By Richard A. Epstein
Right now, the Senate is anxiously considering HR-SA 3590, the Patient Protection and Affordable Care Acta.k.a. the Reid Billwhich builds on earlier efforts in the Senate and House to reach a new consensus on health-care reform. Many legislative uncertainties remain, but its key characteristics seem fixed in stone, and they highlight the radical nature of this legislation.
Senator Orrin Hatch has long urged that the legislation is unconstitutional for its overreaching on individual choice. This paper focuses on the constitutional question in the ratemaking context, by comparison to analogous regulations in the context of public-utility regulation.
One telling sign of the relevance of this analysis comes from the Congressional Budget Office ("CBO"). In a recent release, it has treated the proposal as if it nationalizes much of the private health insurance industry, most specifically because it may well require that rebates to customers kick in whenever, in its words, "medical loss ratios are less than 90 percent." In plain English, the Reid Bill assumes that health-care administration, which is always costly, can be done cheaply even in the new legal environment, so cheaply in fact that these health-insurance rebates kick in whenever insurers' administrative expenses exceed 10 percent of their premium dollar. As the CBO has concluded, "this further expansion of the federal government's role in the health insurance market would make such insurance an essentially governmental program ..."
In effect, the onerous obligations under the Reid Bill would convert private health insurance companies into virtual public utilities. This action is not only a source of real anxiety but also a decision of constitutional proportions, for it systematically strips the regulated health-insurance issuers of their constitutional entitlement to earn a reasonable rate of return on the massive amounts of capital that they have already invested in building out their businesses.
In order to make out this argument, let me proceed as follows. In part I, I shall give a general overview in order to place in context the system of health-care regulation that shall be operated through the State Exchanges that would be formed under the Reid Bill. In part II, I shall give a detailed analysis of some of the major provisions of the Reid Bill. In part III, I shall give a brief analysis of the economic assumptions that underlie the Reid Bill, and the way in which they are likely to lead to extensive price fixing. In part IV, I shall flesh out the constitutional implications of the above analysis. I shall then close with a brief conclusion, which recommends that the Reid Bill be scrapped.
I. AN INSTITUTIONAL AND CONSTITUTIONAL OVERVIEW The concern that I wish to address at the outset deals with the unprecedented level of systemic coercion that the Reid Bill exerts on the various firms that supply health-insurance coverage in the small-group and individual health-insurance markets. Constitutional concerns with these provisions arise even if one assumes that Congress has the power under the Commerce Clause to pass comprehensive regulation of health care in this form. Independent of any question about the scope of Congressional power, it is critical to consider that the Fifth Amendment affords regulated health-insurance companies protection against the taking of property without compensation and without due process of law.
These overlapping guarantees bind both the federal government and the state governments in all of their activities, whether undertaken jointly or independently.
These constitutional provisions have been subject to extensive interpretation in the Supreme Court in ratemaking cases, which must be taken into account in dealing with the legislation. The Supreme Court's basic constitutional requirement is that any firm in a regulated market be allowed to recover a risk-adjusted competitive rate of return on its accumulated capital investment. See Duquesne Light Co. v. Barasch, 488 U.S. 299 (1988).
The Reid Bill emphatically fails this test by imposing sharp limitations on the ability of health-insurance companies to raise fees or exclude coverage. Moreover, the Reid Bill forces on these regulated firms onerous new obligations that they will not be able to fund from their various revenue sources. The squeeze between the constricted revenue sources allowable under the Reid Bill and the extensive new legal obligations it imposes is likely to result in massive cash crunch that could drive the firms that serve the individual and small-group health-insurance markets into bankruptcy.
Although the Constitution requires that regulations permit regulated firms to recover a risk-adjusted competitive rate of return, the Supreme Court has left it up to federal regulators to decide which approach to rate regulation they wish to take. On page 20, I discuss the two major tests that are used to determine whether rates are confiscatory. In one instance, the risk of imprudent investments is left on the firm, and in the other, it is imposed on the ratepayers. Here, the combined effect of all the provisions in HR-SA 6590 makes it unimportant to choose between these two tests, because the blunt truth is that the Reid Bill flunks both tests.
To make this analysis more concrete, it is important to understand the pervasive transformation that the Reid Bill, if passed, will work on both suppliers and users of all health-insurance services. On the one hand, the Reid Bill depends on a combination of huge general tax increases, which is coupled with special levies on industries such as medical-device and pharmaceutical companies. These tax revenues are then used to fund subsidies for large segments of the population in order to allow them to purchase qualified health-care plans that are sold through a set of State Exchanges that the Reid Bill creates. In order to prevent these subsidies from flowing through to the various health-insurance issuers, the Reid Bill imposes extensive obligations on any health-insurance issuer or health-plan provider that wishes to participate within the system in order to keep them from capturing subsidies meant for others. The effect of the subsidies is to increase the level of health care that will be demanded in the United States. The effect of the regulations is likely to be to impose huge costs on various health-insurance companies as they struggle to meet the influx of demand when they are at the outer limit of their capacity.
There are at this point enormous uncertainties about how this entire scheme will play out. My view is that it will prove ruinous on all three fronts. The general public tax increases will be so sharp that it is unlikely that they will generate additional revenues. The subsidies will be so large that the demand for medical services will be left largely unsatisfied, so two consequences are likely. First, an increased queuing for various health care services is to be expected. Second, there will be increased pressure to exclude large groups of people from the system, on the lines of Massachusetts's recent decision to cut from its system 31,000 legal immigrant aliens (who pay taxes but do not vote).
Furthermore, on the supply side of the market, all health-insurance companies will find themselves in an impossible dilemma. If they decide to offer their health-insurance plans outside the State Exchanges, they will be unable to compete for the subsidized consumers who are only able to spend their tax dollars within the framework of the State Exchanges. Their position will be worse because they shall continue to be subject to all present mandates and regulations that have an impact on their business. Insurers outside the Exchanges also face the likely prospect that they will still be further taxed and regulated to help finance the intolerable burdens that arise under the subsidized insurance supplied within the State Exchange system.
Therefore, it is impossible in my view to look in isolation at the regulations of the health plans and health-insurance issuers that operate under the State Exchange system. This is not a case in which a lonesome competitor complains about a subsidy that some private person gives to its competitor. This is a case in which the party that provides the subsidy to health-insurance consumers for use within the State Exchanges also has the power to regulate and tax the non-Exchange competitors in whatever fashion it sees fit. These impositions are, of course, not only applied at the federal level, for the full consideration of the regulatory burden must also take into account any additional regulations and taxes that the states, with explicit Congressional blessing, are allowed to impose on health-insurance plans and insurance issuers that remain outside the State Exchange system.
This level of systemic coercion frames the debate about the constitutionality of the Reid Bill. Those parties that do not wish to suffer the Bill's regulations in order to gain access to a subsidized consumer base are not free to compete in an unregulated market. Direct federal and state government regulation remains a fixed feature of their life. Government regulators at the state and federal levels have both the power and the motive to hit non-Exchange health insurance issuers with a range of taxes and regulations that could quickly make their economic position intolerable. I reached this conclusion before I read the recent CBO report, which concluded that the level of government regulation with the new proposal to require rebates when the Medical Loss Ratio was under 90 percent left so little flexibility for private carriers that they should no longer be treated as nongovernmental entities. It is worthwhile to quote the CBO's words here:
Certain policies governing MLRs, particularly those requiring health plans whose MLR falls below a minimum level to rebate the difference to enrollees, can be a powerful regulatory tool. Insurers operating at MLRs below such a minimum would have a limited number of possible responses. They could change the way they provide health insurance, perhaps by reducing their profits or cutting back on efforts to restrain benefit costs through care management. They could choose pay the rebates, but if they raised premiums to cover the added costs they would simply have to rebate that increment to premiums later. Alternatively, they could exit the market entirely. Such responses would reduce the types, range of prices, and number of private-sector sellers of health insurance...
The CBO of course is not charged with drawing the constitutional implications from its findings. But its report reaffirms what should be evident from the Reid Bill itself, namely, that health-insurance issuers subject to the rebate provisions are practically forced to operate within the State Exchange system where the guaranteed-issue and renewal provisions coupled with the onerous requirements of the essential-benefit plans put them in this impossible position: They cannot earn a reasonable return on their investment, which is required under the Takings and Due Process Clauses of the Fifth Amendment to the Constitution. Let me now offer a more detailed analysis of how this looming tragedy is likely to play itself out.
II. AN ANALYSIS OF H.R.-S.A. 3590THE REID BILL The most obvious feature of the Reid Bill is the incredible level of coercion it imposes on the private companies that supply health-insurance coverage, levied in a coordinated one-two attack. On the one hand, the Reid Bill imposes major requirements on how they do business. On the other, it imposes powerful financial limitations on the revenues that such firms can collect for the provision of their services. Yet the Reid Bill contains no mechanism that guarantees that the revenues in question will be sufficient to cover the new obligations that it imposes. Instead, the Reid Bill relies on extensive but standardless delegation to the Secretary of Health and Human Services to fill in the gaps of the legislation. (The Reid Bill does not create, as does section 241 of H.R. 3962, a new Health Choices Administration with its own commissioner.)
The range of matters that are subject to administrative control under the Reid Bill transforms a large sector of the insurance industry. The traditional law of insurance gave the insurer the complete power to determine whether to accept or reject a given risk, or to determine the premiums to be charged to an insured, the policy limits, and the terms and conditions on which the policy was issued. The duties to disclose were extensive but these were correctly imposed on the insured who alone possessed the relevant knowledge about the nature and scope of the risk.
In this traditional environment, regulation of insurance companies was directed to two different issues. The first was a general form of consumer protection, which requires a full disclosure of the terms of policies, which of course does not negate the critical duties to disclose information about material risks imposed on insured parties. The second went to the issue of solvency, in order to counter the real risk that the insurance carrier that had accepted a premium today might not be around to pay off the health care bills that it had promised to pay tomorrow. Competitive forces were generally used to determine premiums. The various efforts to impose mandatory price controls and coverage on insurance contracts in both the individual and small-group markets have typically driven up the cost of business and have resulted, for example, in a reduction of the number of individual employees who are covered by voluntary plans.
What is most striking about the combined effect of the various provisions in the Reid bill is its cavalier disregard for the long-term stability of all segments of the private health-insurance market, which are likely to be caught in a pincer between the heavy mandates for coverage on the one side and their inability to exercise any underwriting control over their book of business on the other. The Reid Bill does not achieve this objective by imposing direct restraints. Instead, its preferred method of social control lies in the power of the Secretary to designate any health plan of a given insurance company as a "qualified health plan" ("QHP"), as defined in section 1301, which allows the health-insurance company to serve customers who are eligible for financial assistance under this bill. The size of these various benefits is sufficiently large that no company that fails to become a QHP issuer is likely to survive in an insurance market in which coverage is offered on the Exchange, as few people will prefer to purchase a full-price plan to a heavily subsidized one. The restrictions imposed on QHPs, however, are so onerous that all health insurance companies are in effect caught in an impossible bind. The only way to reach subsidized customers is to submit to ruinous financial regulation. The system, therefore, operates in effect as a direct set of controls on virtually all companies that wish to remain in the marketplace.
Let me set out some of the key provisions that are likely to have negative impact on firms, which are added to the Public Health Service Act by section 1201 of the Reid Bill. Section 2704 provides: "[a] group health plan and a health insurance issuer offering group or individual health insurance coverage may not impose any preexisting condition exclusion to such plan or coverage." Unlike the earlier House and Senate Finance Committee bills, there appear to be no exceptions to this particular rule. Yet section 2705 makes it clear that the "health status, medical condition, claims experience, receipt of health care, medical history, genetic information, evidence of insurability, disability, or source of injury (including conditions arising out of acts of domestic violence) or any other health status-related factor determined appropriate by the Secretary" cannot be taken into account in setting rules concerning enrollment eligibility.
In addition, section 2702 provides that every health insurance issuer in the individual or group market must accept all applicants. Section 2703 imposes the similar requirements on health-insurance issuers to renew or continue in force all individual and group plans. These sections create the risk that some firms will be inundated with applications that they will be unable to serve. Unfortunately, the mechanisms that the Reid Bill relies on for dealing with the capacity questions are unequal to the challenge of regulation in any fast-moving and complex market.
The technical amendments found in section 1562 of the Reid Bill would make applicable to all issuers in the group and individual health-insurance markets certain provisions in section 2711 of the current Public Health Service Act, 42 U.S.C. § 300gg-11, which are currently applicable only to issuers in the small-group health-insurance market. These provisions are codified at the end of new section 2702.
To begin with, the provision, as amended by the Reid Bill, would require issuers in the group and individual health-insurance market to accept every employer and eligible individual who applies for coverage, unless such an issuer could demonstrate to the applicable state authority that it lacks the financial capacity to underwrite additional coverage. No firm has the capacity to underwrite all the business that comes its way. At some point, its marginal cost of supplying additional coverage becomes prohibitively high. It is relatively easy for any health-insurance issuer to determine that point for itself. It is far more difficult for it to have to incur the time and expense to demonstrate that point to some applicable state authority, which will then be given the power to determine, without bearing financial responsibility of its mistakes, just how much business the health-insurance issuer must underwrite, and which individuals should be eligible for that coverage.
The amended provision also requires the health-insurance issuer to further show that its decisions to deny coverage are not based on the differential costs of supplying that coverage to various individuals and groups under new section 2702 of the Public Health Service Act. And once it makes those determinations, it must remain out of the market for 180 days unless it gets state administrative approval. When market conditions change, or some individuals or groups opt out of coverage, yet another hearing is required to reenter into the market. There is no assurance that these applicable state authorities have either the resources or capacities to determine these multiple questions for huge numbers of health-insurance issuers within the short period necessary to allow them to participate effectively in the health insurance market, including in the Exchange. There does not appear to be any federal funding to help out the states in the discharge of these responsibilities.
The overall level of federal control is heightened by the requirement that all health-insurance issuers in the individual or small-group market provide an essential benefits package that includes a wide range of "ambulatory patient services, emergency services, hospitalization, maternity and newborn care, mental health and substance abuse disorder services, including behavioral health treatment, prescription drugs, rehabilitative and habilitative services and devices, laboratory services, preventive and wellness services and chronic disease management, pediatric services, including oral and vision care." Section 2707. These essential benefits are subject to explicit minimum requirements that the Secretary "shall ensure that such essential health benefits reflect an appropriate balance among" the required services set out in section 1302(b)(1), not discriminate in any way "against individuals because of their age, disability, or expected length of life," and cover the full needs of "diverse segments of the population, including women, children, persons with disabilities and other groups." No health-insurance issuer that participates in this market will be able to skimp on coverage.
In addition, the cost of compliance is heightened by the requirements of section 2711, which covers all group health plans and health-insurance issuers and prohibits the use of "lifetime limits on the dollar value of benefits for any participant or beneficiary" (except for certain exempt plans that are not otherwise regulated under state or federal law) and "unreasonable" annual limits on the value of benefits for any participant or beneficiary. At the same time that these coverages are guaranteed, the Reid Bill limits the amount of cost-sharing that can be required of plan participants and the size of the deductibles ($2,000 for an individual and $4,000 for a family in the small-group market). Such is the command of section 2707(b), making section 1302(c)(2) applicable to health insurance issuers in the individual and small-group markets. Finally, termination of enrollees is often difficult under section 2712, which allows for the rescission of coverage only with respect to any individual enrollee "who has performed an act or practice that constitutes fraud or makes an intentional misrepresentation of material fact as prohibited by the terms of the plan or coverage."
The combined impact of these interconnected provisions is clear: there is no feasible way that an insurance carrier can respond to the increased costs of servicing of its book of business either by declining coverage or by reducing services. With all escape hatches closed, the critical question is whether the health-insurance issuer is in a position to raise rates in order to offset the risks in question. On this question, section 2794 introduces a complex system of de facto price controls that depends on the close cooperation of state and federal officials. The initial process that goes into effect in 2010 requires the Secretary and the states to develop a plan to look for "unreasonable increases" in charges for insurance coverage. At this point, all health-insurance issuers must submit to the state insurance commission authority "a justification for an unreasonable premium increase prior to the implementation of the increase." (It is not stated as to how one justifies increases that are, by definition, unreasonable.) Thereafter, once the information has been submitted and evaluated, it appears that the state insurance commissioner shall make appropriate recommendations "to the State Exchange about whether particular health insurance issuers should be excluded from participation in the Exchange based on a pattern or practice of excessive or unjustified premium increases." In effect, it appears that the State Exchanges can exclude health-insurance issuers from offering their plans through the Exchanges, at which point the subsidies to insurers will be lost.
As of 2014, the Secretary is put in a position to "monitor premium increases of health insurance offered through an Exchange and outside of an Exchange." Section 2794(b)(2)(A). Of equal importance, the combined effect of section 2794(b)(2)(B) and section 1312(f)(2)(B) would seem to permit the states to influence the premiums charged in the large-group market, even when no large-group coverage is offered through the State Exchange. And the states are put in the position under the Reid Bill to deny Exchange access to those small-group and individual plans that they determine have excessive rates. That power to exclude from the Exchange remains a death knell for all plans in health-insurance markets in which coverage is offered through the Exchange, so that the power to exclude again converts itself into a de facto power to set maximum rates.
In addition, after 2014, it is at least possible that the power of the Secretary to "monitor" could be read as a way to introduce price controls through the back door. The natural reading of the section seems not to support that view, but administrative officials often receive deference in their statutory interpretations under Chevron, Inc. v. National Resources Defense Council, 467 U.S. 837 (1984). More specifically, the primary meaning of "monitor" is to "observe" or to "keep an eye on." But a secondary meaning is to supervise, which in this context might be read to give the Secretary those more expansive powers.
The level of regulation over prices is not confined to these provisions. As currently configured, the Reid Bill contains global caps on profits that operate on a heads-I-win-tails-you-lose principle. Thus, section 2718(a) of the Public Health Service Act, as added by the Reid Bill, imposes on "[a] health insurance issuer offering group or individual health insurance coverage" reporting obligations on the amount of premium dollars that are spent on "clinical services," activities to "improve health care," and all other "non-claim costs." Up to the end of 2013 (unless extended), if this information reveals that the non-claim charges exceed 20 percent of total costs in the group market, or 25 percent in the individual market, the Reid Bill provides that, in the first instance, there "shall" be an annual rebate in the amount of the excess over that level, section 2718(b)(1)(A) & (B). Section 2718(b)(1)(A) & (B) also provide that the individual states may order a reduction in the 20 or 25 percent non-claim figures, so that the annual rebate kicks in at a lower percentage.
In making that determination, under section 2718(b)(2), the states "shall seek to ensure adequate participation by health insurance issuers, competition in the health insurance market in the State, and value for consumers so that premiums are used for clinical services and quality improvements." The amount of any state-adjusted rebate under section 2718(b)(1)(B) with respect to coverage in the individual health insurance market can be adjusted only if the Secretary determines that such rebates will "destabilize" the market. It is difficult to understand how these inconsistent commands can be simultaneously achieved. "Adequate participation" suggests that rates must be kept high enough to keep firms in the market, while "value for consumers" pushes strongly in the opposite direction. In the middle, "competition in the industry" is effectively gutted by the extensive system of regulation that prevents firms from gaining extra profits from valuable new innovations in health-care management or delivery systems.
The most insidious feature of this provision, however, lies in its unconstitutional insistence on a global limitation on the profits obtained by any firm that runs this regulatory gauntlet. The first point here is that no firm would be able to show that its non-claim expenses do not exceed the maximum statutory allowances. Given the high level of administrative costs that the Reid Bill imposes, this is, to say the least, a strong possibility. It could easily be that the health-insurance companies will find themselves in the unenviable position of having to issue rebates at a time when they are operating at a loss. In addition, neither bill takes into account the strong likelihood that overall costs will vary from year to year. These statutory provisions are strictly one-way ratchets, such that the gains, if any, in one year will be taxed away even if there were losses in previous years, or even if losses are expected in future years. The lack of any averaging provision, therefore, has the unhappy effect of placing a hard cap on earnings that is unrelated to the overall risk of the venture.
There is yet another feature that requires some brief notice, for it goes to the stability of preexisting plans. One of the ways that the legislation allows for "acceptable" coverage to be obtained is by enrollment in a "grandfathered" health-insurance coverage plan. That sanctuary is far narrower than is commonly supposed. Under section 1251 of the Reid bill, the only new enrollees that allow the plan to keep its protected status relative to the State Exchanges are those that admit new dependents of an employee or new employees. The admission of any other person into the plan, such as retirees, will eliminate the preferred grandfathered status under section 1401 of the Reid bill, which adds section 36B to the Internal Revenue Code; section §36B(1)(b)(3)(c)(iii) determines the eligibility of plan enrollees for various refundable tax credits or other premium assistance.
Section 1251 is, however, silent with respect to other possible features. The definition of "grandfathered Health Insurance Coverage" in section 202 of the House Bill made it crystal clear that this grandfather status would be lost when an existing plan added or removed a benefit, or increased the premium for one group of employees unless the same proportionate increases were imposed for all plan participants. The Reid Bill is silent about the effect of these changes, which ordinary plans necessarily make dozens of time each year. The question, therefore, remains unanswered as to whether most grandfathered plans will be able to maintain their preferred status, or whether they will lose that preferred status and be required to meet all the substantive and procedural requirements necessary for new plans to participate in the State Exchanges. Questions of this magnitude should not be left to administrative discretion, but should be resolved in the Reid Bill itself.
III. FROM COMPETITION TO PRICE CONTROLS: THE ECONOMIC ANALYSIS Whether or not the Secretary is able to claim more extensive powers, it appears that the complex price-control mechanism implicit in the Reid Bill operates on the indefensible economic assumption that price controls are needed to wring inefficiencies out of the operation of health-insurance issuers. Yet economic theory unambiguously leads to the opposite conclusion. To take only the extreme cases, the logic behind pure competition is that it leads to efficient outcomes because the ability of customers to go elsewhere leads firms to reduce the costs required to meet any specified level of service. In general, a competitive market is regarded as socially optimal because there is no movement in price, quality, or quantity that could make one party better off without making another party worse off. Any assumption that price controls, however implemented, offer some hidden road to the efficient allocation of resources has been repeatedly exploded. One need only think of the systematic cut back in services that is the hallmark of rent-controlled apartments to understand this basic economic principle.
The logic of competition is clear. In competitive markets, the firm is always engaged in a delicate balancing act whereby it must ask whether the additional services that it could supply will generate revenues equal to, or greater than, its cost of providing those services. The pressure of competition could never force a firm to offer its products or services for sale at a loss. At the same time, the ability of customers to go elsewhere will drive prices down to the marginal cost for the provision of those products or services. Even the existence of monopoly power does not allow any firm to make any money if it is forced by regulation to provide goods and services below their costs. The existence of monopoly power only speaks to the possibility of a firm raising its prices above marginal costs, for which rate regulation may be an appropriate response, although always difficult to implement. But there are clear caveats here that need to be observed first.
First, to justify rate regulation, there needs be some evidence of the existence of monopoly, which is not likely in health-insurance markets in which multiple parties are already in competition with each other. To be sure, that condition is not uniform, in part because of the state restrictions on entry that are allowed under the 1945 McCarran-Ferguson Act, 15 U.S.C. §§ 1011-1015. But the existence of this entry barrier does not require any form of rate-of-return regulation. It is a simple matter to repeal McCarran-Ferguson to the extent that it authorizes state barriers to out-of-state competition. That one legislative fix should reduce prices and expand access, but not cost the federal government a dime.
Second, the rate regulation imposed by the federal government cannot be allowed to become confiscatory by denying the firm the ability to recover an appropriate return on its capital. There is nothing that a system of price controls can do to lower costs. In fact, price controls generally increase the costs of production to firms by forcing them to meet heavy compliance costs. Under the Reid Bill, the costs of providing service will necessarily increase because of the heavy compliance costs imposed on health-insurance issuers, the uncertainty of their business position, and their inability to select or decline customers or to set premiums in accordance with known risks of various individuals or groups. There is no reason whatsoever to think that any firm operating in this heavily controlled environment could eliminate inefficiencies from its current operation to offset these losses. To give an analogy, one of the major problems of rent-control and rent-stabilization systems is that once landlords are revenue constrained, they cut services to save costs. They were already doing their best before the regulation, which offers no magic bullet. The logic same holds here. Services will be cut or delayed, in either visible or covert fashion, just as the recent CBO report indicates. As far as I know, there has never been a price-control system that can improve quality of output, and there never will be.
Third, it is wholly unclear as to how private firms will be asked to price their services under these new mandates dealing with guaranteed service and preexisting conditions. One possibility, which seems inconsistent with Section 2794, is to allow for competitive pricing without allowing the state to set the prices which are required. But given the requirements under the Reid Bill, that position will not lead to an offer to supply health insurance at a price that is lower than the blended cost incurred for serving all potential customers. Those numbers could easily make the insurance unaffordable for all but the most sick people. As healthier individuals either stay out of, or abandon, the health-insurance market because of high premiums, the blended rate will have to increase. Quite simply, the risks of adverse selection by insureds are enormous: those individuals whose health prognosis improves could leave the system, while those whose condition has worsened will continue to demand coverage at the same rates as before.
Fourth, persons who choose to stay out of health-plan coverage when healthy (even by paying some tax) will migrate to the plans quickly once their own health condition deteriorates. They would have complete and accurate knowledge of their own condition that they would not need to disclose to the insurer, which is in effect under a statutory duty to enter into a losing contract. The scope of the potential liabilities is only increased by the huge number of individuals to whom this option is made freely available. Wholly without any other consideration, the key requirements prohibiting the use of pre-existing condition exclusion and requiring guaranteed issue and renewal could easily impair the success of health insurance issuers. Ironically, the programs that have the best coverage are the ones that are most at risk, as there is nothing in either the Senate or House bill that appears to alter the rates to account for the differential level of services offered. Other things being equal, the dominant response under this type of mandate would be to reduce the level of coverage across the board, thereby decreasing the options available to many plan recipients. But even this option is blocked by the statutory requirements for "essential health benefits," with their mandatory minimums. On this score, the degrees of freedom to vary rates that the Reid Bill allows health-insurance issuers in the individual and small-group markets relate to "only" four factors, whether or not the coverage is offered through the Exchange. Section 2701(a)(1), as added by section 1201. In addition, these provisions apply with equal force to all health-insurance issuers in the large-group market if the state allows any insurer to sell large-group health insurance coverage through its Exchange.
What is noteworthy about these rate-setting provisions is that they do not allow the health-care-insurance issuers the ability to accurately price their products. The first factor stipulates that coverage for individuals may differ from that offered to families. The second calls for states to establish one or more rating areas within their respective states, subject to review by the Secretary if its areas are found to be "not adequate," under criteria that are nowhere specified in the bill. The choice of these areas--are cities and suburbs in the same or different areas, for example--could be critical for individual plans because the total premium is likely to be sensitive to the areas that are chosen. A health insurance carrier that has an expensive book of business will not be able to adjust the rates accordingly once the geographical boundaries are determined. That nonindividualized treatment could easily create substantial losses for some firms and competitive advantages for others. Finally, some variation is allowed for age up to a 3-to-1 ratio, and for tobacco use up to a 1.5-to-1 ratio. Both these numbers are smaller than the actuarial difference among these groups, so that this provision also requires cross subsidies among plan participants under the guise of "prohibiting discriminatory premium rates," set out in section 2701(a).
At this point, the only possible response of companies is to raise prices to levels that could easily prove economically and practicably unacceptable. Yet once that is done, the prohibitions against "unreasonable" premium increases of section 2794 kick in to make it highly likely that the offending heath insurance issuer will be thrown off the Exchange. But what other alternative is possible? Piling one mandate on top of the other places powerful pressures to impose price controls on the health insurance issuers. After all, the requirement for guaranteed renewal will not satisfy the purposes of universal coverage with the State Exchanges if it is only offered at a price that is beyond the reach of the individuals whom it is supposed to benefit. In the end, therefore, I think that the implementation of the Reid Bill will lead key government officials to impose direct and comprehensive price controls.
This point requires some elaboration. Every scheme that denies a firm the ability to refuse to deal with potential customers has to have either a nondiscrimination rule or a price-restriction rule or both. Thus, standard public utilities have to take all comers. In some instances, they do so on a first-come, first-served basis, as was the case typically when railroads were so regulated. Or these utilities need to articulate some rule that requires a cut back in services offered to earlier customers to make way for later ones, as was typically the case with public-utility hook-ups for gas and electricity. But those provisions will not work in an environment that imposes specific duties, for customer access could easily be denied by what some government administrator deems to be systematically high prices. The only way to make sure that these regulated plans provide access is through some system of oversight on the rates that can be charged.
At this point, the Reid Bill exacerbates the major difficulties of government regulation. The voluntary market under competition will never price goods and services below their cost to the firm. As these costs go up, the health-insurance markets will shrink, for it is quite likely that this mega-mandate will provide many people with services that they do not want and cannot afford. The only two options then are to take some benefits out from the mandate, or to impose price controls at either the state or federal level. Clearly the latter is more likely to be chosen, but there is nothing in either regulatory scheme to rule out the risk that the prices charged will not cover the full costs of providing the benefits. Once again, the risk of price controls is close at hand, even without any explicit authorization on the point.
The situation is still made worse because the federal standards are best understood as creating floors and not ceilings. Indeed, section 1311(d)(3)(B) of the Reid Bill coordinates federal and state programs by providing that state benefit mandates continue to apply to Health Insurance Exchange participating plans so long the state agrees to reimburse the federal government for any increase in premium credits that is attributable to the premium increase arising from the mandate. These additional demands could prove to be extremely costly to companies that seek to acquire nationwide coverage for their employees in the face of specialized mandates that often vary in their economic impact across state lines.
IV. CONSTITUTIONAL ANALYSIS This detailed analysis of the Reid Bill helps to set up the appropriate constitutional analysis. The applicable standards for constitutional review have usually been developed in connection with rate-making procedures in natural monopolies. Within this context, the social objective is to limit the monopoly returns to public utilities, which do not face the risk of competition from new entrants, because they operate in a market in which the declining marginal cost of the initial entrant prevents a new entrant from gaining a toehold. In such a situation, one permissible legislative response is to impose some form of regulation that brings that established player back to a competitive rate of return. I shall pass by all the difficulties in implementing such a program. It is important to note, however, that it is never a satisfactory response for regulators to drive the rates of return down to zero, for then no one would ever be prepared to provide services.
Since it is necessary to compete for capital across the entire range of activities, the constitutional protection afforded under both the Takings and the Due Process Clauses provides that the rate of return cannot fall below that which the investors in the firm could obtain in a competitive market. That calculation has to take into account the level of risk associated with the business, which in general is low with respect to public utilities that have at least de facto protection against new entry.
The hard question, therefore, is what kinds of systems of rate regulation will pass constitutional muster. Within the traditional ratemaking system, the first issue concerns what goes into the rate base. One view is to allow the firm only to include those investments that remain used and usable in the business, which means that the firm has to take the risk of investments that go to waste. For taking this risk, it receives a higher risk-adjusted rate of return. See Smyth v. Ames, 169 U.S. 466 (1899). The alternative is to permit the use of a broader rate base, and to allow therefore a lower rate of return because the risk of poor investments falls on the ratepayers. See Federal Power Commission v. Hope Natural Gas, 320 U.S. 591 (1944). In other instances, it is possible to avoid the cumbrous ratemaking proceedings by instituting a system of rate caps, which in effect tell a firm in an industry like telecommunications that its rate increases will be capped because the increased efficiencies in doing business mean that the unit costs of supplying services will always be on the decline.
What is striking is how far the ratemaking system for health insurance is from all the above. There is no natural monopoly in health insurance, and there is a powerful way to open up health-insurance markets by knocking down the state barriers to entry that have been in effect since 1945 under the McCarran-Ferguson Act. Once it is clearand it is generally clearthat the health insurance industry is competitive or could easily be made competitive, the entire rationale for government ratemaking is undermined. The point of ratemaking was to require the firm to accept competitive rates of returns in a market setting where it enjoyed monopoly power. Here, the market is either competitive already, or easily can be made so. In this environment, ratemaking no longer serves any useful function.
Contrary to the implicit assumption behind the Reid Bill, ratemaking cannot induce further efficiencies once competitive forces have driven out all elements of monopoly power. Yet all firms are trapped, for the only way in which they can escape ever more onerous requirements and restrictions is to render themselves ineligible to enroll new groups or individualswhose health insurance, of course, their tax dollars will continue to fund. In addition, right now the Reid Bill subjects plans outside the Exchanges to certain other legal requirements, which could easily be tightened down the road. The non-Exchange health insurance issuers are, therefore, placed in an untenable position that exposes them to the multiple strategies in the Reid bill that control rates and set the terms of service and that will have three unacceptable consequences: (1) to reduce the rate of return of health insurance companies below competitive levels, (2) to pile expensive administrative mandates on them, and (3) to generate major uncertainties as to how the federal obligations on such companies will pan out.
At this point, there is a near mathematical certainty that the scheme of health-insurance market regulation contemplated by the Reid bill will reduce the risk-adjusted rate of return below the level needed to keep these firms in the individual and small-group health-insurance markets. I am not aware of a single provision in the Reid Bill that looks to ensuring a minimum rate of return. And there are countless provisions in the bill that impose new obligations to cover services while eliminating the revenue sources to deal with them. It is just this combination of regulatory programs that leads the CBO to treat private health insurance issuers as part of a federal program--as though they have been subject to de facto nationalization.
This systemic regulation of both Exchange and non-Exchange carriers shows, moreover, that those health-insurance issuers that participate in the Exchange are shorn of all constitutional rights. The requirement that the states order rebates of money spent on non-claim expenses is not constitutionally permissible unless and until the Reid Bill makes some allowance for earning a reasonable rate of return. That return, moreover, must take into account the extra riskiness that flows from the grant of broad delegated authority to the Secretary.
In addition, the decision to order rebates in good years without adjustments for the losses in bad years makes it impossible for a firm to earn a reasonable rate of return. In utility rate regulation, it is not constitutionally permissible to impose an annual rate cap just at the competitive level, while leaving the carrier obligated to eat the losses in poor years. Section 2718 of the Reid bill goes even further than such unconstitutional provisions in the utility context: it imposes a hard cap, without any accurate accounting for administrative costs or any explicit recognition of the constitutional right to earn a reasonable profit.
To make matters worse, these overall caps apply on top of all the restrictions on the ability to decline coverage or vary rates that are involved in other provisions of the Reid Bill. These provisions necessarily raise the administrative costs of providing insurance. There are no upper bounds on what can be required by various federal and state officials who are charged with oversight of individual and small-group plans in many instances, and all health-care plans in others. At this point, it is only a matter of time before the cost obligations are so enormous that even complete freedom in setting prices would not allow the firm to remain in business. Nor will this problem be cured by the vast pattern of subsidies and taxes that permeate the rest of the bill. Quite to the contrary, the subsidies may put greater pressure on the capacity of health insurance companies to operate, given that these firms have no capacity to choose which plans to provide to which customers.
Given these facts, it is impossible for the rate regulation of firms in the competitive health insurance industry to recover the constitutionally permissible rate of return. So long as competitive rates of return remain the constitutional benchmark, rate regulation necessarily fails. The unregulated rates are already at the competitive level. Any system that reduces revenues, raises costs, and increases uncertainty cannot possibly meet the applicable constitutional standard.
To my mind, the only serious question about the legislation is whether a facial challenge will be allowed to the Reid Bill when it does not contain explicit price-control features. Such facial challenge are often denied in land-use cases, but in rate-regulation cases the result has usually been otherwise. To wait until the program has run its course is to consign a health-insurance company to the substantial risk of bankruptcy just for trying to stay in business. It does not have the option to hold off development until the legal uncertainties are resolved. Since neither the United States nor the individual states will pony up the huge losses sustained by the regulated firms, the challenges have to be allowed before the statute is implemented and not afterwards. How this issue will play out in litigation no one can say for sure. But it would be, in my view, irresponsible for the Senate to pass any health reform legislation that does not address the serious constitutional infirmities found in the Reid Bill.
CONCLUSION: This ill-conceived legislation has many provisions that regulate different aspects of private health-insurance companies. Taken together, the combined force of these provisions raises serious constitutional questions. I think that these provisions are so intertwined with the rest of the legislation that it is difficult to see how the entire statute could survive if one of its components is defective to its core. How courts will deal with these difficult issues is of course not known, but rate-regulation cases normally attract a higher level of scrutiny than, say, land-use decisions.
There is, moreover, no quick fix that will eliminate the Reid Bill's major constitutional defects. It would, of course, be a catastrophe if the Congress sought to put this program into place before its constitutionality were tested. Most ratemaking challenges are done on the strength of the record, and I see no reason why a court would let a health-insurance company be driven into bankruptcy before it could present its case that the mixture of regulations and subsidies makes it impossible to earn a reasonable return on its capital. At the very least, therefore, there are massive problems of delayed implementation that will plague any health-care legislation from the date of its passage. I should add that the many broad delegations to key administrative officials will themselves give rise to major delays and additional challenges on statutory or constitutional grounds.
The health of the American people should not be held hostage to such unwise legislation. The Senate should reject the Reid Bill because of the unsustainability of the statutory scheme regulating health-insurance markets. But there is also little doubt that its central arrangements are unconstitutional, and will face serious legal challenge for years to come. Rather than embarking on a fundamentally flawed course of action, sure to spark litigation, the Senate should start over with other reforms that go in the opposite direction: simplify the system so that market forces can increase both quality and access in ways that no system of government mandates can hope to do. Deregulation is a word that has been forgotten in the current debate. It should be returned to center stage.
Richard A. Epstein is the James Parker Hall Distinguished Service Professor of Law at the University of Chicago, the Peter and Kirstin Bedford Senior Fellow at the Hoover Institution, a visiting professor at the NYU Law School, and a visiting scholar at the Manhattan Institute.
I wish to thank Paula Stannard for her prompt, gracious and expert assistance in guiding my way through the statutory quick sands. Any remaining errors are, of course, entirely my own.
1. H.R. 3962, The Affordable Health Care for American Act, passed the House this past November 7, 2009. A second bill, S. 1796, the America's Health Future Act of 2009, was reported in the Senate from the Senate Finance Committee on October 19, 2009. A third variant, S. 1679, the Affordable Health Choices Act, was reported in the Senate from the Senate Health, Education, Labor and Pensions Committee on September 17, 2009.
2. Congressional Budget Office, "Treatment of Proposals to Regulate Medical Loss Ratios," December 14, 2009. "A Medical loss ratio, or MLR, is the proportion of premium dollars that an insurer spends on health care: it is commonly calculated as the amount of claims incurred plus changes in reserves as a fraction of premiums earned." Id.
3. If we were to assume that the health insurance company gets, at most, a competitive return in good years--an assumption more generous than what the Reid Bill provides--it still gets less than a competitive return in poor years. The situation becomes like a coin flip, in which the regulator wins with "heads" and the regulated health-insurance company loses with "tails." Over the long run, the firm is necessarily deprived of a competitive rate of return except in the wildly improbable scenario that the firm earns exactly the competitive rate of return in all years.
4. See, e.g., Richard A. Epstein & David A. Hyman, Controlling the Costs of Medical Care: A Dose of Deregulation, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1158547.