This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hall, M. A. Articles by Lawlor, J. S. Search for Related Content PubMed PubMed Citation Articles by Hall, M. A. Articles by Lawlor, J. S. Related Collections Health Reform Insurance - Employer-Based System Legal/Regulatory Issues

Health Tracking

MARKETWATCH

HealthMarts, HIPCs,MEWAs, And AHPs: A Guide For The Perplexed

Abstract

 
This paper considers how pending proposals to authorize new forms of group purchasing arrangements for health insurance would fit and function within the existing, highly complex market and regulatory landscape and whether these proposals are likely to meet their stated objectives and avoid unintended consequences. Cost savings are more likely to result from increased risk segmentation than through true market efficiencies. Thus, these proposals could erode previous market reforms whose goal is increased risk pooling. On the other hand, these proposals contain important enhancements, clarifications, and simplification of state and federal regulatory oversight of group purchasing vehicles. Also, they address some of the problems that have hampered the performance of purchasing cooperatives. On balance, although these proposals should receive cautious and careful consideration, they are not likely to produce a significant overall reduction in premiums or increase in coverage.

The Legislation

Top The Legislation Meeting Objectives The Magnitude Of Concern NOTES

 
Proposals for AHPs and HealthMarts have been circulating in Congress since the Clinton health care reform initiative in 1993.² Their overriding purpose is to enable the small-group market to function more like the large group market, to increase coverage among small-firm workers. The small-group market reforms that swept the country in the early-to-mid-1990s, culminating in the federal Health Insurance Portability and Accountability Act (HIPAA), attempted to address the same issue, but many small firms still believe that these reforms fail to confer all of the advantages enjoyed by large employers. Primarily, these are (1) the economies of scale and bargaining power that arise from bulk purchasing; (2) the ability to use Employee Retirement Income Security Act (ERISA) preemption to avoid costly benefits mandated by state law; (3) the ability to self-insure, which avoids other costs of state regulation (such as premium taxes) and eliminates the profit margin and "risk premium" earned by insurers; and (4) the use of experience rating to achieve lower rates for a pool of healthier subscribers.³

The newer AHP and HealthMart proposals seek to confer these benefits on smaller purchasers by (1) authorizing various forms of pooled purchasing for small groups and individuals; (2) amending ERISA to preempt states’ mandated-benefits laws for coverage offered through these arrangements; (3) allowing association plans to offer self-insured coverage, subject to reduced solvency protections; and (4) allowing rates to be set based on the claims experience generated by each association pool rather than as part of the regular small-group rating structures required by state laws.

Many of these objectives have been pursued already by existing market innovations and by states’ small-group reform laws in the 1990s, so it is important to note how the pending federal proposals are different. Small employers have long purchased insurance through private trade or business associations. Moreover, recent reform laws in a dozen states authorize pooled purchasing arrangements in a form known as health insurance purchasing cooperatives (HIPCs). AHPs and HealthMarts differ from these arrangements in a number of important respects, some of which are summarized in Exhibit 1.

Contrasts also exist with respect to existing private associations. Contrary to the current rule in most states but similar to existing multiple employer welfare arrangements (MEWAs), AHPs would be allowed to offer self-insured products, subject to federal oversight of solvency guarantees. Also, AHPs would not be subject to many aspects of small-group reform laws. State rating rules would not apply, and although AHPs would have to meet HIPAA’s portability and guaranteed-issue requirements for sales inside the association, insurers that sell through AHPs would not have to sell in the regular market, and they could cancel their AHP contracts without offering alternative coverage.

These proposals also differ from existing state laws that allow for so-called bare-bones plans. State laws lift certain mandates but replace them with a more limited set of standardized (and therefore mandated) benefits, and such plans often are offered in a fashion that makes them attractive primarily to higher risks. The AHP and HealthMart proposals waive virtually all mandates, leaving plan design entirely to insurers, associations, and their purchasing membership.

Finally, the pending proposals make critical changes in regulatory jurisdiction. They create new oversight responsibilities for the federal government, diminish the role of state insurance regulators, and clarify and expand the solvency oversight of self-insured associations.

Although this particular package of reforms is unique, none of its components is completely novel. Therefore, there is a large body of experience from which to evaluate whether the new proposals are likely to achieve their objectives or cause unintended disruption of other public policy objectives. This paper undertakes such an evaluation, bringing to bear lessons learned from our extensive study of insurance markets and pooled purchasing arrangements in a dozen states.⁴

Meeting Objectives

Top The Legislation Meeting Objectives The Magnitude Of Concern NOTES

 
Advocates of AHPs and HealthMarts claim that they can reduce costs by 30 percent and draw in as many as 8.5 million previously uninsured purchasers. Others, including independent analysts, claim that cost decreases will be offset by cost increases in the regular market, resulting in either no change in enrollment or only a very small increase.⁵ Even if cost reductions are achieved, others dispute whether this will in fact increase coverage.⁶ We choose here to set aside questions about increased coverage and focus instead on the potential for cost reductions.⁷ Reducing costs is an admirable goal, even if the benefit accrues mainly to those already purchasing coverage. However, cost reductions are problematic problematic if achieved by sacrificing other public policy objectives. Opponents claim that savings through these routes will not be as great as projected; instead, they argue, cost reductions will occur because AHPs and HealthMarts will be used for risk selection and segmentation. We explore both sides by focusing on four sources of potential cost savings: increased purchasing power and economies of scale, avoiding mandated benefits, avoiding other unnecessary state regulations, and risk segmentation.

Increased economies of scale. HIPCs. By amassing small groups into a large pool, proponents hope to achieve economies of scale that increase purchasing power and lower administrative costs. HIPCs have the same objectives, but so far they have not seen much lower prices. Examples of attractive rates cited by some existing HIPCs cannot be readily compared with other market options because they are based on different benefit packages or risk pools. Most HIPC administrators and insurers report that HIPC prices are no lower, and often are somewhat higher, than those in the regular market for equivalent coverage. They explain that HIPCs have been unable to realize administrative economies of scale or to bargain aggressively over price because their enrollment has remained very low, usually much less than 5 percent of the total small-group market.⁸ This points to a chicken/egg dilemma: A HIPC needs substantial enrollment to have the bargaining power to negotiate price discounts, but without discounts a HIPC has trouble attracting the requisite enrollment. Those we interviewed generally believe that there is nothing wrong with HIPCs that greater volume wouldn’t fix.

Part of what is lacking is the up-front enthusiasm and long-range commitment from insurers. Especially in recent years, HIPCs have had difficulty attracting and keeping well-known health plans. Insurers (like many businesses) prefer to avoid competition that focuses on price, especially within a structure that makes it easy for buyers to compare prices for standardized products, as HIPCs do. Health plans would rather compete on the basis of variations in benefit design, provider networks, service levels, brand-name recognition, and so forth. Also, insurers have no reason to favor a structure that brings many small employers, who individually lack any bargaining power, together into an entity that represents their economic interests and has the market share to demand price discounts and other concessions. Further, insurers worry that HIPCs are magnets for higher risks and often operate under rules that create the potential for adverse selection.

AHPs and HealthMarts. In several respects, AHPs and HealthMarts have features that might make them more attractive to insurers than HIPCs are. First, they are expected to attract lower, not higher, risks. Unlike HIPCs, they are not structured to represent exclusively purchasers. Rather than excluding insurers from the governing boards, as HIPCs do, HealthMarts’ boards would be required to include both insurers and medical providers —stakeholders whose economic interests conflict with those of employers. Although AHPs bar insurers from governing boards, in practice, insurers or brokers often establish associations and contract to administer them. Moreover, AHPs and HealthMarts may offer the products of only a single insurer, rather than forcing insurers to compete as HIPCs do. Thus, these proposals create the potential for insurers to use AHPs or HealthMarts as sales vehicles for specialized products under favorable regulatory rules. This seems more likely to generate the level of insurer participation and "buy-in" needed to induce aggressive marketing, which, if successful, may produce the economies of scale that have eluded HIPCs.

Administrative costs. Still, skeptics question the potential for lower prices. They observe that with thin profit margins there is little room to reduce price purely by negotiation. Instead, price reductions are more feasible if real economies can be achieved by lowering the cost of sales and administration. Our detailed case studies of HIPCs reveal much less opportunity for administrative cost reductions than first appears possible. A major component of these costs is agents’ commissions. Some HIPCs at first hoped to operate without agents but quickly found that agents are essential to reaching small employers. Even without agents, HIPCs have had to replicate many of the same advertising, informational, and processing costs. These and other costs are built into the HIPCs’ administrative charges, which typically run about 2–3 percent of premiums. Similarly, private associations charge both administrative fees and membership dues, and they frequently receive a commission or royalty from the insurer.

The goal of achieving the efficiencies that exist in the large-group market remains elusive. A purchasing organization of smaller groups inherently has higher costs for the very reason that it is not a single large group. Each employer continues to make its own purchasing decisions, which means that proportionate sales costs continue to be higher, as are the costs of billing and collections. Moreover, for larger employers, many of these informational and administrative costs are internalized within human resource departments and so do not appear as part of the insurance premium. premium. For small employers that lack this staff, these functions must be "outsourced," either to agents paid by commission or to association staff paid through dues and fees. Necessarily, this means that small employers will pay higher premiums than large employers do for the same coverage.

On balance, there is no clear evidence that group purchasing arrangements will achieve much savings through economies of scale. Some savings are likely, but these are probably marginal at best. The Congressional Budget Office (CBO) estimated no savings through this route, based on a survey of employers that failed to find any documented savings from using group purchasing arrangements.⁹

Eliminating mandated benefits. It not clear that eliminating mandated-benefit requirements would lower the cost of health insurance very much. This depends on whether employers will prefer pared-down coverage. Even proponents of these new entities observe that large employers, whom they hope to emulate, tend to offer most of the benefits covered by statemandates.¹⁰ They envision that small employers, especially when facing a tight labor market, will eliminate only peripheral benefits such as hair transplants, vitro fertilization, and chiropractic services. Naturally, to the extent that this is true, the achievable cost savings are lower. Regardless, proponents argue, the choice should be made by purchasers and not be imposed by the government. Opponents respond that there are good reasons to require certain benefits such as maternity care or mental health and substance abuse treatment. If most people are not required to purchase these socially important benefits, people who need them will not be able to afford them. We do not enter into this well-established debate here but simply observe that if mandates were lifted, existing evidence is not conclusive about whether small employers would opt to purchase much cheaper benefit packages. Studies have shown that barebones plans sell poorly and that self-insured employers, free of benefit mandates, offer many of these benefits because workers desire them.¹¹

Reducing state regulation. Freeing the new entities from a variety of state regulations is seen as another source of savings. Insurers now must register, meet solvency standards, submit to inspections, and obtain regulatory approval for contract forms in every state in which they do business. The proposed AHP legislation would greatly simplify and expedite this process by consolidating jurisdiction either in the Department of Labor (DOL) or in one state to which the DOL chooses to delegate regulatory authority. The DOL may choose either the state where most association members reside or the state that the association chooses to declare as its legal domicile (not necessarily where it is headquartered). Also, the bill provides for an application fee of only $5,000 to defray the costs of inspecting and certifying AHPs, which is less than many states charge and much less than paying filing fees in several states at once.

Eliminating duplicative state jurisdiction makes sense and will save costs. To prevent associations from selecting the state with the most lenient rules, the DOL would determine which state has jurisdiction. However, as discussed below, insurers might be able to use AHPs as away to exploit regulatory gradients that exist among states by quickly entering and leaving markets according to the best prospects for risk selection. Also, opponents object that the application fee is sufficient for only one-time inspection, not ongoing oversight.

In other respects, the proposed legislation would reduce but not eliminate the costs of state regulation. First, in contrast with self-insured employers, states could continue to impose premium taxes on insurance sold through HealthMarts and new AHPs. Only existing associations would be exempted. Another area of potential regulatory cost savings is lower solvency standards imposed on self-insured associations. There are good reasons for not eliminating these protections for associations, which unlike large self-insured employers have no source of assets to make good on their promises other than the premiums they collect. Associations also may lack the cohesion, commitment, and sense of community within a large employer group. They are not subject to the same labor-market forces and collective bargaining pressures that induce large employers to keep their self-funded benefits promises. Therefore, the proposed solvency protections are substantial.

Improving solvency oversight of self-insured associations is one of the primary purposes of the AHP legislation, in response to continuing problems with what are currently known as MEWAs. MEWAs’ complex legal status has evolved over time, largely in response to problems of mismanagement or fraud resulting in bankruptcies that have left hundreds of thousands of people with unpaid medical bills.¹² Despite amendments to ERISA in 1983 clarifying state authority and creating federal oversight of MEWAs, problems persist because of the following uncertainties and gaps in regulatory authority: (1) Regulators have poor information about the existence and operation of MEWAs, making it difficult to monitor or detect problems, especially by unscrupulous operators. (2) The extent of allowable state regulation is uncertain, allowing MEWAs to resist regulatory efforts by challenging them in courts. (3) ERISA exempts from state regulation single employers with several divisions and MEWAs that are sponsored by collective bargaining arrangements or by employee leasing firms. Some MEWAs have configured themselves as bogus or questionable versions of these exempt categories to evade regulation.

AHP proposals seek to rectify these problems with additional amendments to ERISA that rename MEWAs as AHPs, require them to register, clarify various ambiguities, close loopholes, and create uniform federal solvency standards to be administered by the DOL (subject to grandfathering provisions for certain existing entities).¹³ As a result, from the employer’s point of view, purchasing through self-insured associations is not the same as self-insuring. Instead, the legislation substitutes one regulated risk-bearing entity (an AHP) for another (an insurer). To this extent, the costs of meeting solvency standards will not be eliminated. However, the solvency requirements for AHPs are not as great as under existing state insurance regulation and can be met through more flexible means, such as stop-loss coverage or reinsurance rather than reserves. Such changes could save costs.

On balance, although opponents may object that lowering these various regulatory costs will lead to insufficient oversight, the proposed legislation imposes far more demands on small-employer pools than are now imposed on self-insured large employers, which are not subject to any regulatory costs. To this extent, the potential savings are not as great as those enjoyed by large employers.

Risk segmentation. The strongest attack levied against these new forms of pooled purchasing comes from those who fear that they will undo existing risk pools, forcing higher-risk groups to drop coverage. These opponents maintain that any cost savings will be achieved primarily by allowing younger, healthier purchasers to form separate risk pools that offer lower rates, at the expense of the elderly and chronically ill. They note that because there is a huge variation in health risks, it is far easier for insurers to lower their price by attracting better risks or avoiding worse ones than by achieving real efficiencies in the sale of insurance or the delivery of care. This is especially true for smaller indemnity insurers, which lack the capital and market concentration necessary to form effective managed care networks. Critics maintain that risk segmentation will occur primarily because the waiver of mandated benefits allows associations to offer more bare-bones plans that appeal more to healthier members. Critics fear that the sick will be stuck in the community-rated, fully regulated market while the healthy flee to the lightly regulated, mandate-free association plans. We consider here various ways in which this might occur.

Benefit differences. Even if association plans offer a wide range of benefits, there are still legitimate reasons to expect some risk segmentation attributable to benefit design. Some associations will likely avoid important mandates such as pregnancy and mental health coverage, which would cause subscribers to sort themselves according to their need for these services. Moreover, benefits can differ apart from mandates. Plans can have widely varying deductibles and copayments. Even if an AHP covered all or most mandated benefits, it could impose a much higher deductible for all services or much larger copayments for pharmaceuticals. Such across-the-board limits discourage higher risks and appeal more strongly to healthier people.

Although these same benefit differences appear in the regular, fully regulated market, small-group rating rules require that price differences between an insurer’s plans reflect primarily the risk-neutral differences in benefits, not the claims experience that results from attracting different risk pools.¹⁴ AHPs, however, are free to do the latter. Current small-group rating reforms require an insurer to treat its entire bloc of small groups in state as a single pool, and the reforms then prescribe allowable variations among groups according to specific demographic and health status factors. More liberal rating rules allow greater variation but only around a single market index rate. Some states go further and allow an insurer to use different index rates for different blocs defined by product lines and sales forces, but these bloc midpoints cannot vary more than a prescribed amount (usually, ±10 percent).

AHPs (and perhaps also HealthMarts) would be subject to these rating rules only within their pools and not across or between different pools in which an insurer might sell.¹⁵ Therefore, insurers selling through AHPs charge different rates to different AHPs based on the claims experience of each pool, similar to rating for larger employer groups.

Even if AHPs offer the same kinds of benefits available elsewhere in the market, their rates will still reflect the risk-selection patterns that result from normal differences in benefits, a result that the regulated market attempts to avoid (although not always with great success). These benefits-related selection patterns can be very pronounced. In our case studies we observed that a number of plans that were identical except for their deductibles attracted substantially different risk pools. For instance, in New Jersey’s individual market, leading insurers in 1999 charged on average $3,312more for single coverage with a $500 deductible than for identical policies with a $1,000 deductible. In other words, the difference in premiums was more than six times greater than the difference in benefits. Some of this disparity is attributable to increased utilization that results from a lower deductible, but most is due to how people naturally sort themselves according to their perceived need for coverage. Similarly, in New York, which in the individual market allows only a standardized plan to be sold in two formats—HMO and point-of-service (POS)—the two leading insurers report that their 1997 claims costs were two-thirds greater under the POS plan, primarily because it attracts higher risks. In North Carolina in 1993, plans with leaner benefits cost two to three times more than those from the same insurer with richer benefits. This effect was clearly the result of risk selection, since only the richer plans were subject to full medical underwriting. These patterns are consistent with large employers’ experience. When they offer plans with widely different benefits, they often find that the better benefits attract workers and families with much worse health risk.¹⁶

Membership criteria and "bona fide" status. The proposed legislation attempts to minimize risk selection in several ways. First, HealthMarts must accept all groups, and AHPs can restrict purchasers only according to valid membership criteria. Membership criteria cannot be used as a proxy for health status. Second, only "bona fide" associations are allowed—those formed primarily for noninsurance purposes and in existence for three years. The effectiveness of these protections can be evaluated by examining the experience of current associations under existing reform laws. This evidence indicates that these protections will avoid only the more overt forms of risk selection such as medical underwriting or excluding high-risk groups. They will not eliminate subtle forms of selection that occur through more natural or covert means.

Under the legislation, associations may restrict their membership to specified occupations. Workers often sort themselves by age, sex, and health condition. This tendency is reflected in the fact that prior to small-group reforms, underwriters rated coverage according to industry classification, with premium variations of 50 percent or more across industries. AHPs would, in effect, allow insurers to resume industry rating. Insurers could charge more to sell through AHPs representing high-risk industries, and they could offer much lower prices to the favored groups based on their actual claims experience.

The requirement of bona fide status and the prohibition of "fictitious groups" have not been effectively enforced. Both HIPAA and state law require that associations exist for noninsurance purposes, yet many associations exist mainly to sell insurance. Typically, these insurance-run associations have virtually no membership limitations, with names such as "Associated Industries," "Business Council," and "Association of Self-Employed." Some observers derisively refer to these as "air-breather" associations: Anyone who breathes air is eligible to join.

Nevertheless, these associations have protected status under state law because they are formed with stated purposes to promote business interests, and they indeed offer member benefits besides health insurance. In substance and form, these insurer-run associations do not differ from broadly inclusive associations formed by Chambers of Commerce and other legitimate business groups. Even where insurers do not operate these associations, we frequently observed in our case studies that offering an insurance product is the primary purpose for the association’s existence.

Destabilizing the market. There is nothing inherently wrong with air-breather associations operated primarily for insurance purposes. However, under the proposed regulatory structure, they are likely to destabilize the market, for two distinct reasons. First, allowing experience-rated associations with fewer regulatory controls to operate alongside a community-rated and more highly regulated market creates a sharp regulatory gradient that will erode, or collapse, the community-rated pool.¹⁷ Second, even if associations were to constitute the entire market, they tend to be unstable because, unlike larger employer groups, nothing binds individual members to the association. Purchasers shop among associations for the best price, and, under an experience-rated system, price differences are driven primarily by the health risk of persons in different pools. We illustrate these concerns with examples from our case studies.

The potential of experience-rated associations to undermine market reforms is the greatest in states with pure or nearly pure community rating, which creates the greatest rewards for separating out low-risk groups. For instance, in Vermont, which has nearly pure community rating (only ±20 percent variation allowed, for all risk factors combined), experience-rated associations account for more than 60 percent of small-group enrollment, and more than 95 percent of enrollment in Blue Cross, which is by far the largest small-group carrier.¹⁸ The portion of small-group business sold as community-rated plans is shrinking at such a rate that one informant described the small-group market as "disappearing" into associations. Two others said that the only people who continue to buy outside of associations are those who are "too befuddled" to have "figured it out yet."

Similarly, in Kentucky, associations were exempted from guaranteed-issue and community-rating requirements in both the individual and small-group markets, making them the only source available for medically underwritten coverage. Following rapid shifts in enrollment, forty-five associations in Kentucky accounted for 35 percent of the combined individual and small-group market in 1997. Several insurers sold only through exempt associations, and most of those that did not left the market altogether.¹⁹

The potential to erode regulated markets also can be seen in differentials between community and association rates in states where they coexist. In New York we found that association rates for individual (self-employed) coverage were 30 percent lower than in the individual community-rated market.²⁰ In Vermont, community rates for some insurers were 30–50 percent higher than what they offered through associations for the same products.

The second form of destabilization is best illustrated by Vermont, where many interview subjects described small-group associations as a "scam" that undermines the purpose of the small-group law, fragments and disrupts the market, and allows insurers to compete by cherry picking. Because many small employers join associations only to buy insurance, they are willing to switch associations frequently to achieve lower rates. For several of Vermont’s largest business associations, this has produced wild swings in both association membership and their insurance rates, through the following cycle: An association that begins with a small, select membership finds that its lower insurance rates are a strong draw for new members (who pay association dues), and so the association uses its low insurance rates to advertise aggressively for more members. As small employers flock to the association, its membership multiplies (as much as twentyfold), but its risk pool becomes less select, so the insurer imposes a steep (20–35 percent) premium increase. This causes a "meltdown," in which employers switch to smaller associations with better risk pools, causing the tumultuous process to repeat. This is not the market dynamic envisioned by AHP proponents.

Durational effects and churning. It might be expected that these disruptive shifts would settle down over time as risk pools become fairly homogenous among associations. However, this market structure might never reach a satisfactory equilibrium. Insurers have strong incentives and opportunities to use associations to engage in purposeful risk segmentation. This can occur through a phenomenon known in actuarial practice as the "durational effect," according to which new subscribers tend to be much better risks than existing subscribers are. Prior to guaranteed issue, it was thought that this was the result of medical underwriting, which selects a healthier-than average pool of new subscribers who, over time, succumb to the law of averages (or, in statistical parlance, "regress to the mean"). Insurers were surprised to find, then, that the durational effect has remained strong even after underwriting was eliminated or drastically curtailed.²¹ This indicates that most of the "select" risk quality of new small-group subscribers was attributable more to natural self-selection behavior than to insurers’ prowess in identifying and excluding bad risks.²²

This self-selection stems from the fact that people with health problems are averse to changing their health insurance. Since most people who buy insurance already have coverage, those shopping for different coverage are healthier than those who are not. Another reason the durational effect might continue to be strong following underwriting reforms is that it arises in part from preexisting condition exclusions. Although insurance reforms purport to limit these exclusions, the one-year limit in HIPAA for small groups is no shorter than what was already standard practice.

Associations give insurers the opportunity to capitalize on this natural durational effect by continually starting new associations and closing old ones. Eliminating this practice, known as "churning," is a primary purpose of the 1990s market reform laws. AHPs (and possibly HealthMarts) reinstate this tactic, for the following complex reasons.

First, insurers would be allowed to experience-rate each association rather than requiring that each insurer’s experience be pooled across associations or across the rest of its small-group business. Second, insurers would not be required to offer their association plans in the regular market.²³ Third, and most important (and largely unrecognized), insurers could freely exit association pools without offering displaced subscribers alternative coverage. In the fully regulated market, states limit insurers’ ability to withdraw from markets, either by requiring them to continue to renew existing subscribers or by refusing them permission to reenter the market for a significant period of time (such as five years). AHP legislation allows insurers to enter and exit markets at will, simply by forming or terminating contracts with associations.²⁴

Given this freedom, insurers can engage in the following strategy: They first find a sponsor to forma new association (or find an existing one with a small membership) and offer an insurance plan at an attractive rate with benefits that appeal to healthier subscribers. Because of the natural selection effects noted above, new subscribers will tend to be healthier than average. As the risk pool worsens over time, the insurer can pull out of the association or simply start a new one. Although the existing association members can switch to the new association, because of inertia they often stay, especially if they are sick. The insurer is therefore able to start over with a freshly selected risk pool, leaving behind the higher risks. Even if the insurer does not cancel its existing association business, it can use the same techniques to segregate its risk pools simply by forming a newer association offering similar coverage. Again, the natural resistance of sick subscribers to change plans means that the new pool will be healthier and can offer lower rates.

The Magnitude Of Concern

Top The Legislation Meeting Objectives The Magnitude Of Concern NOTES

 
Regardless of market rules, insurers in a competitive market have a powerful incentive to be skillful risk selectors, owing to the vast range in health risks. It should be no great surprise, then, to learn that extensive opportunities for risk selection and segmentation would exist under the AHP proposals (and perhaps to a lesser degree under the HealthMart proposals, depending on how rating rules apply to them). These opportunities exist even under current market rules.²⁵ Therefore, these concerns should be kept in perspective. If risk selection effects operate only at the margins and do not overwhelm more socially productive market forces and regulatory rules, these concerns should not be grounds for rejecting a promising initiative.

The case studies we have conducted indicate that this may be so. Although existing associations we studied do offer lower rates, mainly as a result of their better risk pools, those rates are not vastly different from market rates. In Vermont the rate differentials tend to be no greater than the rating flexibility allowed in the community-rated market (±20 percent), and some non-AHP insurers offer community rates that are much lower than AHP rates from other insurers. In New York the association rates in the individual market are equivalent to the small-group community rates. Therefore, it cannot be said that the association exception has completely undermined the community-rating laws in these states. Moreover, in New York, agents said that associations offer the only affordable option for the self-employed.

Nevertheless, the proposals move distinctly away from existing market-reform rules and toward older market structures in which competitive forces focused mainly on risk selection issues. This shift runs contrary to the stated purpose of promoting lower prices through bargaining power and administrative efficiencies. Instead, insurers are encouraged to become skillful risk selectors rather than effective managers of care. In short, the potential of these new purchasing entities to lower prices through risk segmentation is greater than through all of the more socially productive mechanisms combined.

A case can be made that HealthMarts and AHPs violate the principle "first, do no harm." They are unlikely to generate true efficiencies, and they partially undermine existing regulatory and market structures. On the other hand, HealthMarts seek to overcome some of the current limitations of purchasing cooperatives. AHPs seek to reduce, simplify, and clarify existing state and federal regulation that in some respects is redundant, ineffective, or obscure. Even if these changes do not greatly reduce prices or increase coverage, some recasting of the law is desirable simply because existing rules are so impenetrably complex that they are at risk of being dysfunctional, especially with respect to MEWAs and other associations. However, it is a tall order to craft legislation that accomplishes these improvements without allowing insurers to exploit regulatory gradients and circumvent market boundaries necessary to maintaining the integrity of previous reform efforts.

Editor's Notes

 
Mark Hall is professor of law and public health at Wake Forest University School of Medicine in Winston-Salem, North Carolina. Elliot Wicks is a senior fellow at the Economic and Social Research Institute in Washington, D.C. Janice Lawlor is project manager, Department of Public Health Sciences, at Wake Forest.

This research was funded by grants from the Robert Wood Johnson Foundation, under its Changes in Health Care Financing and Organization Program. The analysis and conclusions presented here do not necessarily represent the views of the foundation. This paper benefited greatly from the comments of two anonymous reviewers.

NOTES

Top The Legislation Meeting Objectives The Magnitude Of Concern NOTES

 

1. For an introductory discussion, see M.A. Hall, "The Geography of Health Insurance Regulation," Health Affairs (Mar/Apr 2000): 173–184.
2. These proposals first came to widespread attention through a bill (H.R. 995) in the 104th Congress championed by Rep. Harris Fawell (R-IL) in 1995. Recently, the most prominent version has been sponsored by Rep. Jim Talent (R-MO), chair of the House Small Business Committee, and these ideas were included in George W. Bush’s campaign proposals. This legislation gained notoriety in October 1999 when conservative Republicans in the 106th Congress attached it to the bipartisan Patients’ Bill of Rights (H.R. 2990), a move President Bill Clinton described as a "poison pill" and "legislative sleight of hand" calculated to defeat the bill. Similar legislation has also been highly controversial at the state level, most notably in California. See J. Yegian, Size Matters: The Health Insurance Market for Small Firms (Aldershot, England: Ashgate Publishing, 1999).
3. By "risk premium," we mean the portion of the premium that compensates for the pure riskbearing function—that is, the uncertainty over the extent of claims.
4. See also the Health Insurance Market Reform Study’s Web site, <www.phs.wfubmc.edu/insure>; and E.K. Wicks, M.A. Hall, and J.A. Meyer, "Barriers to Small-Group Purchasing Cooperatives," March 2000, <www.esresearch.org/Documents/HPC.pdf> (11 September 2000).
5. Congressional Budget Office, Increasing Small-Firm Health Insurance Coverage through Association Health Plans and HealthMarts, January 2000, <www.cbo.gov> (11 September 2000); and Len NicholsLen Nichols, principal research associate, Urban Institute, testimony before the House Commerce Subcommittee on Health and Environment, 16 June 1999, <www.urban.org/TESTIMON/nichols6-16-99.html> (11 September 2000). See, generally, Mary Nell Lehnhard, senior vice-president, Blue Cross and Blue Shield Association, testimony before the House Committee on Small Business, 16 February 2000, <www.house.gov/smbiz/hearings/106th/2000/000216> (11 September 2000).
6. For a sampling of recent articles pro and con on related topics, see P. Ketsche and W. Custer, "The Effect of Marginal Tax Rate on the Probability of Employment-Based Insurance by Risk Group," Health Services Research 35, no. 1 (2000): 239–252; and D. Goldman, J. Buchanan, and E. Keeler, "Simulating the Impact of Medical Savings Accounts on Small Business," Health Services Research 35, no. 1 (2000): 53–75. See generally E. Wicks and J. Meyer, "Small Employer Health Insurance Purchasing Arrangements: Can They Expand Coverage?" May 1999, <www.nchc.org/releases/stevesedit> (11 September 2000).
7. We also sidestep other important aspects of the debate over these proposals, such as the adequacy of the solvency standards and the shift from state to federal regulation.
8. E. Wicks and M. Hall, "Purchasing Cooperatives for Small Employers: Their Performance and Prospects," Milbank Quarterly 78, no. 4 (2000).
9. CBO, Increasing Small-Firm Health Insurance Coverage; and S.H. Long and M.S. Marquis, "Pooled Purchasing: Who Are the Players?" Health Affairs (July/Aug 1999): 105–111.
10. Hearings on Association Health Plans, before the House Committee on Small Business, 10 June 1999, <www.house.gov/smbiz/hearings/106th/1999/990610> (11 September 2000).
11. G.A. Jensen and M.A. Morrisey, "Employer-Sponsored Health Insurance and Mandated Benefit Laws," Milbank Quarterly 77, no. 4 (1999): 425–459; and M.S. Marquis and S.H. Long, "Recent Trends in Self-Insured Employer Health Plans," Health Affairs (May/June 1999): 161–166.
12. See R. Tillman, Broken Promises: Fraud by Small Business Health Insurers (Boston: Northeastern University Press, 1998); F. Damon, "Multiple Employer Trusts: A Historical Perspective from ERISA to the California Approach," Journal of Insurance Regulation 2, no. 1 (1983): 20–29; U.S. General Accounting Office, Employee Benefits: MEWA Regulation, Pub. no. GAO/HRD 92-40 (Washington: GAO, March 1992); K. Polzer and J. Jones, "Multiple Employer Purchasing Groups (METs, MEWAs, HINs, HIPCs): The Challenge of Meshing ERISA Standards with Health Insurance Reform," National Health Policy Forum Issue Brief no. 604 (Washington: NHPF, 1992); K. Polzer, "Preempting State Authority to Regulate Association Plans: Where Might It Take Us?" National Health Policy Forum Issue Brief no. 707 (Washington: NHPF, 1997); A. Martin et al., "MEWAs: An Exception to ERISA Preemption: Why, What, and When" (American Law Institute–American Bar Association, February 1992), C724, 363; and C. Forrelli, R. Jones, and C. McHugh, "Regulation of Multiple Employer Welfare Arrangements: The Dilemma of Dual Federal/State Regulation," FICC Quarterly (Fall 1995): 45–63.
13. See Polzer, "Preempting State Authority."
14. M. Hall, "The Structure and Enforcement of Health Insurance Rating Reforms," Inquiry 37, no. 4 (2001).
15. The legislation creates much ambiguity. It states that HealthMarts are subject to state small-group rating rules but that policies must be rated on a "product-specific" basis, without defining what this means. One interpretation is that insurers are required to use a different rating structure for plans they sell only within a HealthMart rather than including these products within their overall small-group rating structure, as state laws require. The purpose may be to allow HealthMarts to pass on savings from leaner benefits than are offered elsewhere in the market, but, as discussed below, the effect may also be to allow insurers to reflect the different health status of those who join a HealthMart pool.
16. J. Robinson and L. Gardener, "Adverse Selection among Multiple Competing Health Maintenance Organizations," Medical Care 3, no. 12 (1995): 1161–1175; and J. Newhouse, "Reimbursing Health Plans and Health Providers: Selection versus Efficiency in Production," Journal of Economic Literature (September 1996): 1236–1263.
17. Here we use "community rating" in a looser sense to include both pure and modified community rating, as well as rating bands that allow some flexibility on price. The distinction we are drawing is not the degree of rating flexibility, but rather whether the rating structure is based on the entire community or instead on the claims experience of a specific pool or product.
18. M. Hall, "An Evaluation of Vermont’s Health Insurance Reform Law," Journal of Health Politics, Policy and Law (February 2000): 101–131.
19. Kentucky Department of Insurance, Market Report on Health Insurance (Frankfort: KDI, April 1997).
20. M. Hall, "An Evaluation of New York’s Health Insurance Reform Law," Journal of Health Politics, Policy and Law (February 2000): 71–99.
21. Hall, "The Structure and Enforcement."
22. Another study found no difference in one large insurer’s claims over six years between small groups that were underwritten and those that were guaranteed issue. J. Glazner et al., "The Questionable Value of Medical Screening in the Small-Group Health Insurance Market," Health Affairs (Summer 1995): 224–234.
23. Technically, the same is true for HealthMarts as well as AHPs, but unlike AHPs, HealthMarts may not restrict membership, so selling through them is equivalent to selling to the entire small-group market.
24. This flexibility already exists in a number of states, primarily in the Midwest. Hall, "The Geography of Health Insurance Regulation." In essence, the AHP proposal seeks to extend the more lenient regulatory approach to all states.
25. M. Hall, "The Impact of Health Insurance Market Reforms on Market Competition," American Journal of Managed Care (January 2000): 57–65.

                      What's this?

This article has been cited by other articles:

				M. Kofman, K. Lucia, E. Bangit, and K. Pollitz Association Health Plans: What's All The Fuss About? Health Aff., November 1, 2006; 25(6): 1591 - 1602. [Abstract] [Full Text] [PDF]