Health Insurance Regulations

Wake Forest University: Health InsuranceMarket Reform Study

Objectives & Scope |  Reports |  Summary of Findings |  OtherResource


Objectives & Scope

This project evaluates the performance of health insurance market reforms. Its focus is on state reforms to the small group and individual markets, but it also includes some preliminary findings about the early impact of the federal Health Insurance Portability and Accountability Act (HIPAA).

Little is presently known about the details of how well these reforms have performed in achieving their various objectives and avoiding predicted harms. Other studies test for aggregate effects on enrollment and premiums using primarily quantitative measures. The objective of this project is to conduct a primarily qualitative evaluation that attempts to understand why these effects have occurred and how these reforms affect that inner workings of the insurance market.

The primary method is an intensive case study of seven states that have enacted varying reforms: Colorado, Florida, Iowa, New York, North Carolina, Ohio and Vermont. This multiple case study consists primarily of two rounds of structured, in-depth, open-ended interviews with over 100 insurance industry and regulatory sources. The study also includes an analysis of quantitative, documentary, and other secondary data, and a market test in which a potential purchaser contacted 144 agents.

The principal reforms studied are:
(1) guaranteed issue
(2) renewability and portability
(3) rating bands and community rating
(4) restrictions on underwriting practices such as risk selection and pre-existing condition exclusions
(5) reinsurance and risk adjustment
(6) public purchasing cooperatives

The results of this investigation are being disseminated through a series of case study reports that are available for browsing or downloading. These are intended to inform lawmakers, purchasers, insurers, agents, and the public policy community whether and how state and federal reforms have achieved their multiple purposes or caused any negative consequences, and how these reforms might be improved.

For additional information, contact
Mark A. Hall
Janice Lawlor


Introduction and Background Analysis

A. The Impetus for Reform

B. The Logic of Reform

  1. The Purpose of Reform
  2. Guaranteed Issue and Continuity of Coverage
  3. Rating Restrictions
  4. Reinsurance or Risk Adjustment
  5. Purchasing Cooperatives

C. The Dangers of Reform

This report provides background information and analysis about the content, logic, and purposes of insurance market reforms. This is intended as a more detailed introduction to these laws for those readers who previously have not had in-depth exposure or understanding. This analysis presents primarily the views of its author and does not constitute findings from the empirical project. For more extensive discussion, see these additional publications by Mark A. Hall:

A. The Impetus for Reform
Insurance market reforms were enacted across the country in the early 1990s in response to growing concerns that the market for individual and small groups was rapidly disintegrating. The portion of the population nationally covered by private insurance declined precipitously during the 1980s -- from 81% of the total population to 70-75% (depending on assumptions made about multiple coverage from different sources). This rate of decay is even more rapid for small groups and individuals.

The instability of the health insurance market is also seen in the rising costs for those who still retain coverage. During the late 1980s and early 1990s, employers' costs for private insurance increased 15-20% each year, much faster than the underlying rate of increase in overall health care spending. According to a survey by a benefits consultant of its clients, the amount employers paid for health care doubled between 1987 and 1992. Despite the tendency to make insurers the scapegoats for these increasing costs, spiraling premiums were not the result of insurance company price gouging. Large insurers consistently lose money or barely break even on their health insurance products and several health insurers went bankrupt over this time period. Even the once invincible Blues have been shaken by financial troubles.

Problems in the health insurance market first began to manifest themselves among individual subscribers who purchase insurance outside the workplace. This end of the market has always been plagued by fairly severe adverse selection. Younger, healthier individuals often decline to purchase insurance until they or a family member anticipate significant expenses, say, the birth of a child or the onset of a chronic disease. Therefore, those who purchase individually tend to be significantly more risky than those insured through employer groups. This raises the price to a level that many individuals cannot afford or are not willing to pay unless they know they are likely to use the benefits. Similar problems then began to spread to the small-group market: that portion composed of employers, but where the group size created the same problems of biased selection. At first, this was employers with 10 or fewer workers, but significant selection problems began to affect firms up to 25 and larger.

To counteract adverse selection and to offer more attractive prices, insurers have long used a number of practices known generically as medical underwriting that limit the availability of health insurance. First, they have rated applicants according to their individual age and health status, which can price insurance out of the reach of older or chronically ill people. Second, to discourage subscribers from enrolling only when they are ill, insurers have imposed preexisting illness provisions that offer insurance but do not cover for a defined period those health conditions that existed any time during the prior 6-12 months before enrollment. This makes health insurance much less attractive to subscribers who are sick or have sick family members. Still, many would choose to pay the premiums for a year or more in order to be covered thereafter, since most states make it difficult for insurers to refuse the renewal of health insurance once it is initially sold. Therefore, insurers were declining to cover individuals or groups of a particularly high risk or they excluded altogether a designated set of health problems, such as cancer or asthma, a practice known as "ridering out."

A second problem created by preexisting exclusion clauses is that employees who have lasted them out are then very reluctant to switch to another insurer and start all over. This can create a condition of "job-lock" in which mildly but chronically ill employees, or those with sick family members, are frozen into their present position by the fear of losing insurance benefits if they were to switch.

Some health insurers were also accused of "churning." This is the practice of offering deep discounts to new subscribers after extensive medical underwriting, based on their initial good health profiles, and then imposing extreme price increases of 50% to more than 100% after a year or two of coverage, or refusing to renew altogether. Frequently, these price increases or terminations were not based on any actual adverse claims history but instead reflected the "wearing off" of the predictive power of the initial screening. Medical underwriting tends to wear off by virtue of the statistical phenomenon known as "regression to the mean," according to which unusually good risks tend over time to become normal risks simply by the operation of the law of averages. Claims costs also rise over time simply because the initial preexisting condition periods expire.

In sum, in the late 1980s and early 1990s, we were witnessing what some have characterized as a not-so-gradual unraveling of the private market, beginning at its smallest end of individual subscribers and very small employer groups and progressing to medium-sized firms and larger. The task that state and federal lawmakers faced was how best to weave these frayed ends of the health insurance rope back together at the same time that they sought to create mechanisms that would restrain health care costs.

Out of these multiple concerns and objectives arose a barrage of state legislation attempting to shore up the market for private health insurance. Between 1991 and 1996, all but four states enacted significant legislation that affects the pricing or marketing of private health insurance. This was followed in 1996 by the federal law known as the Health Insurance Portability and Accountability Act (HIPAA), whose provisions are discussed below. The primary focus of this study is on state-level insurance reforms, of both the small-group and individual markets, but it also provides a preliminary look at the small-group portions of the federal HIPAA legislation.

B. The Logic of Reform
1. The Purpose of Reform

Health insurance reforms have multiple purposes, but they all tend to converge on this central aim: to create a marketplace dynamic that will both promote health insurance coverage and restrain health insurance costs. This broad objective has a number of subsidiary components, which include increasing insurance availability, increasing its affordability, reducing risk selection and medical underwriting, leveling the playing field among insurers of different types with different operating strategies, and stabilizing the market. Others view the purposes in more political terms, as a means to improve the functioning of the private market so as to lessen the threat of a complete government takeover, or in some instances as a desire by one part of the industry to force others to bail them out or to compete on their terms.

Thus, there is no single, cohesive view that characterizes all insurance reform efforts. Some reformers viewed the existing market structures and practices as being mostly acceptable and socially responsible, except for certain excesses that were giving the industry a bad name. This view tends to prevail among smaller, indemnity-based insurers. Other reformers, however, had much more sweeping goals. The more sweeping reformers felt that private insurers were dissipating too much of their competitive energies in socially unproductive or counterproductive efforts to screen, select and measure individual health risks. This set of reformers thought that the private health insurance market would serve its best social function if insurers would quit competing on the basis of risk selection and begin competing on the basis of risk management. This viewpoint is obviously more sympathetic to larger, managed care insurers.

According to this second viewpoint, risk selection has its strongest economic function where insurance behaves in the classic fashion of reducing risk by pooling large numbers of subscribers with similar risk profiles. And risk selection has its strongest social purpose when the risk in question is controllable, so that the cost of insurance motivates efforts to reduce the risk. These are some, but not the central, functions of health insurance. Health insurance pools serve not just to enhance statistical predictability of uncertain events through the law of large numbers. Instead, pooling through health insurance serves the function of spreading the known costs of predictable health risk across a heterogeneous pool of both healthy and sick people, for both routine and catastrophic expenses. Health insurance has thus become in large part a prepaid financing mechanism for predictable health-care expenses. Moreover, many of the genetic and environmental factors that influence health risk are not controllable by individuals, so it is seen as unfair and unproductive to vary insurance cost by risk status. Therefore, the more fundamental reformers believe that insurers serve their most useful social function by covering as many people as possible and by setting their prices to reflect their ability to control the underlying costs of treatment rather than their ability to accurately measure and project those costs. Reforms enacted with this mindset were intended, in part, to harness the energies that private insurers previously applied to assessing individual health status (medical underwriting) and redirect them toward managing the costs of treatment.

Insurance market reforms vary considerably in their details but they share four essential components: (1) guaranteed issue and continuity of coverage, (2) rating bands and community rating, (3) administered reinsurance or risk adjustment, and (4) purchasing cooperatives. These laws also vary in their scope: most states have so far enacted comprehensive reforms of the small-group market, usually defined as employers with 2-50 workers, but a significant number have extended these reforms to individual purchasers. A greater number cover a hybrid category known somewhat paradoxically as "one-life" groups. These are self-employed individuals who purchase insurance through their business. The following overview is based on the common features of this set of laws across all of the states.

2. Guaranteed Issue and Continuity of Coverage

The starting point of reform is to make sure that any willing purchaser has access to insurance and can retain that insurance through subsequent renewal periods. "Guaranteed issue," sometimes known as open enrollment, requires all insurers who participate in the small-group or individual markets to accept any applicant.1 An important distinction exists between states that require only designated policies to be issued on a guaranteed basis and those that require this of all policies. Prior to HIPAA, only 15 states had imposed the broader requirement. But since 1997, states have had to conform their laws to HIPAA, which requires all small-group policies to be guaranteed issue. However, HIPAA applies only to groups of 2-50 employees, and so many states retain the limited guaranteed-issue requirement for the self-employed or for individual insurance.

Where guaranteed issue applies only to some policies, it is usually to one or more standardized plans whose benefits are set by a government committee. Typically, these state-mandated plans will come in a basic, bare bones version, and one with a more standard or common benefits package, and versions of each will often exist for both indemnity and health maintenance organization (HMO) products.

Enabling any group to obtain insurance is coupled with a "whole group concept, which requires the employer to offer coverage to all individuals within a group. This prevents employers from angling for lower-cost policies by excluding sicker individuals in the group and minimizes the selection problems that result if healthier individuals are allowed to drop out of the risk pool and purchase individual insurance. The latter technique is sometimes called list billing, meaning that the insurer bills for the cost of individual coverage for a list of certain employees. This use of the term is not meant to be in contrast with composite billing, which captures a different distinction, namely, whether groups are billed based on average group characteristics or instead are billed in a way that identifies each employee's unique costs. That use of the term list billing occurs within the purchase of group insurance. We mean here to identify the use of list billing that allows the sale of individual insurance through payroll deduction. To the extent this attempts to avoid reform laws by employers paying for less-regulated individual insurance, this practice is often prohibited. State laws also typically set limits on the extent to which insurers can set minimum participation and contribution requirements, that is, the minimum number of employees who must purchase before the insurer is willing to sell, and the minimum amount the employer must contribute before the policy is considered to be subject to the small-group laws.

Another source of complication is determining which members of the group are legitimate employees. In order to keep employers from disguising sick friends and family members as employees, or from hiding healthy employees who are purchasing cheaper insurance individually, state laws allow, or sometimes require, the insurer to demand certain tax and payroll records.

These laws promote continuity of coverage in three ways. These aspects once varied considerably among the states, but are now mostly uniform for small groups as a consequence of HIPAA, although states are still permitted to impose requirements that are more demanding than the federal law. First, insurers are prohibited from refusing to renew insurance except for fraud, nonpayment or similar malfeasance. Insurance theory tells us that, once good and bad risks are pooled together randomly and remain together, and a fair premium is set, actuarial soundness does not require that insurers be able to drop the bad risks when claims begin to accrue. Moreover, if insurance does not remain available when the need for it arises, it is not attractive either to the sick or to the healthy.

The second aspect of continuity is to regulate the use of preexisting condition exclusion clauses. Insurers are prohibited from ridering out specific health conditions altogether. Under HIPAA, they are allowed for small-group insurance to place only an initial 12-month preexisting exclusion on any condition manifested within six months before the date of coverage. This recognizes that some form of preexisting exclusion is necessary in a market where the purchase of insurance remains voluntary, in order to counteract the tendency of subscribers to delay purchase until they are sick. The judgment being made here is that a 12-month exclusion period is sufficient rather than the two-year or longer periods that had come into use.

Third, these reforms address the problem of "job lock" that arises when employees are afraid to change jobs for fear of having to undergo an additional exclusion period. "Portability" or continuity of coverage is promoted by requiring that subscribers, once enrolled, be able to transfer coverage to a new insurer, either by changing jobs or changing insurers within the same workplace, without undergoing a new exclusion period, so long as the gap in coverage does not exceed two months.

Again, the logic is that of adverse selection. If insurance is being acquired in a setting that is demonstrably not driven by selection concerns -- such as a change of job -- risks should distribute themselves evenly and predictably and therefore no special protection against adverse selection is required. Insurers should be able to cover their risks simply by setting the initial premium appropriately. Also, easing the ability to switch is obviously critical to allowing insurers to compete on the basis of price and service quality.

3. Rating Restrictions

Guaranteed issue and continuity of coverage eliminate the worst effects of medical underwriting -- denial of coverage, job lock, and churning -- but, standing alone, they would increase price variations and fluctuations by forcing insurers to take on the most extreme risks and allowing them to price their policies accordingly. Therefore, the second component of the reforms is to construct a variety of rating restrictions that compress the degree of price variation among subscribers. HIPAA purposefully does not impose rate restrictions, leaving this to state law. States have adopted two types of restrictions: (1) limits on the amount insurers may increase the price for a specific subscriber over time; and (2) limits that compress the range of prices an insurer can charge across its entire block of business at any point in time.

The National Association of Insurance Commissioners' (NAIC) model limits year-to-year premium increases for any given group to 15% above the insurer's "trend." Trend is defined as the increase in the insurer's rates for new business. The concept is to allow marketwide cost increases that are driven by technology advances, inflation in the medical sector and the like, but to limit those increases that reflect group-specific health risk. (Trend is keyed to new business rates because this is where insurers are the most competitive.)

The second component of the rating reforms would prevent any insurer from varying its prices among subscribers at any point in time more than a defined amount above or below its midpoint for policies with similar benefits and "case characteristics."2 States vary considerably in these rating rules. There are three basic approaches: rating bands, adjusted community rating, and pure community rating. While each requires successively greater degrees of rate compression, the key distinguishing factor is the extent to which insurers may reflect health characteristics in their rates. Rating bands allow health status to affect rates, but only to a defined extent above or below the midpoints. States originally allowed ranges of 25-35%, but many have since tightened the range to 10-20%. Allowing no rate variation based on individual health status is called modified community rating. This is modified or adjusted, rather than pure, community rating because full or substantial adjustment is still allowed for age and sometimes gender factors. States typically allow adjustments for other factors such as which industry the employer is in, or whether employees smoke.

Pure community rating eliminates most of these other factors (including age/gender factors), and retains only location and family size as rating factors. Some states with this form of community rating are not totally pure, to the extent they allow some small additional variation in rates, but if the rating rules greatly restrict the degree of age/gender rating, we classify them as community rating. This is the critical distinction because unconstrained age/gender ratings allow variations of five-fold or more, even if individual health status is entirely removed.

Proposals that allow separate blocks of business add another layer of complexity. Traditionally, many private insurers have treated all small-group business as a single block (or "book" or "class") of business for purposes of rating, product design and marketing, but some have maintained distinct blocks when their products are sold through separate sales forces, when they are acquired from another insurer, or when they have a fundamentally different design, such as HMO versus indemnity products. In keeping with this tradition, many states follow the older NAIC model by applying the rating limits separately to a limited number of blocks defined in this manner. In order to prevent circumvention of the rating limits by block gerrymandering (e.g., defining a block of business to be all the older policies), a number of states also limit the pricing variation among block midpoints. For this purpose, a 20% spread is typical. Finally, rating differences are allowed to reflect the actuarial value of differences among benefits in different policies.

As a consequence, considerable rating flexibility remains even in states with modified community rating or tight rating bands. Because each allowable factor can, in theory, be added to each of the others, and because demographic factors can be very large, at the extreme these restrictions can still allow more than a 10-fold difference in the rates charged two groups at either end of the possible combinations of risk factors. On the other hand, these distant outliers might be very rare.

4. Reinsurance or Risk Adjustment

The third major component of the insurance market reforms is an administered reinsurance mechanism that allows individual insurers to reinsure any risks that are expected to generate costs exceeding the prices they may charge. Reinsurance encourages insurers to accept all applicants by allowing them to pass their highest risks over to an industry-funded reinsurance pool. This outlet suppresses the incentive to engage in risk selection in various indirect and surreptitious ways such as targeted marketing, gerrymandered benefit packages, selective poor service, or "field underwriting" (informal screening by agents). This also makes it less risky for insurers with small market shares to remain in a market with guaranteed issue.

The principal funding for the reinsurance entity is from the reinsurance premium paid when an insurer cedes a group or individual to the pool. Under the NAIC model, which is used by most states with reinsurance, insurers may cede either high-risk groups, or individuals within groups, on payment of a reinsurance premium that for groups is 150% and for individuals is 500% of the market average for the particular plan and case characteristics in question. Since insurers will reinsure only those groups and individuals they predict will have higher expenses than these prices, the reinsurance entity is expected to suffer losses, which are spread back to the insurance market through assessments against participating insurers in proportion to their small-group market shares.

This reinsurance mechanism differs from state high-risk pools that cover individual uninsurable persons because reinsurance is invisible to the public. The decision to reinsure is made solely by the insurer (the "ceding" insurer) when the policy is issued. The ceding company remains responsible for plan administration; the reinsurer merely indemnifies the issuing insurer for its claims expenses. This administered reinsurance mechanism also differs from conventional, private market reinsurance because it is used selectively for groups that are expected to be higher risks than the allowable premium reflects. In contrast, conventional reinsurance covers all of an insurer's risk pool for the unpredictable chance that an actuarially accurate premium will not be sufficient. Commercial reinsurance also does not have a redistributive funding mechanism.3

Reinsurance in effect is an option for insurers to decide individually whether to keep the premium and assume the risk of a particular subscriber or forego the premium and pass the risk on to the rest of the market. Without reinsurance, different companies would randomly, or perhaps systematically, suffer the full brunt of bad risks above the rating bands, with no avenue for relief. Reinsurance also reduces the need for intensive policing of possible tactics for gaming or circumventing the open enrollment and rate compression requirements. In essence, it creates an incentive-based system for controlling the worst effects of medical underwriting without having to carefully monitor the way in which the industry conducts its business.

However, reinsurance is subject to criticism for minimizing the degree of cost sharing borne by the ceding insurer. Ceding insurers retain only a $5,000 deductible and a 10% risk corridor under the NAIC model. This low retained risk provides insurers the greatest encouragement to accept high-risk cases, but this also removes some of the incentive to control costs and manage care for the most expensive cases. However, if this cost-sharing percentage is raised further, it undermines the very purpose of reinsurance, which is to relieve the insurer of the burden of bearing the costs of predictably higher-risk cases. It is very difficult to find a satisfactory compromise between the two goals of cost protection and cost sharing because they are so fundamentally at odds.

Another source of difficulty arises from the considerable disagreement among various sectors of the insurance industry over the existence and structure of the reinsurance mechanism. Some commercial insurers insist that reinsurance is essential if they are to engage in guaranteed issue subject to rating restrictions. HMOs view it as unfair that they would be taxed proportionately for the excess costs of reinsured claims when, in their view, their managed care techniques contribute lower-than-average costs to the pool. Yet other insurers, including Blue Cross, think reinsurance is fine for those who need it, but they are large enough to bear their own risks and do not want to participate in a subsidy system they do not benefit from. Therefore, in most states, participation in the reinsurance system is optional.

An alternative to voluntary reinsurance is a system of administered risk adjustment. In this approach, insurers with higher risks are protected against adverse selection and those with lower risks are prevented from receiving a windfall by calculating how much greater or lesser each insurer's risk pool is than the market average, and requiring insurers with lower risks to make transfer payments to those with higher risks. This approach in theory is more accurate and fine-grained than reinsurance, but the techniques for making the required measurements are still not well-developed and tested, so risk adjustment is being used only on a limited basis in a few states.

5. Purchasing Cooperatives

The health insurance market reforms discussed so far are aimed only at "availability, not affordability," meaning they are only designed to offer insurance to any willing purchaser at prices that do not far exceed the market average, not to impose rate regulation or reduce prices across the market. Reformers hope that focusing competitive pressures on the efficiency of medical care delivery will eventually lower prices relative to covered benefits. But in the short term the reforms may have just the opposite effect by drawing higher risks into the market.

So far, these proposals do not address the higher administrative and marketing costs that make individual and small-group insurance more expensive than large group insurance. This is reflected in the fact that the medical loss ratio for this part of the market is often 10-20 points lower than for larger employers. The medical loss ratio is the portion of the premium that is returned in medical claims expenses; the remaining portion of the premium is absorbed by marketing and administrative overhead, and as profit. Medical loss ratios for large-group insurance are typically 90% or above. For individual insurance, they are in the 70s or lower. Small-group insurance tends to run in the mid-70s to 80s.

There are three reasons why the medical loss ratio is lower for individuals and small groups. First, the per-enrollee cost of marketing is much higher simply because these costs are spread over fewer people. These costs are compounded by the administrative expenses incurred in having to medically underwrite small groups in order to counteract adverse selection. Second, smaller groups have less bargaining clout than large groups, so there is some possibility that insurers are able to realize a higher profit margin (although the profit margin for health insurance generally is quite low). Third, individuals and smaller groups present a greater threat of adverse selection because they are more prone to making their insurance purchasing decision based on individual health characteristics and expected medical needs. Therefore, added effort is required to counteract adverse selection or a higher price is necessary to anticipate its effects.

As a partial solution to many of these diseconomics of scale, several states have created purchasing cooperatives for the small-group market, included self-employed individuals. By pooling purchases of many small buyers, they enjoy some of the same economies of a large group. These public purchasing cooperatives go under several names, but health alliances or health insurance purchasing cooperatives (HIPCs) are the most common. Purchasing cooperatives were a cornerstone of President Clinton's 1993 health care reform plan, as well as of several of the competing proposals.

It is important to distinguish these public purchasing cooperatives from the voluntary, private purchasing groups that have existed in a number of other forms for a number of years under the name of multiple employer welfare arrangements (MEWAs), or in the form of industry association groups such as those sponsored by chambers of commerce. The public purchasing cooperatives differ in several crucial respects. First, sponsorship and management is public, not private. Second, membership is not restricted. Third, administration is more proactive, in that public purchasing cooperatives usually actively oversee marketing, pricing and enrollment. Finally, only the public coops usually have several or many insurers to choose from.

A public purchasing cooperative contracts with a number of insurers who agree to abide by certain marketing and pricing rules. The core idea is to streamline the marketing function and to form a larger risk pool in a fashion that creates for small groups the same bargaining clout, administrative expertise, and employee choice that are typical for large employers. Purchasing cooperatives potentially create several decided advantages for both small employers and insurers. First, they reduce search costs. Small employers do not have the size or money to justify hiring benefits managers. Independent agents typically perform this function, but their commissions add to the premium. Although most coops also use agents and charge administrative fees, they hope to perform this information function for less. Second, purchasing cooperatives can offer individual employees within small groups the same (or an even better) menu of choice as that typically given to employees of large firms, in contrast with the practice of most small employers to offer only a single option. Third, purchasing cooperatives can reduce the administrative costs of enrollment and billing. Finally, a purchasing cooperative can, if it is permitted, wield considerably greater bargaining clout than individual small employers.

Purchasing cooperatives have some potential shortcomings (discussed below), however, and they are seen as threatening to many groups in the insurance industry. Insurance agents are among those who are the most opposed, since coops threaten to displace their function entirely. Smaller insurers, and those who specialize in indemnity offerings, also fear that they will be left out of cooperatives. And, when government authority is used to establish, fund and sometimes administer cooperatives, this is thought to portend growing government involvement in the private insurance industry.

C. The Dangers of Reform

So far, we have drawn for the most part only the most optimistic picture by surveying the need for reform and the prospects for various components of reform. Now, we acknowledge that all of these ideas might not come up roses. Throughout this period, these reforms have attracted some vocal critics that warn about possible adverse consequences, and a number of quieter voices that warn against setting hopes too high. The strongest fear is that these reforms could be counterproductive, since they have the potential to increase prices and decrease coverage. These reforms may raise prices because they make insurance most attractive to the highest-risk subscribers by holding prices to less than the policy's actuarial value to them. The excess is built into the premiums paid by all purchasers, which will drive an undetermined number of lower-risk purchasers out of the market, thus raising the market average even more. This phenomenon is known as adverse selection against the market as a whole. This potential exists because the decision to purchase insurance remains voluntary, and existing purchasers are thought to be highly price sensitive.

What critics oppose most vehemently is pure community rating since this has the most severe rate compression and therefore has the greatest potential to drive out younger, healthier subscribers. This, critics fear, will result in not simply a one-time loss in coverage but an escalating, destabilizing dynamic that may destroy the market. When the first round of subscribers drop out and the community rate increases, this might force still more cycles of subscribers to drop out and subsequent price increases, thus setting into effect an adverse selection spiral that eventually could result in insurance that is so expensive that almost no one would buy it.

A related but different concern is that adverse selection will occur among insurers (in contrast with against the market as a whole). Even if the market as a whole does not lose enrollment, some insurers may be disproportionately hit by loss of subscribers, resulting in some portion going out of business. This affects not only their private welfare, but could compromise the competitive dynamic of the market if too many insurers fold or withdraw. This form of biased selection can occur even without insurers' trying to select good risks because good and bad risks do not always sort themselves randomly among competing insurers. Instead, attributes of some insurers' plans will be more attractive to healthier patients and other attributes more attractive to sicker patients. For instance, it is likely that patients who have strong ties with their physicians because of their past use of medical care will be more hostile to the restrictions in choice of providers imposed by managed care plans such as HMOs. Conversely, younger and healthier patients with no established physician relationships or without anticipated heavy use of medical services find these restrictions to be less of an inconvenience. On the other hand, patients who anticipate more medical expenditures might prefer the lower deductibles and copayments that are typical of HMOs. Therefore, there is considerable debate over whether indemnity plans systematically attract riskier patients than do HMOs. Similarly, subscribers who have mental illness will gravitate towards plans with this coverage, and patients with chronic illness will choose packages that offer generous pharmacy benefits.

Insurers are able to engage in countless techniques to encourage this strategic behavior to their benefit. Some are devious and improper such as poor claims service for sicker patients, or what is known as "field underwriting" -- tacit encouragement of field agents to keep high-risk applications from ever reaching the home office. Other techniques are quite innocuous. Well-baby visits and an ample supply of pediatric specialists tend to attract younger subscribers, and specialists in sports medicine may attract healthy subscribers. Lesser coverage of prescription drugs discourages sicker patients, and deleting from an HMO's drug formulary certain very expensive specialty drugs that treat rare conditions will obviously deter subscribers who suffer from those conditions. Simply choosing one advertising medium rather than another or marketing more aggressively in one geographic area than another is likely to produce widely varying mixes of age and health status. Similarly, a managed care plan's attractiveness to different groups is strongly influenced by its choice of where to locate treatment facilities.

In sum, biased selection can occur both naturally, through patients' choice among different benefit plans, and as a result of insurers' calculated use of often covert selection devices. Insurance reforms heighten this potential since they prevent insurers from accurately pricing individual health risk. This is worrisome behavior that, at a relatively low level, may need to be policed. If it occurs on a larger scale, certain insurers may be forced artificially to leave the market, or certain desirable insurance products and benefit packages may be withdrawn.

Selection problems might manifest themselves in yet another way: at the borders of the regulated market. These reforms necessarily create an unlevel regulatory field, either at the larger end or the smaller end, or both. If small groups are regulated and individual insurance is not, then these reforms may encourage healthier small groups to move into the individual market, or unhealthy individuals to purchase as small groups, in ways that are less than honest. At the larger end, it is not feasible to impose these state-law regulations on large groups since they can easily avoid them by self-insuring, thereby invoking the protection of ERISA preemption.4But, small groups might be tempted to self-insure as well. This too can bleed good risks out of an already troubled market, and it can result in employers taking on too much risk. Finally, small employers might find ways to artificially aggregate into groups that appear larger than the 50-worker threshold in order to avoid these laws, such as by forming into private associations.

All of this creates the potential for a great deal of complex strategic behavior or outright cheating, which in turn requires a good measure of administrative oversight. These concerns raise the question of whether these laws are capable of effective administration.

This outline of the purposes of these reforms and their potential harms and failings points to four central criteria that can be used to evaluate the success of these reforms: the extent to which they promote (1) insurance availability, measured through increased enrollment; (2) affordability, measured through average prices; (3) market competition, measured in a variety of ways; and (4) regulatory administrability, also assessed in a variety of ways. The individual state reports focus on these four criteria rather than the various components of the reform laws just outlined. This is because each of these components interact with each other and have importance across each of these categories of concern. For instance, guaranteed issue, which points primarily to availability, also might increase prices or lead to various circumvention techniques that affect administrability. Or, rating restrictions, which address affordability, might result in less insurance being purchased. Many components of the reforms affect market competition, and some components, such as purchasing cooperatives, affect each of the criteria in equal measures. Therefore, the complexity of this regulatory scheme prevents any neat pairing of components and effects, since each component interacts with all the others and with market and social conditions that are independent of these laws. It is our task to follow the interwoven threads of information in this complex tapestry as we assess whether these reforms have worked as intended, and, if so, why, and, if not, why not.


  1. Guaranteed issue describes a continuous open-enrollment requirement. Open enrollment is usually used to described a more limited period of guaranteed issue, such as one lasting only a month. But, sometimes these terms are used interchangeably.
  2. Observe that, during the first year of implementation, this limit can create a conflict with the restriction on annual increases for those insurers with existing wide spreads who would be forced to raise the rates at the low end of their rating bands more than the 15% annual increase limit.
  3. For an insightful discussion that makes many of these same points, see Randall R. Bovbjerg, "Reform of Financing for Health Coverage: What Can Reinsurance Accomplish?" Inquiry, vol. 29 (Summer 1992), pp. 158-175.
  4. This is true with respect to state reform laws. It is notable the HIPAA for the first time imposes parts of these regulations on self-insured employers as well.


Return to the top >>