emanuel sounds especially fair to us. Ezekial J. Emanuel & Victor R. Fuchs, Health Care Vouchers - A Proposal for Universal Coverage, 352 New Eng. J. Med. 1255 (2005) (suggesting value-added tax as revenue source).