ERISA Preemption and
Kentucky Ass’n of Health Plans, Inc. v. Miller :
Before and After

By: Kamyar Mehdiyoun
© Aspen Publishers



On April 2, 2003 the Supreme Court decided Kentucky Ass’n of Health Plans, Inc. v. Miller1. The court ruled that Kentucky’s so called ‘Any Provider Law’ was not preempted by Employee Retirement Income Security Act (ERISA).2 The relatively brief decision is a landmark in the field of ERISA preemption. Writing for a unanimous court, Justice Scalia’s opinion held that states may require HMOs and other insurers to provide their customers with an open network of health care providers.3 More significantly, the court discarded the two-part test that had guided its previous preemption decisions, and replaced it with a simpler and much broader preemption standard.4 The opinion stated that the previous two-part test had proved to be unworkable, and had led to confusion in lower courts.5 It, therefore, appears that the Court hoped that its newly adopted standard would resolve the uncertainty in the field of ERISA preemption, which in turn, would result in less litigation.6 This article will first examine the Supreme Court’s pre-Miller preemption jurisprudence. It will then turn to a discussion of Miller, and will conclude that the extraordinary breadth of the new preemption standard is likely to lead to even more uncertainty and litigation.


ERISA’s Preemption Provisions

ERISA preempts any and all State laws insofar as they may “relate to any employee benefit plan”.7 There are few exceptions. Section 514(b)(2)(A), the so called ‘saving clause’, exempts from preemption, state laws that “regulate insurance, banking or securities”. In order to prevent state regulation of plans on the theory that plans are insurers, ERISA’s ‘deemer clause’ provides that “Neither an employee benefit plan… nor any trust established under such a plan, shall be deemed to be an insurance company or other insurer, bank, trust company, or investment company or to be engaged in the business of insurance or banking for purposes of any law of any State purporting to regulate insurance companies, insurance contracts, banks, trust companies or investment companies.”8

Both the Supreme Court and appeals courts have repeatedly criticized the “complex”9 and “unhelpful”10 structure of ERISA’s preemption language, referring to it as “not a model of legislative drafting”11 which leaves discerning its scope as difficult as unraveling the fabled Gordian knot12 and consuming a “substantial share” of the Supreme Court’s time.13

One source of difficulty in interpreting and applying ERISA’s preemption clause is that it applies to state laws regardless of whether those laws concern matters governed by ERISA. Instead the standard for preemption is whether the state law relates to any employee benefit plan.14 The Congressional Conference Committee in charge of drafting the final version of the bill that became ERISA, altered the original, narrower language applicable only to state laws relating to the specific subjects covered by ERISA, and instead adopted the much broader language of 514(a).15

Yet another difficulty arises from the fact that, as the Supreme Court put it “while the general preemption clause broadly pre-empts state law, the saving clause appears broadly to preserve the States’ lawmaking power over much of the same regulation. While Congress occasionally decides to return to the States what it has previously taken away, it does not normally do both at the same time.”16

Reflecting the statutory scheme of ERISA’s preemption section, the courts begin their analysis of state laws by first determining whether such laws ‘relate to’ to an employee benefit plan. If this requirement is satisfied, the analysis then proceeds to whether the ‘saving clause’ allows state regulation of the activity in question. Finally, the courts inquire whether application of the ‘deemer clause’ changes the result of the analysis.


State Regulation of Insurance and Supreme Court’s Pre-Miller Jurisprudence

The U.S. Constitution empowers Congress to ‘regulate commerce with foreign nations, and among the several states.”17 During the early 1940’s, and especially in the aftermath of the Supreme Court’s landmark decisions in United States v. Darby18 and Wickard v. Filburn19, in which the Court established the broad reach of the commerce clause, the extent of state regulation of insurance emerged as a contentious topic.20 An early pronouncement of the Supreme Court on the subject had been made in 1869 in Paul v. Virginia21. In that case the Court decided that issuing a policy of insurance did not qualify as commerce22 and, thus upheld Virginia’s regulation of fire insurance contracts against a commerce clause attack.23 For the ensuing seventy-five years the courts followed Paul v. Virginia and consistently held that regulation of insurance transactions was exclusively a state matter.24

In 1944, in United States v. South-Eastern Underwriters Assn., the Supreme Court once again confronted the issue.25 The defendants in Underwriters were fire insurance companies that had engaged in anti-competitive practices by fixing premium rates and agents’ commission.26 The question was whether the Sherman Anti-Trust Act applied to the insurance companies’ practices.27 In their defense, the insurance companies argued that they were not required to conform to the standards of business conduct established by the Sherman Act because ‘the business of fire insurance is not commerce.’28 Departing from its well-established precedents, the Supreme Court held that the business of insurance was interstate commerce and was, therefore, subject to the Sherman Act.29 The Court examined both the language of the Sherman Act and its legislative history and concluded that Congress had not intended to exempt the insurance industry from the Act’s broad reach.30

In 1945, in response to Underwriters, and concerned about the detrimental impact of the decision on the states’ ability to regulate insurance, Congress adopted what came to be called the McCarran-Ferguson Act.31 In passing the Act, Congress declared that “the continued regulation and taxation by the several States of the business of insurance is in the public interest.”32 Congress’ intent in passing the McCarran-Ferguson Act was to turn back the clock to pre-Underwriters era, and to assure the States’ supremacy in regulating the business of insurance within their territories.33

Section 2(a) of the Act provided that “[T]he business of insurance, and every person engaged therein, shall be subject to the laws of the several States which relate to the regulation or taxation of such business.”34 As the Supreme Court has noted, the statute did not provide a blanket exemption for the activities of insurance companies35; instead, the states were given a free rein only when their laws regulated the ‘business of insurance.’36

In Group Life & Health Ins. Co. v. Royal Drug Co., the court confronted the task of defining the phrase ‘business of insurance’.37 Royal Drug is an especially significant case both because it illustrates the Court’s interpretation of that key phrase, and for its factual similarities to Miller.

In Royal Drug, Blue Shield provided a benefit in the form of prescription drugs.38 The insured could receive the drugs in exchange for a $2 payment.39 In order to satisfy its obligations and meet the demands of the insureds, Blue Shield had entered into mutual contracts with a group of pharmacies.40 The contracts at issue provided that having dispensed the drugs for the nominal amount, the pharmacies would then receive the remainder of the drug cost directly from Blue Shield.41 Thus, “only pharmacies that [could] afford to distribute prescription drugs for less than this $2 markup [could] profitably participate in the plan.”42 While large pharmacy chains could profitably operate under the contracts, smaller, independent pharmacies could not do so. A group of these independent pharmacies brought an anti-trust action against Blue Shield, alleging that the insurance company’s contracts with participating pharmacies constituted price-fixing and group boycott in violation of Sherman Act.43 It is worth noting that there were no state laws at issue in the case. The question before the court was whether the private agreements between Blue Shield and participating pharmacies were the ‘business of insurance’ within the meaning of §2(b) of McCarren-Ferguson Act.44

The Court’s opinion in Royal Drug began by emphasizing that the main element of an insurance contract was risk spreading.45 The agreements between the insurer and providers did not spread any risk, because they did not affect the insurer’s promise to the insured that they would be able to obtain their prescription drugs in exchange for the $2 payment.46 Highlighting another characteristic of the ‘business of insurance’ the Court noted that “the relationship between insurer and insured, the type of policy which could be issued, its reliability, interpretation, and enforcement– these [are] the core of the ‘business of insurance.'”47 Since the pharmacy agreements were not between insurer and insured, and were instead separate contracts between the insurer and providers, they failed to satisfy this core element of ‘business of insurance’.48 The Court also cited authority for the proposition that exemptions from the antitrust laws were to be narrowly construed49, and concluded that the application of that principle was especially appropriate to the pharmacy agreements at issue, because those agreements involved parties wholly outside the insurance industry.50 The Court, therefore, concluded that the pharmacy agreements did not qualify as ‘business of insurance’ and were instead subject to federal anti-trust laws.51

Three years later, in Union Labor Life Ins. Co. v. Pireno the Court explicitly adopted Royal Drug’s construction of the term ‘business of insurance’, and for the first time formulated a three-part test which would consistently guide its jurisprudence both in anti-trust and ERISA preemption cases.52 As the court put it, “[I]n sum, Royal Drug identified three criteria relevant in determining whether a particular practice is part of the “business of insurance” exempted from the antitrust laws by 2(b): first, whether the practice has the effect of transferring or spreading a policyholder’s risk; second, whether the practice is an integral part of the policy relationship between the insurer and the insured; and third, whether the practice is limited to entities within the insurance industry.”53

Pireno like Royal Drug arose in the anti-trust context. Eleven years after the passage of ERISA, the Supreme Court in Metropolitan Life Ins. Co. v. Massachusetts, for the first time interpreted ERISA’s preemption provisions as they related to state regulation of insurance.54 At issue in Metropolitan Life was a Massachusetts law that mandated insurance companies to provide mental health benefits in certain insurance policies.55 In deciding that Massachusetts’ mandated benefit law fell within ERISA’s saving clause, the Court explicitly adopted the three-part McCarren-Ferguson test which had been previously used exclusively in the anti-trust cases.56

The McCarren-Ferguson test, however, was only employed to bolster the Court’s earlier conclusion that as a matter of common sense, mandated benefit laws regulated insurance within the meaning of the saving clause.57 As the Court explained in Rush Prudential HMO, Inc. v. Moran, in order to satisfy the ‘common sense’ test “a law must not just have an impact on the insurance industry, but must be specifically directed toward that industry.”58 Implicit in the inquiry is the requirement that the law regulate the main elements of the insurance contract which are “the spreading and underwriting of a policyholder’s risk.”59

After Metropolitan Life, the Court routinely began its preemption analysis by first inquiring whether the state law ‘related to’ an employee benefit plan.60 The Court then proceeded to ask whether as a matter of common sense, the state law regulated insurance.61 The Court then verified the result of this inquiry by employing the three-part McCarren-Ferguson test.62

This tacking-on of the ‘common sense’ test to the three-part McCarren-Ferguson inquiry, presented a rather awkward structure which resulted in confusion in lower courts. There initially developed a split in the circuits as to whether all three parts of the McCarren-Ferguson test needed to be met.63 In Unum Life the court declared that McCarren-Ferguson’s three factors only served as ‘checking points’ or ‘guide posts’64 and that satisfaction of all its parts was not required.65 The reason for the confusion seemed to be the fact that some of the McCarren-Ferguson factors replicated the common sense inquiry, and one could not be met without the other. For example, the third part of the McCarren-Ferguson inquiry (i.e. whether the practice was limited to entities within the insurance industry) seemed to be identical to the common sense test.66 In addition, as the Court noted in Rush Prudential “[t]he common-sense enquiry focuses on ‘primary elements of an insurance contract [which] are the spreading and underwriting of a policyholder’s risk.”67 This inquiry seemed to substantially overlap with the first part of the McCarren-Ferguson test which asked “whether the practice has the effect of transferring or spreading a policyholder’s risk”.68 Despite the similar inquiries of the common sense test and the first prong of the McCarren-Ferguson test, the Court both in Rush Prudential and in Unum Life declined to address the issue of whether the state law regulated a risk-spreading practice.69 The court’s refusal to discuss risk-spreading was especially troubling in light of the Court’s repeated identification of risk-spreading as “a critical determinant in identifying insurance.”70


“Any Willing Provider” Laws and Circuit Courts’ Preemption Analysis

The pre-Miller confusion at the appeals courts level as to whether the so called ‘Any Willing Provider’ (AWP) laws were preempted by ERISA, illustrated the unsatisfactory nature of the Supreme Court’s preemption jurisprudence.

At present, twenty-one states have a variation of AWP laws.71 A typical AWP law requires all health insurers to allow any willing provider who practices within the geographic area served by the insurance company and who meets the insurer’s general qualification requirements, to enter into contract with the insurer.72 There exists a substantial body of scholarly literature discussing merits and drawbacks of AWP laws.73 While the advocates argue that AWP statutes provide the patients with a greater choice of health-care providers, the opponents point out that AWP laws do so at the expense of undermining the fundamentals of managed care plans and driving up the cost of health care in the United States.74 Critics of AWP contend that a key element of controlling health care costs is the insurers’ ability to secure lower fees from providers by offering them the privilege of participation in a closed network. Forcing the insurers to open these provider networks removes the economic utility of managed care plans. The empirical evidence on the impact of AWP laws on health care costs, however, is inconclusive.75

Before Miller was decided, there existed a circuit split on the question of whether ERISA’s saving clause applied to AWP laws. The extent of the confusion in lower courts is illustrated by the fact that they disagreed on every single element of the test: appeals courts not only diverged as to whether the common sense inquiry was satisfied, but also were divided on each of the three parts of the McCarren-Ferguson test.

Thus, while the Eighth Circuit held that Arkansas’ APW failed the common sense inquiry because it did not specifically target the insurance industry76, the Fifth Circuit concluded that an early version of Texas’ AWP regulated insurance as a matter of common sense because it “directly regulate[d] the terms of health insurance policies” at issue.77

The Fourth and Eight Circuits disagreed as to whether AWP laws met the first prong of the McCarren-Ferguson test which asked whether the law resulted in transferring risk between the insured and the insurer.78 The Fourth Circuit noted that policyholders’ risk consisted of several components, including the types of illness that the insurance policy covered, and the cost of treatment.79 Before AWP was enacted a policyholder who obtained medical care from a nonparticipating provider, had to shoulder out-of-pocket costs without being reimbursed by the insurance company.80 By opening up the provider network, AWP transferred the risk of paying this out-of-pocket cost from the insured to the insurer, who in turn spread it among all the policyholders.81 This argument was supported by the Fifth Circuit.82 The Eighth Circuit, however, held that AWP did not spread risk.83

The Eighth and the Fourth circuits also diverged as to whether AWP laws affected an integral part of the insurance policy relationship and thus met the second part of the McCarren-Ferguson inquiry.84 The Fourth Circuit based its conclusion that Virginia’s AWP statute met the second prong of the McCarren-Ferguson test on the argument that the law regulated treatment and cost of health insurance and these were integral parts the insured-insurer relationship.85 In Fourth Circuit’s view, mandatory-provider regulations were conceptually similar to mandatory-benefit laws in Metropolitan Life and thus, were immune from ERISA preemption.86 Again, the Eight Circuit disagreed and held that Arkansas AWP was not an integral part of the policy relationship between the insurer and the insured.87 The Eight Circuit supported this conclusion by noting that instead of defining the insurer-insured relationship, AWP defined the relationship between the insurer and a third party (i.e. the provider).88

The third prong of the McCarren-Ferguson test required that the state statute be limited to the insurance industry.89 The Fifth and the Sixth Circuits disagreed as to whether AWP statutes which included self-insured plans within their definition of insurer, satisfied this part of the test.90 The Sixth Circuit held that self-insurers engaged in the business of insurance, while the Fifth Circuit concluded that they did not.91


The Miller  Decision

The prevailing confusion among the circuits prompted the Supreme Court to grant certiorari in order to resolve the conflict. The AWP law at issue in Miller was a pair of Kentucky statutes which applied to health care providers and chiropractors. The first statute stated: “[a] health insurer shall not discriminate against any provider who is located within the geographic coverage area of the health benefit plan and who is willing to meet the terms and conditions for participation established by the health insurer, including the Kentucky state Medicaid program and Medicaid partnerships.”92 Another similarly worded AWP statute applied specifically to chiropractors.93

In order to control the quality and cost of health care delivery, several HMOs and a Kentucky-based association of HMOs, had created exclusive ‘provider networks’ by entering into contracts with doctors, hospitals and other health care providers.94 In exchange for furnishing discounted rates, the providers enjoyed a higher volume of patient traffic than non-network providers.95 Kentucky’s AWP, which required an open network of providers, threatened the viability of these closed networks and prompted the HMOs to bring suit against the Commissioner of the Kentucky Department of Insurance in order to stop the AWP’s enforcement.96 The trial court granted summary judgment in favor of the defendant and held that Kentucky’s statute was exempt from preemption, because it regulated insurance within the meaning of ERISA’s saving clause.97 On appeal, a divided panel of the Sixth Circuit affirmed.98 Seeking certiorari from the Supreme Court, the parties agreed that Kentucky’s AWP ‘related to’ ERISA-covered health plans99 and the question presented to the Court was formulated as follows: “Are state ‘Any Willing Provider’ statutes preempted by ERISA, or are they saved from preemption because they are laws which regulate insurance?”100

During the oral arguments, justices focused on several points which shed light on their written opinion.101 Probably the most significant of these was the issue of risk spreading.102 In its prior decisions, the Court had described risk-spreading as “an indispensable characteristic”103 and a “feature distinctive of” insurance.104 As such, the Court had identified risk-spreading as both a subject of the common sense test and more specifically, the focus of the first prong of the McCarren-Ferguson inquiry.105

In finding that Kentucky’s AWP spread risk, the Sixth Circuit had relied on the reasoning of the Fourth Circuit in Stuart Circle that AWP “spread the cost component of the policyholder’s risk among all the insureds, instead of requiring the policyholder to shoulder all or part of this cost when seeking care or treatment from an excluded doctor or hospital of his or her choice.”106 The Sixth Circuit, however, had qualified its support of this argument by noting that whether Kentucky’s AWP law spread risk was “certainly a debatable issue.”107 HMOs countered by pointing to the Supreme Court’s decision in Royal Drug, which on essentially similar facts had found that insurer-provider contracts did not spread risk.108 The Court in Royal Drug had warned against confusing the insurer-provider agreements with the insurer-insured contracts.109 The former did not spread insurer’s risk, because they did not affect the promise that the insurer had made to the policyholders: as long as the insured was able to get his medications for a $2 payment, he was “basically unconcerned with arrangements made between [insurer] and participating pharmacies.110 Instead of spreading the risk among many policyholders, the bulk purchase arrangements between insurer and pharmacies had the effect of reducing the risk that the insurer would pay a higher sum for the medications.111

During the oral arguments, the justices displayed their skepticism as to whether this crucial element in the definition of ‘insurance’ was present in Kentucky’s AWP statute. Justice O’Connor, for instance, pointed out that although Kentucky’s law by sanctioning open provider networks, afforded the policyholders a greater choice among health care professionals, this was simply a ‘practical benefit’ to the insured, and its connection to risk-spreading was unclear.112

The Solicitor General, whose office had filed an amicus brief in support of the Kentucky’s position, advanced a more sophisticated argument as to why risk-spreading requirement was satisfied. The brief pointed out that risk-spreading could only be accomplished if the insurance contract itself was carried out.113 Kentucky’s AWP statute conditioned the performance of the insurance contract on the requirement that the insured be given access to an open network of health care providers.114 The brief concluded, therefore, that since the imposition of such a requirement made risk-spreading possible in the first place, Kentucky’s AWP met the first prong of the McCarren-Ferguson test.115 Under questioning during the oral arguments, Associate Solicitor, Allen Feldman, conceded that Kentucky’s statute did not in fact spread risk, and instead was a “regulation of the goods or services that an insurer provide[d]”.116 One of the justices, however, reformulated Solicitor General’s argument as follows: “So, [you’re] saying the first McCarran-Ferguson factor includes a provision that determines the way the insurer manages the risk, even though it may not affect the risk as between the insurer and the insured.”117 In other words, in order for the Solicitor General’s argument to be accepted, the Court had to abandon its traditional inquiry into risk-spreading, and instead focus on the effect of the state statute on the overall risk management by the insurer. As it will be seen below, this is exactly what the Court did.

Another indication that the Court’s traditional saving clause analysis was in jeopardy, surfaced during the oral arguments, when the justices expressed their dissatisfaction about the effectiveness of the common sense test. The transcript reads as follows:

QUESTION: [Justice Ginsburg] –I’ll tell you frankly what my problem is. I read the Sixth Circuit opinion, I said, yes, that makes common sense, and I read Judge Kennedy’s dissenting opinion and said, yes, that’s common sense, too, so what–
QUESTION: These–these are rational judges on both sides, they both made good arguments, and they both conformed to some sense of what goes on in the real world, so what is the common sense test?

QUESTION: [Justice Scalia] And I don’t like the, you know, common sense test, I know it when I see it. What I worry about, the–the common sense test is that we will approve those things that we like, and disapprove those things that we don’t like. I mean, who likes a private antitrust arrangement that–that limits choice, so you just say, common sense, that’s not the business of insurance, and who doesn’t like something that enables–enables the insureds to–to have a greater selection in–in doctors, so we say, common sense says, that is the business of insurance. I–I don’t trust common sense.

Justice Scalia, of course, was also the author of the Miller opinion.120 Writing for a unanimous Court, he acknowledged that the two-part inquiry which had guided the Supreme Court’s saving clause analysis in the past, had resulted in much confusion in the lower courts.121 He then announced “a clean break from the McCarran-Ferguson factors” in favor of a new two-part test.122 The first prong of the new test asked whether the state law at issue was “specifically directed toward entities engaged in insurance.”123 If this inquiry was satisfied, the state law had to meet another condition: it must “substantially affect the risk pooling arrangement between the insurer and the insured.”124 The Court held that Kentucky’s AWP was saved from preemption because it met both requirements.125

Without explicitly stating it, in addition to the McCarren-Ferguson test, the Court had made a ‘clean break’ from the ‘common sense’ inquiry as well. In order to begin its analysis of the Kentucky’s statute, therefore, the Court had to go back to the drawing board by examining the language of ERISA’s saving clause. The operative language in that clause exempted from preemption, any state law that “regulates insurance”.126 In order to distinguish insurance from other practices, the Court in Royal Drug had identified risk-spreading as its “indispensable characteristic”.127 Now, however, the Court simply discarded risk-spreading as an identifying characteristic of insurance.128 Instead, the Court required that state laws which were specifically directed toward entities engaged in insurance, must also substantially affect the risk pooling arrangement.

In order to appreciate the breadth of the category of entities which after Miller may be regulated by states’ insurance laws, one may examine the language of Kentucky’s AWP statute. That statute defined the term ‘insurer’ as “any insurance company; health maintenance organization; self-insurer or multiple employer welfare arrangement not exempt from state regulation by ERISA; provider-sponsored integrated health delivery network; self-insured employer-organized association, or nonprofit hospital, medical- surgical, dental, or health service corporation authorized to transact health insurance business in Kentucky.”129 In addition to ‘insurers’, Kentucky’s AWP also applied to HMOs that provided administrative services to self-insured plans.130

Examining the broad reach of Kentucky’s AWP laws, and using the risk-spreading standard in order to identify ‘insurance’, the dissenting Sixth Circuit judge had stated that “although HMOs, [Health Service Corporations], and Insurance Companies may accept risk in some situations, as third party administrators they would merely be contracting to handle paperwork and plan administration for a self-insured ERISA plan.”131 The dissent, therefore, concluded that Kentucky’s AWP was not directed at the business of insurance, as “no insurance was involved.”132

The Miller Court disagreed with this conclusion. As to self-insured plans, Justice Scalia pointed out that the AWP’s application to these entities did not forfeit its status as a law which ‘regulated insurance’ because self-insured plans engaged in the “same sort of risk pooling arrangements as separate entities that provide[d] insurance to an employee benefit plan.”133 Turning to noninsuring HMOs whose only connection with insurance was to provide administrative services to self-insured plans, Justice Scalia noted that these too fell within the definition of insurance for the purposes of ERISA’s saving clause.134 And even if they did not, Rush Prudential allowed some overbreadth of the state law regulating insurance.135

It is worth noting that in order to determine whether Kentucky self-insurers satisfied the first prong of the new test- that is whether they were ‘entities engaged in insurance’- the Court appears to use the ‘risk pooling’ standard, which is the second prong of the test. If the first part of the new Miller test is indeed conceptually independent from its second part, the question still remains as how to distinguish ‘entities engaged in insurance’ from those that are not so engaged. Before Miller, the risk-spreading standard was employed for exactly this purpose. After Miller, the answer to that question is unclear.

But even having used ‘risk pooling’ standard in order to bring self-insurers within the ambit of ‘entities engaged in insurance’, it is not clear how noninsuring HMOs that, as the dissenting Sixth Circuit judge pointed out, only handled paperwork and plan administration for self-insured ERISA plans, may be included within ‘entities engaged in insurance’. Although, state regulation of these noninsuring entities may arguably affect the overall risk pooling arrangements of insurers, it is hard to see how these entities themselves engage in insurance. One consequence of Miller’s extraordinary broad definition of ‘insurance’ is likely to be state statutes regulating entities with only the most attenuated connection with insurance industry. The states’ attempts to indirectly regulate these entities in the guise of insurance regulation, will likely spawn much litigation, with the noninsuring entities arguing that they can not be properly categorized as ‘entities engaged in insurance’.

In Miller, the HMOs’ main argument, however, was not that Kentucky’s AWP was overinclusive because it applied to both self-insurers and noninsuring plan administrators. HMOs instead insisted that on essentially similar facts, the Court in Royal Drug had found that insurer-provider contracts did not qualify as ‘business of insurance’.136 Justice Scalia answered this argument by pointing out that Royal Drug arose in anti-trust and not ERISA context.137 The ‘business of insurance’ standard of the McCarren-Ferguson Act, therefore, should not be applied to the present case, since ERISA only required that the state law ‘regulate insurance’.138 Under the new Miller test, it no longer mattered whether the insurer-provider contracts, which were the subject of Kentucky’s AWP, qualified as ‘business of insurance’.139 The Court analogized Kentucky’s regulation of insurer-provider contracts to state regulation of practice of law through CLE requirements.140 Insurer-provider contracts did not have to qualify as ‘business of insurance’, in the same way that attending CLE classes did not have to qualify as practice of law.141 After Miller, therefore, it appears that as long as the state regulation substantially affects the insurer’s risk pooling arrangements, the immediate subject of the regulation need not qualify as insurance practice.


What Has Miller  Wrought?

The most important aspect of Miller is its extraordinary broad interpretation of the term ‘regulates insurance’ in ERISA’s saving clause. This expanded interpretation, provides the states with a free hand to regulate a broad range of activities within their jurisdiction. As discussed above, after Miller, not only the regulated entity need not be the insurer, but also the regulated activity need no longer qualify as ‘business of insurance’.

The Supreme Court justified its abandonment of the McCarren-Ferguson factors by pointing out that consideration of those factors had “added little to the relevant analysis” of ERISA preemption.142 The Court found this “unsurprising, since the statutory language of [the saving clause] differs substantially from that of the McCarran-Ferguson Act. Rather than concerning itself with whether certain practices constitute “[t]he business of insurance,” or whether a state law was “enacted … for the purpose of regulating the business of insurance,” (Court’s emphasis), [ERISA’s saving clause] asks merely whether a state law is a “law … which regulates insurance, banking, or securities.”143

It is noteworthy that this ‘substantial’ difference between the saving clause and McCarren-Ferguson Act, had not prevented the Court in the past from consistently using the McCarren-Ferguson factors as a guide in its ERISA preemption analysis. In fact, in Metropolitan Life the Court had found occasion to note that “[t]he ERISA saving clause, with its similarly worded protection of “any law of any State which regulates insurance,” appears to have been designed to preserve the McCarran-Ferguson Act’s reservation of the business of insurance to the States. The saving clause and the McCarran-Ferguson Act serve the same federal policy and utilize similar language to define what is left to the States.”144

As discussed in the first part of this article, the appeals courts’ attempts to apply the McCarren-Ferguson factors to the saving clause analysis, had led to inconsistent results and confusion. Risk spreading, for example, had proven to be less than a reliable guide in determining whether a certain practice qualified as ‘business of insurance’. Having said this, however, it is difficult to see how the adoption of the new test would resolve the matter. The new ‘substantial effect on risk pooling arrangement’ standard will probably prove to be as difficult to interpret and apply as the old ‘risk spreading’ test. Thus, several post-Miller courts have concluded that the required analysis under the second part of the Miller test is essentially similar to the old ‘risk-spreading’ inquiry.145 For example, in Morales-Ceballos v. First Unum Life Ins. Co. of Am., the court apparently equated Miller’s ‘substantial effect on risk pooling arrangement’ standard with the McCarren-Ferguson’s ‘risk spreading’ test, and concluded that since Pennsylvania’s bad faith statute did not spread or transfer risk, it did not substantially affect risk pooling arrangement either.146 If courts conclude, as some already have, that the new Miller standard does not do much to clarify the pre-Miller confusion, one might expect that ERISA preemption will continue to be the source of much litigation.

The same might be said about the first part of the new test. Identifying entities that ‘engage in insurance’ will certainly prove a difficult task, especially in light of the fact that the Miller Court included noninsuring service providers to self-insurers as such entities.147 How remote the connection of an entity to the insurer must become, before it no longer qualifies as an ‘entity engaged in insurance’? That question promises to be widely litigated in the future.

What is certain, however, is that Miller greatly expands the regulatory reach of the states. In addition to traditional areas of state insurance regulation, such as laws governing insurer solvency, marketing, disclosure, underwriting, premiums, and other typical insurance activities148, Miller opens the door to state regulation of other activities that up to now have not been commonly understood as insurance practice.

The likely effect of Miller’s upholding of AWP laws on health care costs in the United States, however, should not be exaggerated. As mentioned above, the evidence concerning the impact of AWP laws on health care costs is inconclusive.149 The contention that the Miller decision will cause an upheaval in health care delivery in the United states seems to be exaggerated. Thus, Karen Ignagni, president of the American Association of Health Plans has stated that the Miller ruling “changes little in the current health care delivery system.”150

Supreme Court’s substantial broadening of the reach of the saving clause in Miller, serves as a corollary to the Court’s increasingly narrow reading of the ‘relate to’ clause in 514(a).151 The cumulative effect of a narrow interpretation of the preemption clause in 514(a), and a broad reading of ERISA’s saving clause, promises to result in states’ unprecedented regulatory power over the health care of their citizens.



1Kentucky Ass’n of Health Plans, Inc. v. Miller, 123 S.Ct. 1471, 155 L.Ed.2d 468 (2003).

2 29 U.S.C. §§ 1001.

3 See Miller, 123 S.Ct. at 1477-8.

4Id. at 1479.

5 Miller, 123 S.Ct. at 1478.

6 See, e.g. California Div. Of Labor Standards Enforcement v. Dillingham Constr., N.A. Inc., 519 U.S. 316, 335-336 (1997). (Scalia, J., concurring ) (noting that the Court had accepted 14 preemption cases since ERISA’s inception in an effort to bring clarity to the field without much success.)

7 ERISA 514(a)

8 514(b)(2)(B)

9 Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724, 740.

10Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355, 364 (2002), See also, New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U.S. 645, 656 (1995) (referring to the unhelpful language of the ‘relate to’ clause)

11 Metropolitan Life, 471 U.S. at 739

12 Prudential Ins. Co. of Am. v. National Park Med. Ctr., Inc., 154 F.3d 812, 818 (8th Cr. 1998).

13 Rush Prudential, 536 U.S. at 365.

14 514(a).

15 Shaw, 463 U.S. at 98.

16 Metropolitan Life, 471 U.S. at 739-40. See also Rush Prudential, 536 U.S. at 365. (ERISA’s preemption language seems “simultaneously to preempt everything and hardly anything.”)

17 Article I, Section 8.

18 United States v. Darby, 312 U.S. 100, 118 (1941) (declaring that “the power of Congress over interstate commerce is not confined to the regulation of commerce among the states. It extends to those activities intrastate which so affect interstate commerce or the exercise of the power of Congress over it as to make regulation of them appropriate means to the attainment of a legitimate end.”)

19 317 U.S. 111 (1942).

20 See e.g. United States v. South-Eastern Underwriters Assn., 322 U.S. 533 (1944).

21 Paul v. Virginia, 8 Wall. 168, (1869).

22 Id. at 183.

23Id. at 185.

24 SEC v. National Securities, Inc., 393 U.S., 453, 458 (1969); See also, Hooper v. People of State of California, 155 U.S. 648, 655 (1895) (“The business of insurance is not commerce.”); New York Life Insurance Company v. Deer Lodge County, 231 U.S. 495, 510 (citing Paul v. Virginia and stating that “contracts of insurance are not commerce at all, neither state nor interstate.)

25 United States v. South-Eastern Underwriters Assn., 322 U.S. 533 (1944).

26 Id. at 535.

27 Id. at 534.

28Id. at 535.

29 Id. at 539.

30 Id. at 557-62

31 National Securities, Inc., 393 U.S. at 458.

32 59 Stat. 33 (1945), 15 U.S.C. 1011.

33 National Securities, Inc., 393 U.S. at 459.

34 59 Stat. 33-34. Section 2(b) provides: “No Act of Congress shall be construed to invalidate, impair, or supersede any law enacted by any State for the purpose of regulating the business of insurance, or which imposes a fee or tax upon such business, unless such Act specifically relates to the business of insurance: Provided, That after June 30, 1948, the Act of July 2, 1890, as amended, known as the Sherman Act, and the Act of October 15, 1914, as amended, known as the Clayton Act, and the Act of September 26, 1914, known as the Federal Trade Commission Act, as amended, shall be applicable to the business of insurance to the extent that such business is not regulated by State law.” Id.

35 National Securities, Inc., 393 U.S. at 459.

36 Id. at 459-60 (“Insurance companies may do many things which are subject to paramount federal regulation; only when they are engaged in the “business of insurance” does the statute apply.”) (Emphasis in original.)

37 Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205 (1979).

38 Id. at 209.

39 See Id.

40 See Id.

41 Id.

42 Id.

43Id. at 207.

44 Id. at 208.

45 Id. at 211.

46Id. at 213-14.

47Id at 215-16, quoting National Securities, Inc., 393 U.S. at 460.

48 Id. at 216.

49 Id. at 231.

50 Id.

51 Id. at 214.

52 Union Labor Life Ins. Co. v. Pireno, 458 U.S. 119 (1982).

53 Id. at 129.

54 Metropolitan Life, 471 U.S. at 724.

55 Id. at 727.

56 Id. at 742-3

57 Id. 740.

58 Rush Prudential, 536 U.S. at 366.

59 Id., quoting Royal Drug, 440 U.S. at 211.

60 Rush Prudential, 536 U.S. at 365; Unum Life Ins. Co. of Am. v. Ward, 526 U.S. 358, 367 (1999)

61 Unum Life, 526 U.S. at 367-8 (1999).

62 Id.

63 See e.g. Cigna Healthplan of La., Inc. v. Louisiana ex rel. Ieyoub, 82 F.3d 642, 650 (5th Cir. 1996) (satisfaction of all three prongs required); Cisneros v. UNUM Life Ins. Co. of Am., 134 F.3d 939, 946 (9th Cir. 1998) (all three parts need not be met).

64 Unum Life, 526 U.S. 374.

65 Id. at 373.

66 See e.g. Rush Prudential, 536 U.S. at 374 (concluding that third factor was satisfied for “many of the same reasons that the law passes the common sense test”); Unum Life, 526 U.S. at 375 (finding that the third factor was “well met” for the same reasons that the common sense test was satisfied.) The Court in Miller explicitly acknowledged this duplication by noting that the third McCarran-Ferguson factor was “a mere repetition of the prior inquiry into whether a state law is “specifically directed toward” the insurance industry under the “common-sense view.” Miller, 123 S.Ct. at 1479.

67 Rush Prudential, 536 U.S. at 366, quoting Royal Drug, 440 U.S. at 211.

68Metropolitan Life, 471 U.S. at 743.

69 See Rush Prudential, 536 U.S. at 373; Unum Life, 526 U.S. at 374.

70 Royal Drug, 440 U.S. at 213; Rush Prudential, 536 U.S. at 367 (stating that risk spreading and underwriting is a “feature distinctive of insurance”); SEC v. Variable Annuity Life Ins. Co. of America, 359 U.S. 65, 73 (underwriting of risk is an “earmark of insurance”)

71 See Health Policy Tracking Service, National Conference of State Legislatures, Any Willing Provider, , (visited July 5, 2003).

72 See, e.g. Ala. Code § 27-45-3 (The Pharmacy Act); Ala. Code § 27-19A-3 (The Dental Act); Del. Code Ann. Tit 18, § 7303b; Ga. Code Ann. § 33-20-16; Idaho Code § 41-3927; Ind. Code Ann. §27-8-11-3(c); Ky.Rev.Stat. Ann. § 304.17A-270 (West 2001); La.Rev.Stat. Ann. §22:1214(15) as amended by LA Legis. 129 (2003) (West); Minn. Stat. § 62Q.095; Miss. Code Ann. § 83-9-6(3); N.H. Rev.Sat. Ann. §420-B:12(V); N.D. Cent. Code § 26.1-36-12.2.(1)(c); S.D. Codified Laws Ann. §58-18-37; Utah Code Ann. § 31A-22-617(7)(b) as amended by Utah Laws Ch. 131 (H.B. 165) (2003) (West); Va. Code § 38.2-4209(c); Wis. Stat. Ann § 628.36(2)(b)1.

While some AWP statutes apply only to a limited category of providers, other statutes have a broader application. Thus, the following state statutes apply only to pharmacists: Alabama, Connecticut, Delaware, Florida, Illinois, Massachusetts, Mississippi, New Hampshire, New Jersey, North Carolina, North Dakota, South Dakota, Tennessee, Texas and Wisconsin.

The following states have broader AWP statues: Georgia (apply only to Blue Cross Blue Shields contracts or rural care providers), Idaho (all qualified health care providers), Illinois (limited to participating providers with public aid); Indiana, Kentucky, Minnesota (any qualified independent health care provider), Wyoming (any qualified Wyoming provider). See, National Conference of State Legislatures, supra note 71.

73 See e.g., Linda H. Aiken & William M. Sage, Staffing National Health Care Reform: A Role for Advanced Practice Nurses, 26 AKRON L. REV. 187; James E. Bowers, Health Care Reform and Antitrust: An Insurer’s Perspective, 694 P.L.I. COMM. 71; Arnold Celnicker, A Competitive Analysis of Most Favored Nations Clauses in Contracts Between Health Care Providers and Insurers, 69 N.C. L. REV. 863; James W. Childs, Jr. You May Be Willing, But Are You Able?: A Critical Analysis of “Any Willing Provider” Legislation, 27 Cumb. L. Rev. 199; David L. Meyer & Charles F. Rule Health Care Collaboration Does Not Require Substantive Antitrust Reform, 29 WAKE FOREST L. REV., 169; Marc A.

Rodwin, Consumer Protection and Managed Care: Issues, Reform Proposals, and Trade-Offs, 32 HOUS. L. REV. 1319, 1321-35.

74 See e.g., Dianne McCarthy, Narrowing Provider Choice: Any Willing Provider After New York Blue Cross v. Travelers, 223 Am. J.L. & Med. 97; Bruce W. Karrh, Health Care Reform in the United States, 13 DEL. LAW. 17.

75 Childs, supra at 212 (“[T]here is some evidence that health care costs have not risen in those states which have enacted AWP legislation”). See also Pat Butler, Kentucky’s “Any Willing Provider” Law and ERISA, National Academy for State Health Policy, June 2003. (visited on July 15, 2003) (Reviewing the literature on AWP’s effect on health care costs and concluding that “independent research

evidence of the effects of [AWP] laws -controlling for a variety of confounding factors- is mixed.”)

76 National Park, 154 F.3d at 829.

77Texas Pharmacy Ass’n v. Prudential Ins. Co. of Am. 105 F.3d 1035,1040 (5th Cir. 1997).

78 Compare Stuart Circle Hosp. Corp. v. Aetna Health Management, 995 F.2d 500, 503 (4th Cir. 1993) (concluding that Virginia’s AWP statute transferred or spread the policyholder’s risk) with National Park, 154 F.3d 812, 829 (Arkansas AWP “plainly does not have the effect of transferring or spreading the policyholder’s risk”)

79 Stuart Circle 995 F.2d at 503.

80 Id.

81 Id.

82Texas Pharmacy Ass’n 105 F.3d at 1041.

83National Park, 154 F.3d at 829.

84 Compare Stuart Circle, 995 F.2d at 503, (finding that Virginia’s statute satisfied the second part of the McCarren-Ferguson test.) with National Park, 154 F.3d at 829 (concluding that Arkansas AWP was not an integral part of the policy relationship between the insurer and the insured.)

85 Stuart Circle, 995 F.2d at 503.

86 Id.

87 National Park, 154 F.3d at 829.

88 Id. at 830.

89 Union Labor Life Ins. Co. 458 U.S. at 129.

90 Compare Cigna Healthplan of La., Inc. v. Louisiana ex rel. Ieyoub, 82 F.3d 642, 650 (5th Cir. 1996) (holding that Louisiana’s AWP failed the 3rd prong of the McCarren-Ferguson test because the statute was not limited to entities within the insurance industry) with Kentucky Ass’n of Health Plans, Inc. v. Nichols, 227 F.3d 352, 371 (6th Circuit, 2000) (concluding that “entities such as HMOs and self-insurers are engaged in the business of insurance.”). Although Louisiana’s statute in Cigna Healthplan did not use the expression ‘self-insurer’ it included the terms ‘self-funded organizations’ and ‘self-funded trusts or programs.’ La.Rev.Stat.Ann. 40:2202(3)(a) & (b).

91 Id.

92 Ky.Rev.Stat. Ann. § 304.17A-270 (West 2001)

93 § 304.17A-171(2)

94 Pet. Br. 1.

95 Pet. Br. 1-2.

96 Nichols, 227 F.3d at 355.

97 Id.

98 Id. at 372.

99 11 NO. 1 ERISA Litig. Rep. (April, 2003)

100 Pet. Br. i. (internal quotation marks removed)

101 2003 WL 145394, U.S. Oral.Arg.,2003.

102 Id. at 11, 18, 26, 27, 40-49.

103 Royal Drug, 440 U.S. at 212.

104 Rush Prudential, 536 U.S. at 367.

105 See supra note 67-68.

106 Nichols, 227 F.3d at 369.

107 Id. at 368.

108 Royal Drug, 440 U.S. at 213-14.

109Id. at 213.

110 Id. at 214.

111 Id. at 215.

112 2003 WL 145394 (U.S. Oral.Arg.,2003) at 42-3.

113 2002 WL 31477704 (U.S.Amicus.Brief,2002), at 19, (citing Department of Treasury v. Fabe, 508 U.S. 491, 503-4 (1993) for the proposition that “[w]ithout performance of the insurance policy, there is no risk transfer at all.”)

114 Id.


116 2003 WL 145394 (U.S. Oral.Arg.,2003) at 46.

117 Id. at 44-5.

118 Id. at 38.

119 Id. at 41-2.

120 Miller, 123 S.Ct. at 1473.

121 See Id. at 1478-9.

122 Id. at 1479.

123 Id.

124 Id.


126 ERISA § 514(b)(2)(A).

127Royal Drug, 440 U.S. at 212.

128 The argument that the abandonment of risk-spreading standard was justified because it was originally adopted by the Court in the pre-ERISA, anti-trust context is unavailing, since risk-spreading was consistently identified by the Court as an important identifier of ‘insurance’, and not of ‘business of insurance’, which was a term used in the McCarren-Ferguson Act. See supra note 70.

129 Ky.Rev.Stat. Ann. § 304.17A-005(23).

130 Miller, 123 S.Ct. at 1476.

31 Nichols, 227 F.3d. at 374.

132 Id. at 375.

133 Miller, 123 S.Ct. at 1476.

134 Id.

135 Id.

136 Royal Drug, 440 U.S. at 214.

137 Miller, 123 S.Ct. at 1476-77.

138 See Id.

139 See id. at 1477.

140 Id.

141 Id.

142 Id. at 1478.

143 Id. References omitted.

144 Metropolitan Life, 471 U.S. at 744.

145 Anderson v. Business Men’s Assur. Co. 2003 U.S. Dist. Lexis 9833 at 23-4 (E.D. La.), (Noting that the preemption analysis under the new Miller test “remains substantially the same” as under the old McCarren-Ferguson inquiry”); Singh v. Prudential Health Care Plan Inc., 2003 U.S. App. Lexis 13499 at 18 (4th Cir. 2003) (Stating that Miller “did not work any fundamental change in the substance of the saving-clause analysis”); But see, Anderson v. Cont’l Case Co., 258 F.Supp. 2d 1127, 1131(E.D. Calif. 2003). (stating that the second prong of the Miller test is a “departure from the previous case law”).

146 2003 U.S. Dist. Lexis 9801 at 8 (E.D. Pa), See also, McGuigan v. Reliance Std. Life Ins. Co., 256 F.Supp.2d 345, 348 (E.D. Pa.) (Citing Pilot Life for the proposition that Mississippi’s bad faith statute failed the old risk-spreading test, and concluding that Pennsylvania’s bad faith statute, for the same reason, failed the second part of the new Miller test)

147 Miller, 123 S.Ct. at 1476.

148 See supra note 75.

149 Id.

150 Quoted in John Carroll, AWP Reimbursement Ruling May be More Than Meets Eye, Managed Care, May 2003.

151 Compare Shaw, 463 U.S. at 97 (adopting an expansive definition of the ‘relate to’ clause and concluding that a state law relates to a plan “if it has a connection with or reference to a plan”); Mackey v. Lanier Collections Agency & Services, Inc., 486 U.S. 825 (holding that Georgia’s garnishment statute was preempted because it expressly referred to ERISA plans, even though the statute specifically exempted such plans from garnishment); with Travelers Ins. Co., 514 U.S. 645, 655 (adopting a narrow definition of ‘relate to’ clause and stating that “if ‘relate to’ were taken to extend to the furthest stretch of its indeterminacy, then for all practical purposes pre-emption would never run its course, for ‘really, universally, relations stop nowhere’)

152 See supra note 75.

153 Id.

154Quoted in John Carroll, AWP Reimbursement Ruling May be More Than Meets Eye, Managed Care, May 2003.

155 Compare Shaw, 463 U.S. at 97 (adopting an expansive definition of the ‘relate to’ clause and concluding that a state law relates to a plan “if it has a connection with or reference to a plan”); Mackey v. Lanier Collections Agency & Services, Inc., 486 U.S. 825 (holding that Georgia’s garnishment statute was preempted because it expressly referred to ERISA plans, even though the statute specifically exempted such plans from garnishment); with Travelers Ins. Co., 514 U.S. 645, 655 (adopting a narrow definition of ‘relate to’ clause and stating that “if ‘relate to’ were taken to extend to the furthest stretch of its indeterminacy, then for all practical purposes pre-emption would never run its course, for ‘really, universally, relations stop nowhere’)

Need A Free Initial Consultation ?

Let us know how we can help!