Part I
Management Issues and Policies in Health Insurance Market Reforms
1
Restructuring Health Insurance Markets
ELLIOT K. WICKS
States and federal legislators are considering major policy changes to make health insurance more affordable and to expand health care coverage. Often this involves changes in the structure of health insurance markets, such as establishing risk pools or insurance exchanges for certain populations and altering the rules for selling coverage in the individual and small-group markets. This chapter examines structural changes that might be made to expand coverage, with special focus on six administrative issues that must be addressed: changes in rating rules, high-risk pools, standard benefit plans, reinsurance, Section 125 plans (Internal Revenue Code âcafeteria plansâ), and insurance exchanges.
The Purpose of Health Insurance
It is useful to begin by recalling the fundamental purpose of health insurance. People buy health insurance to ensure that the cost of paying for medical services will not be a barrier to receiving necessary care and to protect themselves against the financial catastrophe of incurring a very large medical bill that they could not afford to pay from their own resources. When they buy health insurance, people choose to incur a small, certain lossâthe insurance premiumâto protect themselves against the possibility of incurring a very large, unpredictable lossâthe cost of paying for very expensive medical care. In other words, people pool their risk, but this form of protection is sustainable only under very specific conditions: Insurance works only so long as most people in the pool incur expenses far below the amount they pay in premiums so there will be enough money to cover the few people who incur large losses.
The role of health insurers is to design and sell health insurance policies that provide financial protection, but the market does not always perform as intended, especially for people whose risk of needing health services is relatively high. If every person bought health insurance and had identical coverageâthe same covered services, the same limits, and the same cost-sharing structureâ the role of insurers would be very different and most of the problems related to the functioning of insurance markets would disappear. Such a system would be essentially a system of social insurance, like Medicare, where everyone is automatically covered, the benefits are essentially identical, and everyone pays the same premium rate. Of course, this is not a description of the current private health insurance system. Under competitive pressures in a voluntary market, where people may or may not buy coverage, insurers face strong incentives to develop many different benefit structures and coverage options to try to attract the most profitable business. They find it profitable to segment the market by risk: If other insurers are pooling high-risk and low-risk enrollees and charging them one rate, a competitor can gain business by pooling only low-risk enrollees and offering them a lower rate. For similar reasons, insurers have incentives to structure their benefit plans to be especially attractive to lower-risk people and to avoid the highest-risk individuals.
Of course, once one insurer adopts these tactics, others must follow to remain competitive. If this process remains unchecked by legal constraints, the result is extensive segmentation of the market by risk, where coverage becomes unaffordable for many higher-risk individuals. Of course, the proliferation of coverage options and the mechanisms insurers must employ to differentiate people on the basis of risk add to insurersâ administrative costs and create burdens for the people who buy coverage. Legislators and others who seek to restructure the insurance market hope to ameliorate some of the problems while still retaining the private insurance system.
Characteristics of a Well-Functioning Market
Next I compare a well-functioning market with the market operating today.
Efficiency
One desirable characteristic of an insurance market is that it would operate efficientlyâthat is, administrative costs, including transactions costs for consumers and providers, would be minimized. Costs associated with reviewing the characteristics of people applying for coverage, determining eligibility for coverage, and setting prices for categories of applicantsâreferred to as medical underwritingâwould be minimized, since resources used for medical underwriting have no benefits for consumers. Marketing costs would be a small portion of the total premium.
Reasonable Choice
A well-functioning market would offer people a choice of different insurance benefits packages and delivery systemsâhealth maintenance organizations (HMOs), preferred provider organizations (PPOs), indemnity plans, and so on. Benefit options would be sufficient to allow people to choose a plan that suited their preferences but not so numerous as to make choice confusing. Consumers would have objective information in a form that would make it easy for them to make intelligent choices as they compared coverage options and plans.
Useful Competition
A well-structured market would promote useful competition among insurers so they were motivated to minimize the cost of coverage and provide timely, efficient, and effective customer service. Competition would induce insurers to reduce not only administrative costs but also the underlying costs of medical expenses. No insurers would be so dominant that they could have control over the market. Insurers would receive a return on investment sufficient to induce them to continue in business, but profits would not be excessive.
Pooling Risk
Market rules would ensure pooling risks to spread them broadly and fairly and not result in high-risk people facing unaffordable premiums. No one who could afford to pay for average-priced insurance would be priced out of the market because of their higher-than-average risk. To the extent that market segmentation exists, it would not threaten the basic insurance principal of having low-risk people subsidize the medical expenses of high-risk people.
Comparing Reality to the Ideal
Comparing present insurance markets to the ideal shows that performance often falls short. Administrative costs are a relatively high proportion of premium costs, especially in the individual market but also in the small-group market. Marketing costs also tend to be high in these markets because each sale adds only a few enrollees. The number and variety of insurance benefit plan options are far more numerous than necessary to meet consumer needs and, along with the complexity of benefit structures, make informed choice difficult. Few people can go to a single source to see and compare all the options. Consumers bear the burdens of trying to keep track of bills and submit claims, worrying about coordination of benefits, and generally trying to ensure that they are getting the financial benefits to which they are entitled. Family or job status changes create a risk that people will fall between the cracks in insurance coverage or even be denied coverage entirely. In states that allow risk rating in the small-group and individual markets, competitive pressures force insurers to devote resources to identifying the relative risk posed by different applicants. In such circumstances, the rewards for being effective at risk segmentation may be as great or greater than those associated with controlling underlying medical costs. As a result, higher-risk people, especially in the individual market, may be priced out of the market, so the basic risk-pooling function of insurance is impaired.
In most states, a small number of insurers account for a very high proportion of the business in the individual and small-group marketsânot the picture of a classic competitive market. Market domination by only a few insurers has been common in many states for a number of years, but recently the degree of market concentration has increased greatly, reflecting, in part, a wave of mergers. For employment-based and individually purchased health insurance, WellPoint and UnitedHealth Group each hold 14 percent of shares of the national market, and Blue Cross plans control 32 percent of the overall market. Concentration is higher in many state and regional markets. According to one recent study, the two largest insurers have at least half the enrollment in forty states (out of forty-four for which data were available), and they hold at least three-quarters of the enrollment in fifteen states.
Deficiencies have prompted some critics to call for wholesale reform, often arguing that only something like a single-payer system can fully address such deficiencies. But the opposition to such massive change is fierce, and such a system brings its own set of problems. Likely, in the near future we will continue to have a system composed of multiple insurers offering competing plans. If that is so, the policy question is, What are the options for streamlining the administrative structure of the present system and bringing its performance closer to the ideal?
Major Options for Structural Change
In this chapter I look at rate compression, high-risk pools, standardized benefit plans, public reinsurance, Section 125 plans, and insurance exchanges or purchasing pools. Although I discuss the policy justification for each approach, my emphasis is on administrative issues.
Rate Compression
In the absence of legal constraints, insurers segment people by risk, creating large premium differences between high-risk and low-risk people. Most states have placed some limits on risk rating in the small-group marketâalthough the limits vary from state to state^âand a few have done so in the individual market. Some states have required insurers to use a pure community rating in one or both of these markets, so that everyone pays the same rate for similar coverage regardless of health status or any other personal characteristics. In states with the least restrictive rate rules, the rate-variation ratio can be as much as 10:1 in the small-group market, and the highest-risk people in the individual market may be denied coverage entirely. The purpose of rate compression, of course, is to broaden the sharing of risk to make coverage more affordable for higher-risk groups and individuals. The problem with this solution is that rate compression, while making coverage less expensive for higher-risk people, makes it more expensive for lower-risk people. In the absence of a mandateâa requirement that everyone have coverageâthe higher premiums may cause some lower-risk individuals or groups to drop coverage and discourage others from acquiring it in the first place. If the exodus is too great, there will be too few people in the risk pool whose medical expenses are substantially less than their premiums. The average cost of medical claims will rise, forcing the insurer to raise the premium, which causes other people with below-average risk to leave the risk pool. The result is a spiral of adverse selection and a continually deteriorating risk pool.
Experience suggests that this phenomenon of adverse selection against the market as a whole is less of a problem in the small-group than in the individual market. It is much easier for an individual than for a group to assess when he or she is most likely to need expensive medical care and to buy coverage only when that need is imminent and to go without coverage at other times.
The policy challenge, then, is how far to go in compressing rates. States have chosen different paths, and their experiences can be a useful guide for those considering whether to move toward greater rate compression. In general, rate compression has seemed to work reasonably well in the small-group market in helping to make coverage more affordable for high-risk groups, but it has not increased the overall rate of coverage. In the individual market, the problems have been greater because of adverse selection against the market as a whole, created when low-risk people drop out.
From a purely administrative standpoint, rate compression does not pose significant new challenges for most states since they already have an administrative structure in place to enforce their current rating rules. The federal government does not now impose rating restrictions in either the small-group or the individual market, but even if this policy were to be changed, it is likely that the responsibility for enforcing the law would fall on the states. To enforce rate limits, states must be able to review individual insurersâ procedures and policies and to collect and analyze premium data to ensure that they are complying with the rating restrictions.
If a state adopts a pure community rating, allowing no variation based on any rating factors, the administrative task is much simpler than when insurers are permitted to use a variety of rating factors. For example, it is not unusual for states to allow rate differences based on some combination of health status, age, gender, geography, industry, health-related behaviors, and class of business, typically with different limits attached to each factor. When there are multiple rating factors and different limits on the extent to which insurers can vary rates based on each factor, both compliance with and enforcement of the law become very complicated and therefore more expensive.
One way to simplify the administrative burden and to make the rating restrictions more comprehensible, while still allowing any degree of variation in premium thought to be advisable, is to set an overall limit on the amount by which the rate can varied based on all factors in combination. For example, if a state decided that no high-risk group should have to pay more than four times the rate paid by the lowest-risk group, the state would stipulate that whatever rating factors were used, the maximum rate variation for all factors in combination could not exceed a ratio of 4:1. Besides being simpler, this approach makes less important the choice about which particular rating factors insurers are permitted to use. Under this approach, state regulators would find it easier to determine whether an insurer is complying with this requirement rather than with one that allows insurers to vary rates based on multiple factors, each with its own limit. Under the latter circumstance, it is sometimes difficult to determine even what total rate variation is the theoretical maximum. Taking this approach also allows states to move gradually to greater rate compressionâfor example, starting with a ratio of 5:1 and moving to 3:1 by reducing the ratio by 0.5 each year for four yearsâwhich will help reduce sticker shock for low-risk people and may make them less likely to forgo insurance.
High-Risk Pools
Rate compression is a way to make coverage more affordable for high-risk people, but it is a tool most commonly used in the small-group market, since, if employed in the individual market, it is more likely to produce adverse selection against the market as a whole. The tool more frequently used in the individual market is a high-risk pool. Like rate compression, it is a way of subsidizing high-risk people. The idea is simple: A mechanism is used to identify people whose risk profile is so high that they could almost surely not afford to pay a premium that would fully cover their risk. These people are then offered coverage with a specified benefit package through a special risk pool composed of just high-risk people. They pay a premium that is higher than what ânormal-riskâ people would pay for such coverage but still not sufficient to cover their full expected medical claims. The shortfall represents a subsidy that has to be funded from some source, typically some kind of surcharge on insurers but sometimes, and preferably, from broader-based revenue sources. Implementing a high-risk pool raises a number of policy and administrative issues.
Eligibility
An obvious question is how to determine who will be eligible, or, to put it another way, how will high-risk people be identified? Most states have given insurers great discretion in deciding who will be denied individual coverage, allowing them to deny coverage to anyone they judge to be especially high risk. In California, an approach is being considered that would leave insurers no discretion: the state would develop a questionnaire that all insurers would administer to applicants, and insurers would be permitted to deny coverage only to those people who âfailedâ the test. Everyone who âpassedâ would have guaranteed coverage. This approach has the advantage of predictably limiting the number of people who will fall into the high-risk category, as well as providing uniform treatment for applicants regardless of the insurer to which they apply.
Some states require the individual to show that he or she has been turned down for coverage by more than one insurer. Denial of coverage does not by itself make one eligible for a high-risk pool in most states. States frequently also require that the person not have available any source of group coverage, including Consolidated Omnibus Budget Reconciliation Act (COBRA) coverage. In some instances a denial is defined to include having available only individual coverage that excludes some major medical conditions. Some states also allow a person to be eligible if the cost of the only coverage that is available exceeds a specified high dollar amount.
This approach involves many administrative issues, mostly related to verifying eligibility. Someone has to verify that the person has no other coverage available, has been denied coverage, or can find only coverage that excludes some condition or is too expensive. For some of the information, the only reasonable verification may be the applicantâs self-report, subject to penalty for false statements, though some states require written proof of having been denied coverage in the form of a letter from the insurer. If written proof is not required, some agency will need to audit a selection of...