Rules of the Road: How an Insurance Exchange Can Pool Risk and Protect Enrollees
March 31, 2009
- Many health reform proposals would establish an insurance “exchange” to provide coverage options that are affordable, comprehensive, and easy to compare. But unless an exchange has certain features, it will not function properly.
- Legislation establishing an exchange will need features which assure that: 1) insurance plans compete on the basis of price and quality, not on the basis of which insurers are best at attracting healthier enrollees and deterring sicker ones; 2) less-healthy individuals are not charged higher premiums for the same coverage simply because they end up in plans that disproportionately enroll less-healthy people; 3) all enrollees get at least a basic level of comprehensive benefits; and 4) consumers are able to compare plans and make informed decisions.
- To accomplish this, the legislation will need to require insurers to offer at least a minimum level of benefits. It also must limit the number of different benefit designs available in order to rule out benefit packages crafted to attract the healthy and deter the sick. And a reasonable limit will be needed on the number of plans an exchange offers in order to help consumers make intelligent choices.
- To prevent less-healthy people from being charged more for the same coverage because they are enrolled in a plan that disproportionately enrolls sicker individuals, the exchange should require each insurer to offer the full range of different benefit designs the exchange provides and to treat enrollees in all of its plans as one risk pool, as Massachusetts does.
- Simply requiring that all plans meet an “actuarial value” standard will not be sufficient to address these matters.
Several leading proposals for major health reform include a new entity — sometimes called an “exchange” — that would offer a choice of health insurance plans to individuals and, if designed well, provide insurance options that are affordable, comprehensive, and easy for consumers to compare.
The proposals on the table offer several elements that are crucial to making a new exchange work effectively. However, they do not spell out all of the steps needed to ensure that the exchange meets the goals of protecting vulnerable enrollees, simplifying the choice of plans so consumers can make intelligent decisions, and promoting competition among insurers based on the cost and the quality of their products — not their ability to maximize profits by attracting healthy, less costly enrollees and avoiding sicker, more costly people. It also is essential for an exchange to limit the extent to which healthy and sick people separate into different insurance plans. Such separation —known as “adverse selection” — can cause plans that attract less-healthy enrollees to become increasingly unaffordable over time.
To ensure that an exchange provides affordable, comprehensive coverage to all enrollees and enables people to make informed choices, it will need four key components:
- minimum standards for the benefit packages offered, including limits on beneficiaries’ out-of-pocket costs;
- limits on the variation in benefit designs that the insurers in an exchange can offer, so benefits cannot be set up to deter less-healthy individuals and attract only those in good health and so consumers can more readily compare plans based on price and quality;
- a limit on the number of different choices of benefit-package design, so individuals can make an intelligent decision about coverage, rather than being overwhelmed by the number of choices they face; and
- a requirement for each insurer to provide the full range of benefit design choices and to set premiums for all its plan options based on a single group that includes all of the people it enrolls within the exchange. Without such a requirement, which the Massachusetts health- reform plan includes, sicker individuals are likely to end up paying higher premiums than healthier people for the same coverage, rather than being charged more only for any additional coverage they purchase.
A strong exchange can greatly reduce the problems many people face today when they must obtain coverage on their own, without the help of an employer. An exchange can ensure access to coverage for people with health problems, simplify the decision-making process for consumers, and promote competition between insurers that is based on price and quality and will lead to better-value benefits. Several recent health-reform proposals take steps to protect people with medical conditions and reduce the incentive for insurers to avoid the sick in order to increase profits. For instance, many proposals include rules that would require health insurers to accept all applicants and avoid considering an individual’s health status when setting plan premiums. In addition, many proposals include “risk adjustment,” which would provide higher payments to insurers whose plans disproportionately attract less-healthy enrollees who cost more to cover. Premium subsidies and other elements, including total out-of-pocket limits expressed as a percentage of family income, would help ensure affordability.
But without additional steps, an exchange would fail to produce the improvements that consumers need and would likely encounter serious problems over time. Without properly structured rules for benefits design, many insurers participating in the exchange likely will create products designed to deter sicker enrollees who are more expensive to cover. Plans also may vary dramatically from one another, resulting in widespread confusion for beneficiaries and increasing the chances that people will unknowingly select coverage that later turns out to be inadequate.
Some have called for using an actuarial-value standard as part of health care reform. This would involve calculating a dollar amount, or percentage, equal to the share of health spending of a typical group of enrollees that a health plan covers. The amount or percentage is known as the “actuarial value.” All plans would then need to offer a benefit package estimated to be worth at least that value. But, an actuarial-value standard alone would be woefully insufficient. It would not equip an exchange with the tools necessary to meet the goal of providing coverage that is affordable, comprehensive and easy to compare, and it would do little to prevent insurers from designing plans fashioned to attract healthy individuals and deter less-healthy people. Two plans with the same actuarial value could have very different coverage levels and out-of-pocket costs, with one plan providing less coverage for certain treatments that sicker people are more likely to need (such as extensive hospital stays or certain chemotherapy drugs) while offering enhanced coverage for services attractive to healthier individuals (such as lower premiums or discounts on membership in a health club). The experience with private plans in the Medicare Advantage program, in which insurers have been able to develop and market plans designed to attract healthier beneficiaries and discourage sicker ones, illustrates the risk.
Furthermore, an actuarial standard would not guarantee that plans offer comprehensive coverage; plans would simply need to provide a benefit package that met the dollar-value standard. A plan could offer a package fashioned to attract people expecting not to be sick by skimping on some basic coverage. If an enrollee then became seriously ill during the year, the individual could be underinsured or even uninsured for needed medical care.
- With a minimum standard for benefits — e.g., a standard that assured that plans provided coverage for a comprehensive set of necessary services, such as physician visits, laboratory tests, inpatient hospital care, and prescription drugs, as well as basic standards for the scope of such coverage — individuals would not have to worry about choosing the “wrong” insurance. No matter what health problems an individual might face during the year, he or she could get the needed care at an affordable price.
- In addition, with appropriate limits on the degree of variation in different benefit designs — a basic feature of the Medigap market — insurers would not be able to easily create insurance products calibrated to deter the sick and attract the healthy, and consumers would be able to compare plans on price and quality and thereby make intelligent decisions.
- An appropriate limit on the number of different plan choices in an exchange also is crucial — to enable consumers to choose among plans without becoming overwhelmed.
- Finally, a requirement that insurers in an exchange offer a full range of plans and set premiums based on a single pool of all the people in their various plans — as is done in Massachusetts — would pool healthier and sicker people together so that people aren’t charged higher premiums just because they are enrolled in a plan that disproportionately serves less-healthy individuals.
By promoting competition based on price and quality, these four steps would hold down the costs in the insurance exchange and help make it as efficient as possible, while making sure that it serves individuals who need to obtain coverage through it.
Health Insurance “Exchanges”
The existing individual insurance market does not pool risk well. Many people, particularly those with health problems, either cannot get coverage or must pay very high prices for it.  In addition, individuals must sort through an often-bewildering array of coverage options and try to determine which one makes most sense for them. Small businesses face similar challenges in many states. 
Large employer-based insurance plans, by contrast, usually do a good job of spreading costs across a diverse group of people, both the healthy and the sick, while incurring lower administrative costs. Large companies often make sure that plan options are adequate and supply their workers with information about the different choices. A new exchange could bring these advantages of large-group plans to small firms and individuals seeking insurance on their own.
Recent comprehensive health-reform plans differ on the details of how an exchange would work and the roles it would play in a reformed health care system. But these proposals also have many similarities. Two plans — one outlined by President Obama during the campaign and another proposed recently by Senate Finance Committee Chairman Max Baucus (D-MT) — both propose retaining and strengthening the employer-based insurance system and existing public programs for low-income people. To reach individuals and businesses not covered by those efforts, both plans propose a new national health-insurance exchange that would offer a choice of coverage options from private insurers as well as a public plan from the government.  A third proposal, from Senators Ron Wyden (D-OR) and Bob Bennett (R-UT), would largely replace employer-based insurance and public programs with new exchange-like purchasing pools at the state level, which would provide a choice of at least two private plans. 
In all three of these proposals, as well as in other recent health reform plans, exchanges would give people a choice of various plan options and provide information about the plans’ benefits and costs. The proposals also would give the exchanges various other duties, such as setting standards for benefits, regulating and approving private insurance plans, facilitating enrollment, and administering a new system of subsidies for people with low and moderate incomes.
This vision of an exchange is consistent with the concept of “managed competition,” developed by economist Alain Enthoven and others, under which multiple health insurers would vie against each other for enrollment, striving to offer better-value products while playing by rules that ensure fairness and protect consumers. As Enthoven has explained, essential to “managed competition” is the idea that an active “sponsor” (such as an exchange) would engage in “continuously structuring and adjusting the market” to ensure competition that is based on price and quality. This approach aims to restrain health spending while improving the quality of care provided.
To make affordable coverage available within an exchange, the Obama, Baucus, and Wyden-Bennett plans would require insurers to accept all applicants and would bar insurers from considering health status when setting plan premiums. The Baucus and Wyden plans also propose a mechanism, known as risk adjustment, to provide higher payments to insurers whose plans tend to attract sicker-than-average enrollees, which is intended to reduce incentives for insurance companies to avoid the sick in order to increase their profits.  The Baucus and Wyden plans also include a mandate for individuals to purchase insurance, which ensures that healthier people purchase coverage along with less-healthy ones.
These provisions are important and desirable. But, on their own, they would be inadequate. Additional steps are needed to ensure that plans in an exchange actually compete on the basis of price and quality, to help consumers make intelligent choices, and to prevent “adverse selection” from separating healthy and less-healthy individuals into different insurance plans, with the plans that disproportionately enroll those with higher costs having to charge much higher premiums.
An Effective Exchange Can Reduce Adverse Selection
A major reason to create a strong exchange is to limit “adverse selection” within it. Any time that individuals are offered a choice of different health-insurance options, there is a risk of adverse selection. Sicker people tend to pick more-comprehensive plan options because they expect to use more health-care services, while healthier people tend to enroll in other plans that are less comprehensive but also less expensive. Sicker people are costlier to cover, so if they concentrate in more-comprehensive plans, premiums for those options would rise and become increasingly unaffordable over time. As premiums rise, the healthier individuals who initially enrolled in the more-comprehensive plans would likely abandon them for other, cheaper coverage, leaving behind an even more concentrated group of sicker people and further driving up premium costs.
The Impact of Adverse Selection on Exchanges
Without protections, adverse selection would have a negative impact on an exchange. Of most concern, sicker people would face unnecessarily high costs for more-comprehensive benefits, possibly to the point where that coverage became unaffordable or unavailable. This is because adverse selection, by segmenting healthier and sicker people into different insurance plans, can drive up the costs of more-comprehensive plans to levels well above the value of the additional benefits they provide.
Adverse selection also diminishes the sort of competition among insurers that can lead to improved benefits and lower costs for all enrollees in an exchange. If insurers can profit relatively easily from avoiding high-cost enrollees or attracting healthier ones, some are likely to use benefits design and other features to “cherry pick” a pool of customers that will be most profitable. This would cause insurers to compete less by restraining costs and improving the quality of their products and services and more by attempting to attract a healthier clientele. The result would be a loss to all participants in the exchange — both the healthy and the sick. As Alain Enthoven has written, rather than allowing cherry picking of customers, the goal of an exchange or similar mechanism should be “to reward with more subscribers and revenue those health plans that do the best job of improving quality, cutting cost, and satisfying patients… The rules of competition must be designed and administered so as not to reward health plans for selecting good risks, segmenting markets, or otherwise defeating the goals of managed competition.”a
A series of events in the Federal Employees Health Benefits Program (FEHBP), which covers federal workers and retirees, provides an example of adverse selection. The program offers multiple private insurance plans, and the federal government (as the employer) subsidizes the cost of the coverage. In the 1980s and 1990s, premiums for certain plans offered through FEHBP rose because higher-cost people were disproportionately enrolled in them. For example, Blue Cross and Blue Shield offered a “high-option” plan and a “low-option” plan, named to reflect the comprehensiveness of their benefits. As the high-option plan increasingly attracted sicker enrollees, its annual premium cost grew to be $2,800 higher than the premium for the low-option plan in 1994, even though the difference in the actuarial value of the two plans’ benefits was only about $80. This occurred because of the high costs the high-option plan incurred while serving the less-healthy group of beneficiaries enrolled in it. Eventually, the costs grew so high that Blue Cross stopped offering the high-option plan altogether in 2002.b
a Alain C. Enthoven, “The History and Principles of Managed Competition,” Health Affairs Supplement 1993.
b The Federal Employees Health Benefits Program: What Lessons Can It Offer Policymakers?” National Health Policy Forum Issue Brief, March 12, 1998. FEHPB does not use risk adjustment, which would have lessened, but not eliminated, adverse selection in this case by providing higher payments to plans that enrolled higher-cost people. See also Leonard Burman, “Medical Savings Accounts and Adverse Selection,” Urban Institute, 1997 and Harry P. Cain II, “Moving Medicare to the FEHB Model, Or How to Make an Elephant Fly,” Health Affairs, July/August 1999.
An insurance exchange should be structured to minimize adverse selection by grouping together people with different levels of medical expenses and health risk. Effective pooling makes it more likely that insurance plans and their costs will be stable over time, and it helps insurers to predict the average medical costs of the group they are covering from one year to the next.  As the American Academy of Actuaries states, “the pooling of risk is fundamental to all types of insurance.” 
Rules that promote good risk-pooling also diminish the incentives for insurers to structure their plans in ways that encourage enrollment by healthier people or discourage sicker people from signing up, a practice known as “cherry picking.” Most importantly, effective pooling protects people in poorer health from having to bear unaffordable medical costs by spreading their costs across a broad group that also includes healthier people. (See the box on page 6.)
Healthier people, too, would benefit. If private insurers compete primarily on the basis of who can most effectively screen out higher-cost enrollees, then healthy people would miss out on the better-value benefits that would result if insurers competed based on cost and quality. 
In addition, healthy people may not always remain healthy. Chronic medical conditions are exceedingly common and become more common with age. Either as part of growing older, or because of an unexpected illness, people who are healthy one day may not be healthy the next and may suddenly need access to more comprehensive benefits. In general, effective pooling within a new insurance exchange would provide access to reasonably priced, comprehensive health coverage for all enrollees. 
The Benefits and Limits of Risk Adjustment
“Risk adjustment” is an essential step in reducing adverse selection in an exchange. But, by itself, it will not be sufficient to prevent insurers from using benefits design and other features to deter enrollees who are in poorer health. As the Congressional Budget Office has explained, existing risk-adjustment systems “tend to overpredict the costs of beneficiaries who end up with low health care spending and to underpredict the costs of those who end up with high health spending.” In other words, risk adjustment is very difficult to do with complete accuracy. When done well, it succeeds in compensating for some — but not all — of the differences in health costs between healthier and less-healthy beneficiaries. As the CBO has warned, the inability of current risk-adjustment systems to fully adjust for differences in health care costs “could cause premiums for enrollees in plans that attract higher-cost beneficiaries to rise substantially over time.”  This is why risk adjustment needs to be combined with the measures outlined here.
Four Steps to Creating an Effective Exchange
For an insurance exchange to fulfill its main goals — promoting effective risk-pooling and facilitating informed decision-making by beneficiaries — the following four steps are essential.
1. Require a Minimum Level of Benefits
To make sure individuals obtain adequate coverage, the insurance plans available within an exchange should be required to meet minimum standards for benefits design. The standards should define a comprehensive set of necessary services — such as physician visits, laboratory tests, inpatient hospital care, and prescription drugs — as well as standards for the scope of such coverage. The minimum standards should also include basic parameters for cost-sharing, such as a maximum deductible and a cap on what individuals have to spend out-of-pocket each year.
The standards for benefits design need not be overly detailed or prescriptive. They should not, for example, specify the treatment protocols that would be covered for a particular illness. However, the standards must be more rigorous than merely requiring benefits designs to have a minimum actuarial value. Relying on an actuarial-value standard would be an inadequate way to assure that beneficiaries have adequate coverage.
Assigning an actuarial value to a plan involves calculating a dollar amount (or percentage) that reflects how much of the medical costs of a typical group of people the plan is estimated to cover; an actuarial-value standard is included in some plans, such as the Wyden-Bennett plan, as a way to set a minimum standard for the adequacy of benefits. But, by itself, such a standard is insufficient to assure access to a minimum level of comprehensive coverage. A wide variety of benefit designs can meet an actuarial value standard. A plan that limits coverage of certain services used primarily by more costly enrollees, such as hospital stays, can still have the same overall actuarial value as a plan that provides more adequate hospital coverage, because the plan limiting hospital coverage can substitute other benefits for it such as lower premiums or coverage for services more attractive to healthier people, such as discounts on membership in a gym. As a result, a plan could meet an actuarial standard while leaving someone who has a serious illness or condition with inadequate coverage for many necessary services.
Setting more specific minimum benefit standards would help to guarantee consumers adequate coverage regardless of which plan they select. No enrollee would have to worry about making the “wrong” choice and ending up with a plan that provides skimpy benefits or imposes unaffordable out-of-pocket costs for a medical condition that the individual develops during the year.
2. Limit Variation in Benefits Design
Even with minimum standards for benefits and cost-sharing, insurers operating in an exchange still could offer widely varying benefit designs. And since minimum benefit standards likely would not be overly prescriptive, plans could still design their benefit packages in ways calculated to attract healthy people and deter enrollment by sicker ones.
For example, insurers might try to avoid people with expensive health costs by sharply limiting coverage of mental-health services or maternity care. Or they might provide an annual physical at no charge, which would likely attract healthier people, while charging high out-of-pocket costs to enrollees who must go to the doctor more frequently or who need lengthy hospital stays. Accordingly, an exchange should limit the extent of such differences to meet three important goals — to help prevent adverse selection, to better assure that plans compete based on price and quality rather than on who is best at cherry picking, and to make it easier for consumers to compare plans and make intelligent decisions.
Wide variation in benefits packages has been an ongoing problem in Medicare Advantage, an alternative to the traditional Medicare program in which the federal government pays private insurers to provide health coverage to Medicare beneficiaries. Medicare Advantage plans have the flexibility to scale back existing Medicare benefits so long as the actuarial value of the overall package they provide is not less than the value of traditional Medicare benefits. This has resulted in an array of benefit designs that vary widely from one another and can have dramatically different impacts on beneficiaries’ out-of-pocket costs.  Some plans have imposed substantially higher cost-sharing for stays in the hospital and costly treatments like chemotherapy drugs than the charges traditional Medicare leaves, evidently as a way to discourage enrollment by sicker beneficiaries, while expanding certain benefits that may hold greater appeal for healthy individuals such as vision and dental care, preventive care, and membership in gyms or health clubs. 
The broad and bewildering variation in Medicare Advantage plans demonstrates that an actuarial-value standard is an inadequate method of holding variation in benefits design within reasonable bounds. Two benefit designs can have the same actuarial value but vary greatly in the coverage they offer.  The plans could have widely different maximum limits on how much an enrollee has to spend out-of-pocket each year, widely different amounts of cost-sharing for various services, and very different rules on whether or to what degree various treatments and services are covered — and still have the same actuarial value.
An actuarial value simply summarizes, in one figure and for a typical group of enrollees, a plan’s overall dollar value or the percentage of health spending that it covers.  Using an actuarial-value standard is merely a simplified method for experts to broadly compare the generosity levels of two or more health insurance plans. It is not a method for reducing adverse selection among plans, nor does it ensure that beneficiaries will be able to easily understand the trade-offs they would face under one benefit design versus another.
If the plans in an exchange need adhere only to an actuarial-value standard, insurers will have ample room to design benefits in ways that attract or discourage particular types of enrollees, resulting in adverse selection. Widely disparate benefit designs would likely spring up, with the potential to cause confusion among beneficiaries. A minimum standard that requires all plans to provide a basic level of comprehensive coverage would go part of the way toward preventing such problems, but would not be sufficient by itself. Adverse selection and beneficiary confusion still would result if there was broad variation among the different benefit designs available in an exchange. If one plan in the exchange were permitted to have a very different benefit and cost-sharing structure that would primarily attract healthier individuals and discourage enrollment by those in poorer health, the risk for adverse selection and beneficiary confusion would remain high. Insurers would have a strong incentive to promote, through marketing and other means, plans that are most likely to attract healthier enrollees and to downplay options that may be more attractive to the less healthy. The wider the benefit-design differences among the plans in an exchange, the greater the opportunity for insurers to take steps that contribute to segmenting the market, with healthier and less-healthy individuals separating into different insurance plans.
Instead of simply being required to meet an actuarial-value standard, even with some minimum benefit requirements, the plans available in an exchange should also meet some specific rules that moderate the degree of variation in benefit design. This is needed to keep adverse selection and beneficiary confusion to a minimum. An example of this approach is the Medigap market, in which private insurers sell supplemental coverage that fills gaps in the traditional Medicare benefits package. In the 1970s and 1980s, the Medigap market had few such rules, and the market was rife with beneficiary confusion and abusive marketing practices by insurers. The high prices charged for some Medigap policies were found to substantially outweigh the value of the benefits those policies provided. Congress responded with legislation that led to the creation of 12 standardized, prototype Medigap policies. Medigap policies generally must follow one of these 12 policy designs.
All of the Medigap policy designs, which are labeled A through L, fill in gaps in the traditional Medicare package. Medigap A policies, for example, cover the cost-sharing associated with Medicare’s hospital and outpatient coverage. Medigap B policies cover these items, plus the annual deductible for Medicare hospital coverage. Medigap C policies cover additional benefits such as medical emergencies during foreign travel.
This system has worked successfully in Medigap for more than a decade. While the Medigap market has its flaws, research has found that these reforms had a major positive impact on the market. The range of premiums narrowed significantly in the years after these rules were implemented, a sign that the market became more competitive and that consumers’ ability to assess the value of policies improved. Surveys of Medigap carriers, consumer groups, and state counseling programs show that beneficiary confusion about Medigap policies diminished significantly as a result of the reforms.
Another important way to limit the differences in benefits design is to rule out certain types of plans that would, by their nature, encourage adverse selection. High-deductible plans, especially those associated with tax-advantaged Health Savings Accounts (HSAs), pose a particular threat of this type. Healthier people, particularly those with higher incomes, are more likely to choose HSAs because they would be attracted by the lower premiums, tend to be more comfortable with the less-comprehensive benefits typically available from high-deductible plans, and can best take advantage of the generous tax benefits that HSAs provide to people in high tax brackets. Because of the strong likelihood that only healthier people would be willing to enroll in such plans, leading to adverse-selection problems, the new exchange should not allow high-deductible plans linked to HSAs.
Ensuring that benefits designs are adequately “standardized” — that is, that they offer a minimum level of comprehensive coverage and that the range and nature of the differences in plan design are appropriately restrained to reduce adverse selection and beneficiary confusion — would substantially improve competition in the exchange. If adverse selection is limited, plans are more likely to compete based on cost and quality, rather than on how successful they are at attracting healthier-than-average enrollees. Many leading health researchers and economists have noted the importance of structuring insurance benefits to foster such competition.
3. Limit the Number of Different Benefit Designs
While it would be helpful to provide a choice of different plan options to enrollees in a new insurance exchange, a bewildering array of choices would make the system difficult for beneficiaries to navigate. To simplify decision-making for individuals and enable decisions to be made more intelligently, there should be a limit on the degree of variation between plans and a limit on the number of different benefit designs any person must consider.
According to recent research in the field of behavioral economics, individuals are often overwhelmed when they are presented with too many options — to the point where they may not make a choice at all or may simply pick the option with the lowest short-term cost. For example, when presented with a choice of employer-sponsored retirement plans, individuals were less likely to sign upat all when too many investment options were offered. 
Under the Medicare prescription drug benefit, which presents several dozen different plan options in many parts of the country, beneficiaries have complained about confusion and difficulty making a decision. In addition, a recent study by leading M.I.T. health economist Jonathan Gruber shows that beneficiaries generally are not choosing prescription drug plans that provide them with the lowest possible premium and out-of-pocket costs for the medications they use, possibly because it is so difficult to evaluate such a wide range of coverage options. 
A limit on the number of plan options in an exchange, along with a minimum benefit requirement and limits on the variation in benefit design, would make individuals more likely to compare the cost and quality of various plans. Information could be presented in a consistent way, reducing the time and effort needed to examine different options and increasing individuals’ ability to choose high-value plans. That, in turn, would aid competition and efficiency in the exchange. 
4. Require Each Insurer to Have One Risk Pool
If the new exchange includes plans with varying levels of comprehensive benefits, people will likely base their choice of plan in part on their health status. Sicker people generally will be more willing to pay higher premiums to get additional coverage, and they may naturally sort themselves into the most comprehensive plans. Therefore, to prevent adverse selection, each insurer operating within the exchange should be required to offer each of the allowable benefit designs and to treat all enrollees in all of its plans as a single group when setting premiums. Insurers should not be able to price each plan based on the population that selects it. This would ensure that people in the most comprehensive coverage option an insurer offers would be grouped with the presumably healthier (and thus less costly) enrollees in the least comprehensive plans. No insurer would be able to offer only the type of plan most likely to attract healthier-than-average people, and sicker people would not have to bear the full brunt of their medical costs.
Massachusetts has adopted this strategy in the Commonwealth Choice program it created to offer multiple, private options to people who lack other sources of coverage. People who want lower premiums can choose plans in the “Bronze” tier, while people who want more generous coverage and are willing to pay more for it can choose a plan in the “Gold” tier.  Each insurance carrier must offer plans in all of the tiers and must treat all of its enrollees as one group, or risk pool, when pricing its products. Insurers are not permitted to price their Gold plans only on the basis of the people enrolling that plan, nor can they price Bronze plans based on the population selecting that option. The result of this is that, while people choosing an insurer’s Gold plan pay more in exchange for the additional coverage they are receiving, they are not paying extra charges simply because the Gold plan may be more likely to attract a sicker population. In effect, the health risks of all enrollees in an insurer’s products are being pooled together, and the costs spread among the group. 
Even with standardized benefits and unified risk pools, insurers likely would try to attract lower-risk enrollees. For example, they could target advertising campaigns to healthier or younger people by placing ads in publications or on websites that healthier-than-average people are more likely to see. Insurers could also design their network of health care providers with an eye toward attracting healthier beneficiaries or discouraging sicker ones, such as by excluding or limiting access to specialists that treat potentially expensive illnesses such as cancer or kidney disease. Finally, insurers could make their plans less appealing to sicker people by imposing more restrictive authorization or appeal procedures compared to other, competing options.  In a reformed health care system, a strong regulatory entity would have to take action in these areas, when necessary, in order to prevent adverse selection and protect beneficiaries.
As part of health reform, lawmakers are considering ways to offer individuals a choice of benefit options. While choice is often viewed as uniformly good, too much choice could have a negative impact. Without a structure to protect against adverse selection, less-healthy people could face unaffordable costs, and insurers would likely compete more on the basis of the types of enrollees they attract than on the cost and quality of their products. To prevent adverse selection and help consumers compare plans, the benefit designs within an exchange should be subject to the “rules of the road” described here.
These rules of the road have several important components, including minimum standards for benefits and limits on the amount of variation among different benefit designs. These steps would ensure that the exchange guarantees a basic level of coverage, while reducing the chance that adverse selection will occur and drive up costs over time.
To provide additional protection against adverse selection, exchanges also should require insurers to offer plans with all levels of benefits and to treat all of an insurer’s enrollees as one risk pool. Limiting the number of different benefit designs that individuals must consider also would simplify the process of selecting insurance and promote competition that would lead to better-value plans in the exchange.
Taken together, these steps would allow an insurance exchange to better achieve its primary goals of promoting risk-pooling, providing affordable, understandable benefit choices to healthier and sicker beneficiaries alike, and developing an insurance market where plans compete based on who offers the best quality coverage for the best price. Families, individuals, and small businesses looking for health insurance could then secure many of the advantages that are now typically available to large employers.
 Karen Pollitz, Richard Sorian, and Kathy Thomas, “How Accessible is Individual Health Insurance for Consumers in Less-Than-Perfect Health?” Kaiser Family Foundation, June 2001.
 Leonard Burman, “Tax Code and Health Insurance Coverage,” Testimony before the U.S. House Budget Committee, TaxPolicyCenter, October 18, 2007.
 Jonathan Gruber, “Covering the Uninsured in the U.S.,” National Bureau of Economic Research, Working Paper 13758, January 2008.
 Barack Obama and Joe Biden, “Plan to Lower Health Care Costs and Ensure Affordable, Accessible Health Coverage for All,” http://www.barackobama.com/pdf/issues/HealthCareFullPlan.pdf , accessed March 11, 2009.
 Max Baucus, “Call to Action: Health Reform 2009,” November 12, 2008, http://finance.senate.gov/healthreform2009/finalwhitepaper.pdf .
 The Baucus plan, which is more detailed than the Obama campaign document, suggests allowing private plans to be offered on a national, regional, statewide, or local basis.
 Healthy Americans Act, S. 391, introduced February 5, 2009. For an analysis of an earlier version of the Wyden-Bennett plan, see Edwin Park, “An Examination of the Wyden-Bennett Health Reform Plan, Key Issues in a New Approach to Universal Coverage,” Center on Budget and Policy Priorities, September 24, 2008.
 Cathy Schoen, Karen Davis, and Sara R. Collins, “Building Blocks for Reform: Achieving Universal Coverage With Private and Public Group Health Insurance,” Health Affairs (May/June 2008) pp. 646-657.
 Alain C. Enthoven, “The History and Principles of Managed Competition,” Health Affairs Supplement 1993.
 Risk adjustment is not specifically mentioned in the Obama campaign plan, but the Administration would likely support such a provision.
 Congressional Budget Office, “Key Issues in Analyzing Major Health Insurance Proposals,” December 2008. p. 78. See also American Academy of Actuaries, “Wading through Medical Insurance Pools: A Primer,” September 2006.
 American Academy of Actuaries, “Wading through Medical Insurance Pools: A Primer,” September 2006.
 See, e.g., Linda J. Blumberg and Len M. Nichols, “First, Do No Harm: Developing Health Insurance Market Reform Packages,” Health Affairs, Fall 1996, pp. 35-53.
 Blumberg and Nichols, op cit.
 Congressional Budget Office, “Designing a Premium Support System for Medicare,” December 2006.
 Paul Precht, David Lipschutz, and Bonnie Burns, “Informed Choice: The Case for Standardizing and Simplifying Medicare Private Plans,” California Health Advocates and the Medicare Rights Center, September 2007.
 Medicare Payment Advisory Commission, “Report to the Congress: Benefit Design and Cost-Sharing in Medicare Advantage Plans,” December 2004. See also Brian Biles, Lauren Hersch Nicholas and Stuart Guterman, “Medicare Beneficiary Out-of-Pocket Costs: Are Medicare Advantage Plans a Better Deal?” The Commonwealth Fund, May 2006; Patricia Neuman, “Medicare Advantage: Key Issues and Implications for Beneficiaries,” Testimony before the House Budget Committee, Kaiser Family Foundation, June 28, 2007.
 See Park, op cit.
 “Study of the Administrative Costs and Actuarial Values of Small Health Plans,” Actuarial Research Corporation for the Small Business Administration Office of Advocacy, January 2003.
 Nora Super, “Medigap: Prevalence, Premiums, and Opportunities for Reform,” National Health Policy Forum Issue Brief No. 782, September 9, 2002.
 Super, op cit.
 Lauren A. McCormack, Peter D. Fox, Thomas Rice, Marcia L. Graham, “Medigap Reform Legislation of 1990: Have the Objectives Been Met?” Health Care Financing Review, September 22, 1996 and Super, op cit.
 Edwin Park and Robert Greenstein, “Latest Enrollment Data Still Fail to Dispel Concerns about Health Savings Accounts,” Center on Budget and Policy Priorities, Revised January 2006. See also Leonard Burman, “Tax Code and Health Insurance Coverage,” testimony before the U.S. House Budget Committee, Tax Policy Center, October 18, 2007.
 Jodi DiCenzo and Paul Fronstin, “Lessons from the Evolution of 401(k) Retirement Plans for Increased Consumerism in Health Care: An Application of Behavioral Research,” Employee Benefit Research Institute Issue Brief, August 2008.
 Jonathan Gruber, “Choosing a Medicare Part D Plan: Are Beneficiaries Choosing Low-Cost Plans?” The Henry J. Kaiser Family Foundation, March 2009.
 Elliot K. Wicks, “Restructuring Health Insurance Markets,” Health Management Associates, January 2009.
 Jon Kingsdale, “Report to the Massachusetts Legislature, Implementation of the Health Care Reform Law, Chapter 58 2006-2008,” The Massachusetts Health Insurance Connector Authority, October 2, 2008.
 The insurers also must include in their risk pools the people they enroll through the individual and small-group markets that operate outside of Commonwealth Choice.
 So far, there have not been reports of adverse selection problems in Commonwealth Choice, though the program is still new and relatively small. It launched in 2007 and covered just 20,150 members as of January 2009. Most individuals newly insured through the Massachusetts health reform plan have enrolled through employer-based plans or through public programs.
 Richard Kronick and Joy de Beyer, “Risk Adjustment is Not Enough: Strategies to Limit Risk Selection in the Medicare Program,” The Commonwealth Fund, May 1997.