Why Managed Competition Requires Extensive Regulation

Managed competition must involve intelligent, active, collective purchasing agents contracting with health care plans on behalf of a large group of subscribers and continuously structuring and adjusting the market to overcome attempts to avoid price competition.... It takes more than mere passive administration of inflexible rules to make this market work.121

"To prevent a disastrous drop in quality, regulation is required."

For the reasons given above, managed competition " and, indeed, any plan that combines community rating with competition " creates two inherent problems: (1) biased risk selection and (2) deteriorating quality. Biased risk selection is a problem in any insurance market that has artificial prices. No applicant is a "good" or a "bad" risk independent of a premium, and if the market works well and risk is priced accurately no applicant is more desirable than any other.122 It is only artificial premiums that create good and bad risks, and where they exist insurers naturally will pursue the former and avoid the latter. To keep that from happening, the managers or sponsors must regulate actively. As Enthoven explains:
In the Jackson Hole proposal, sponsors are the final arbiters of risk selection. ...[This may require] periodic face-to-face meetings with the marketing directors of all participating health plans, with the sponsor serving as honest broker. If Plan A is skimming, that hurts the other plans. The sponsor should lead a discussion on how risk selection can be defined, measured, and compensated for. This is an ongoing process, not a single event.123

Jackson Hole proponents have devoted almost all of their attention to the problem of biased risk selection and very little attention to the problem of deterioration in quality.124 Nonetheless, to deal with both problems, proponents usually advocate limited open enrollment periods, although this feature is absent from at least three of the managed competition bills in Congress. In addition, proponents invariably propose a complex government bureaucracy designed to (a) tightly regulate the content of health insurance policies and require everyone to have the same basic benefits package, (b) prevent insurers from offering higher deductibles or features likely to attract healthier subscribers and (c) redistribute funds from profitable to unprofitable insurers.125

Important Regulatory Bureaucracy: Health Alliances.

"Is a regulatory body, or health alliance, the alternative to 13 trillion pages of regulations?"

The Cooper alliance would also be mandatory,127 but would have very few powers. Chafee’s alliances would be neither mandatory nor monopolistic, and the Nickles plan has no alliances. In Florida’s new reform plan, the alliances are voluntary and a similar approach has been taken in California. And some powerful participants in the health care debate in Congress would like to strike alliances altogether from the Clinton bill. For example, Senate Republican leader Bob Dole of Kansas has said that alliances have to go, and Representative Pete Stark, the California Democrat who heads the House Ways and Means Subcommittee on Health, has declared that "There is not one chance in 100 that mandatory alliances will survive."128

Alain Enthoven, however, argues that voluntary alliances will not work.129 Managed competition requires regulation and there cannot be competing regulators. A recent editorial in the New York Times explains why:
Did the insurer recruit only in Scarsdale? Did the insurer answer phone calls from potential applicants in Harlem? Did the insurer tailor its benefits package so that AIDS patients would not apply?

Congress could, of course, enact 13 trillion pages of rules to stop these practices. But a more effective, less regulatory answer is to require most individuals or their employers to buy coverage through a cooperative, or an alliance.130

The only thing missing from this editorial is an appreciation of the fact that the alliance also will have to impose 13 trillion pages of rules, or the regulatory equivalent thereof.131

Important Restriction: Same Health Benefits for Everyone.

Important Restriction: Same Deductible and Copayments for Everyone.

"Risk adjustment payments would switch funds back and forth among the health plans."

Important Regulation: Risk-Adjustment Mechanism.

It might seem that the logical way to start constructing a risk-adjustment mechanism would be to tax or subsidize health plans based on the health of people at the time they joined a plan. Thus, sicker people would have a subsidy added to their premium payments and healthier people would have a tax deducted from theirs. Although enrollees would pay the same community-rated premium, health plans would receive a risk-adjusted premium. In theory, this would make the health plans indifferent between potential enrollees.

The problem with this approach is that it doesn’t solve the problem. Health economist Joseph Newhouse notes that in the Rand Health Insurance Experiment, 1 percent of the patients accounted for 28 percent of the total costs, but most of the high-cost patients could not have been identified in advance. In fact, Newhouse found that only 15 percent of the variation in health care costs among individuals could be predicted in advance, even when researchers had full knowledge of the patients’ demographic characteristics.138

"Only 15 percent of the variation in health costs among individuals can be predicted in advance."

This finding is consistent with other studies, which conclude that at most 20 percent of the variation in health expenditures for individuals can be predicted by such observable factors as health status and prior health expenditure.139 Therefore, no risk adjustment mechanism, even in principle, can compensate health plans for more than 20 percent of the potential adverse selection.

If adjustments cannot solve the problem based on prior knowledge of patients, the only alternative is to base them on past knowledge, the experiences of patients after they enroll.140 In other words, we could wait and see who gets sick before we start shifting money among insurers. For example, consider again the cost-of-care line in Figure IV. If the net amount insurers received for each applicant were based on this line rather than on the artificial premium line, then insurers would have no reason to prefer one patient over another. The problem is that if we reimburse health plans for what they spend, we are merely replicating the cost-plus system of health care finance that helped to create the crisis we now face.141 On the other hand, if we pay health plans based not on actual costs but on fixed fees determined by the patient’s diagnosis, we would have all of the problems we have in Medicare’s current system of hospital reimbursement.142

"As a result, a risk-adjustment mechanism can’t solve the problems."

A perfect risk-adjustment mechanism " which doesn’t exist " would eliminate the problem of competitive pressures to underprovide care for the sick and overprovide it to the healthy. It would do so by assuring each health plan exactly the right payment for each patient. But such a system would be totally based on regulation, not competition - thus undercutting the entire Jackson Hole rationale for reform. In the language of the New York Times editorial quoted above, a perfect risk-adjustment mechanism is supposed to solve problems that would otherwise require 13 trillion pages of regulations. Yet in order to make the mechanism work, we would probably need even more regulations. Put another way, risk-adjustment mechanisms don’t remove the need for regulation, they simply rename it.