Am. Drug Stores, Inc. v. Harvard Pilgrim Health Care, Inc., 973 F. Supp. 60 (D. Mass. 1997)

Richard A. Johnston, Brian E. Whiteley, James J. Nicklaus, Hale & Dorr, Howard Wayne, Anne Marie Anderson, Wayne, Richard, Hurwitz & McAloon, Boston, MA, for American Drug Stores, Inc., dba Osco Drug, Inc., Plaintiff.

John C. Kane, Theodore W. Ruger, Michelle Garvin, Ropes & Gray, Boston, MA, Daly D.E. Temchine, Epstein, Becker & Green, P.C., Washington, DC, Thomas I. Elkind, Epstein, Becker & Green, P.C., Boston, MA, for Defendants.

*61 Loretta M. Smith, Attorney General's Office, Jamie W. Katz, Assistant Attorney General, Boston, MA, for Commonwealth of Massachusetts, Intervenor-Plaintiff.

John C. Kane, Theodore W. Ruger, Michelle Garvin, Ropes & Gray, Boston, MA, Daly D.E. Temchine, Epstein, Becker & Green, P.C., Washington, DC, Thomas I. Elkind, Epstein, Becker & Green, P.C., Boston, MA, for Intervenor-Defendant.

LASKER, District Judge.

This case raises the question whether the Massachusetts "Pharmacy Freedom of Choice Any Willing Provider Act," Mass. Gen.Laws ch. 176D, § 3B, is preempted by the federal Employee Retirement Income Security Act, 29 U.S.C. §§ 1001-1461. Plaintiff American Drug Stores, Inc., d/b/a Osco Drug Stores ("Osco"), has brought this suit seeking to gain admission to the restricted pharmacy networks through which the defendants, Harvard Pilgrim Health Care ("HPHC") and the Prudential Insurance Company of America ("Prudential"), contract to supply their respective patient-customers. Osco argues that the defendants have excluded it from their pharmacy networks in violation of the Massachusetts Act. The defendants move for summary judgment on the grounds that the Act is preempted by ERISA. The Commonwealth of Massachusetts has intervened as a plaintiff to defend the validity of the Act.^[1]

For the reasons set forth below, the motion is denied.

The Act requires any "carrier^[2] that offers insureds^[3] a restricted pharmacy network^[4]" to comply with specified requirements "in soliciting, arranging, competitively bidding and contracting for such a network." Mass. Gen.Laws ch. 176D, § 3B. The Act provides further that retail pharmacies or associations of retail pharmacies which are not part of a carrier's restricted pharmacy network "shall nevertheless have the right to provide drug benefits to the carrier's insureds provided that such non-network pharmacies reach [certain specified] agreements with the carrier" concerning matters such as billing, reimbursement, eligibility determinations, protection of proprietary and confidential information, quality assurance and auditing. Id. The Act does not dictate terms of the relationship between carrier and pharmacy, but instead uses the agreement a carrier reaches with its network pharmacies as a benchmark against which to measure its relationships with non-network pharmacies. In particular, the Act precludes a carrier from "impos[ing] any agreements, terms or conditions on any non-network pharmacy ... which are more restrictive than those required of any pharmacy in the carrier's restricted pharmacy network." Id. The Act does allow a carrier to "impose a cost-sharing charge for the use of a non-network pharmacy not to exceed five percent more than the charge for using" a network pharmacy. Id.

The policy goals fostered by the Act include increasing patient access to and *62 choice of pharmacies, and protecting pharmacies which have been excluded from carriers' networks. The defendants challenge both the appropriateness of considering the purpose of this legislation and Osco's arguments as to the policies underlying the legislation, but neither challenge is tenable. In New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Insurance Co., 514 U.S. 645, 657-59, 115 S. Ct. 1671, 1678, 131 L. Ed. 2d 695 (1995), the Supreme Court examined "[b]oth the purpose and the effects" of the New York surcharge statutes which it upheld. The legislative purpose of a statute is relevant to a preemption analysis in part because it pertains to whether the law involves an area of traditional state concern. Moreover, while the defendants dispute the admissibility of affidavits Osco submitted as to the purpose of the Act, they offer no substantial challenge to Osco's claims as to the nature of the underlying policies. Because the purposes of the Act are clear from its title, language and effects, the challenged affidavits are unnecessary to this opinion. The Act, like other any willing provider statutes, serves the dual purposes of protecting providers and ensuring patients of greater access to services.^[5]

Both defendants are "carriers" under the terms of the Act. HPHC is a health maintenance organization which provides health care benefits to nearly 500,000 members in Massachusetts, approximately 70% of whom are entitled to services because of their membership in ERISA plans. A few of the 3,3000 ERISA plans with which HPHC contracts in Massachusetts are self-funded, uninsured plans to which HPHC provides administrative services only. Approximately 90% of HPHC members are entitled to prescription benefits; nearly 200,000 of these members receive their benefits pursuant to a contract between HPHC and the retail chain CVS.^[6] More than 60,000 Massachusetts residents receive health care services through Prudential. Approximately 90% of these individuals secure such benefits by virtue of their participation in ERISA plans. Slightly more than half of these members belong to self-funded, uninsured plans to which Prudential provides administrative services only. Approximately 75% of the ERISA plan members who receive services through Prudential are entitled to prescription benefits. While Prudential offers coverage with open provider structures, much of its pharmacy coverage is provided through restricted networks.^[7]

The parties dispute whether it is the ERISA plan or the carrier who actually chooses to use a restricted network. Osco and the Commonwealth argue that the defendants' contracts with ERISA plans demonstrate that the defendants make such decisions, and claim that ERISA plans do not in fact "care" whether restricted networks are used. The defendants maintain that ERISA plans make the decision whether to use restricted networks, and argue that the fact that ERISA plans contract for services which include restricted networks shows that the plans do "care" about the matter. Despite the energy the parties have devoted to this dispute, it is not germane to the ultimate question of whether a state law may prevent ERISA plans from contracting with carriers for use of restricted networks. It is clear that ERISA plans do contract with third parties for services which include restricted *63 pharmacy networks. Whether the ERISA plans truly "care" about having these networks, or whether they simply accept them as part of a total package put together by the carriers, does not affect this analysis.

After passage of the Act, Osco sought to participate in the networks offered by HPHC and Prudential. When HPHC and Prudential continued to exclude Osco from their networks, Osco brought this suit.

ERISA provides in relevant part that "the provisions of this subchapter ... shall supersede any and all State laws insofar as they may now or hereafter relate to any employee benefit plan." 29 U.S.C. § 1144(a). The defendants argue that the Act "relates to" ERISA plans, and is thus superseded by ERISA, because it "mandates employee benefit structures or administration."^[8] They explain that the Act "directly limits the structures through which [ERISA] plans may provide pharmacy benefits" by precluding ERISA plans from contracting with carriers for insurance or administrative services which incorporate restricted pharmacy networks. The defendants emphasize that ERISA preemption sweeps broadly in order to allow uniformity of plan administration and to minimize the administrative burdens facing ERISA plans by eliminating potentially conflicting state regulations. They claim that the Massachusetts Act is subject to preemption because it imposes regulatory demands which prevent ERISA plans from establishing nationally uniform administrative schemes.

The plaintiffs maintain that the Act is entitled to a strong presumption of validity because its focus on health care and the insurance industry reflects a legitimate state interest in an area of traditional state concern. They deny that the Act "mandates" employee benefit structures or administration, arguing that the Act does not affect the structure or administration of ERISA plans themselves, but only of the third party vendors which service the plans. The plaintiffs point out that ERISA plans need not comply with the statute, but may in fact use restricted pharmacy networks so long as they do not rely on a carrier to do so. The plaintiffs contend that the Act's connection with ERISA plans is "too tenuous, remote and peripheral" to be deemed "related to" the plans within the meaning of ERISA. They explain that the Act does not affect any of the activities central to plan administration such as eligibility determinations, benefit calculations, disbursements, record-keeping and fund management and stress the difference between administration of an ERISA plan and operation of a pharmacy network.

The defendants reply that the plaintiffs' argument mistakes the relations among plans, carriers and ERISA. They note that fiduciary relationships often exist between carriers and ERISA plans, and argue that the Act impermissibly purports to direct the conduct of ERISA fiduciaries in dealing with plans. They claim that the requirements of the Act may prevent carriers from fulfilling their fiduciary obligations under ERISA by preventing them from establishing restricted networks where they believe such networks are in the interest of the plan beneficiaries. The defendants also dispute the plaintiffs' claim that the Act concerns an area of traditional state regulation; they maintain that the Act does not regulate health care, but merely interferes with the contractual relationship between carrier and provider.

The parties also disagree as to whether, assuming the Act "relates to" ERISA plans, it is nonetheless saved from preemption as an insurance regulation pursuant to 29 U.S.C. § 1144(b) (2) (A). The defendants emphasize that the Act will affect many self-insured plans which rely on carriers for administrative services only. They argue first that application of the Act to these uninsured plans clearly cannot be a regulation of insurance. They claim further that the Act as a *64 whole cannot be considered an insurance regulation because it is not limited to entities within the insurance industry. The plaintiffs respond that all carriers regulated by the Act are entities within the insurance industry, and explain that the Act is saved from preemption because it concerns the business of insurance.

Finally, the parties disagree as to whether, if the Act is preempted, it should be preempted in its entirety or severed so as to continue to apply in situations which do not concern ERISA plans or their beneficiaries. The defendants argue that because the Act does not differentiate between insured and self-insured plans, or between ERISA plans and other sources of coverage, and because there is no reason to believe the Massachusetts legislature would have passed the Act knowing it would apply only to the small percentage of the population whose coverage is not tied to ERISA plans, the Act should be preempted in its entirety. The plaintiffs reply that the chapter in which the Act appears includes an express severability provision which states that

[i]f any provision of this chapter, or the application of such provision to any person or circumstances, shall be held invalid, the remainder of the chapter, and the application of such provision to person or circumstances other than those as to which it is held invalid, shall not be affected thereby.

Mass.Gen.Laws ch. 176D § 14. The plaintiffs argue further that the legislature intended the Act to be enforceable on behalf of Massachusetts residents covered by non-ERISA plans even if it was held unenforceable on behalf of residents covered by ERISA plans.

ERISA "supersede[s] any and all State laws insofar as they may now or hereafter relate to any employee benefit plan." 29 U.S.C. § 1144(a). State laws which "regulate insurance" are saved from preemption, 29 U.S.C. § 1144(b) (2) (A), but such laws cannot be applied to regulate self-insured ERISA plans because ERISA plans may not "be deemed" insurance companies or "engaged in the business of insurance," 29 U.S.C. § 1144(b) (2) (B).

The Supreme Court has explained that a state law "relates to" an ERISA plan "if it has a connection with or reference to such a plan." Shaw v. Delta Air Lines, Inc., 463 U.S. 85, 97, 103 S. Ct. 2890, 2900, 77 L. Ed. 2d 490 (1983). There is no suggestion here that the Massachusetts Act "refers to" ERISA plans, so the question is whether the Act has a "connection with" such plans. A state law may be subject to preemption even if it "is not specifically designed to affect [ERISA] plans, or the effect is only indirect." Ingersoll-Rand Co. v. McClendon, 498 U.S. 133, 139, 111 S. Ct. 478, 483, 112 L. Ed. 2d 474 (1990). Nonetheless, it has always been clear that preemption has limits and does not extend to any law which "has only a `tenuous, remote or peripheral' connection with covered plans." District of Columbia v. Greater Washington Board of Trade, 506 U.S. 125, 130 n. 1, 113 S. Ct. 580, 583 n. 1, 121 L. Ed. 2d 513 (1992) (citing Shaw). Preemption often stops short of state laws of general applicability, which "impose some burdens on the administration of ERISA plans but nevertheless do not relate to them within the meaning of the governing statute." DeBuono v. NYSA-ILA Medical and Clinical Services Fund, ___ U.S. ___, ___, 117 S. Ct. 1747, 1752, 138 L. Ed. 2d 21 (1997). In this, as in all preemption analyses, "the purpose of Congress is the ultimate touchstone." Ingersoll-Rand, 498 U.S. at 138, 111 S. Ct. at 482.

While the law in this area has been in flux since the Supreme Court commented on the potentially infinite reach of "relations" and "connections" and observed that "prior attempt[s] to construe the phrase `relate to' [do] not give us much help drawing the line," New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Insurance Co., 514 U.S. 645, 655, 115 S. Ct. 1671, 1677, 131 L. Ed. 2d 695 (1995), two recent Supreme Court cases DeBuono and California Division of Labor Standards Enforcement v. Dillingham Construction, ___ U.S. ___, 117 S. Ct. 832, 136 L. Ed. 2d 791 (1997) affirm and elaborate on the approach set forth in Travelers. The Supreme Court has emphasized that while it is not disavowing its earlier *65 decisions, those were clear cases which did not test the limits of, and do not reveal the proper way to analyze, ERISA preemption.

The most recent Supreme Court pronouncements recognize the impossibility of extracting meaningful limits from the statutory language alone, and reject such "uncritical literalism" in favor of analyzing "the objectives of the ERISA statute as a guide to the scope of the state law that Congress understood would survive." DeBuono, ___ U.S. at ___, 117 S. Ct. at 1751; Travelers, 514 U.S. at 656, 115 S. Ct. at 1677. These opinions also recognize the importance of analyzing the purposes of the challenged legislation and the nature of the effect of such legislation on ERISA plans. Dillingham, ___ U.S. at ___, 117 S. Ct. at 838; Travelers, 514 U.S. at 658, 115 S. Ct. at 1678.

The Supreme Court has emphasized the importance of "the starting presumption that Congress does not intend to supplant state law." Travelers, 514 U.S. at 655, 115 S. Ct. at 1676 ("[W]e have worked on the `assumption that the historic police powers of the States were not to be superseded by the Federal Act unless that was the clear and manifest purpose of Congress.'") (citation omitted).

In the last five years, several courts have considered whether ERISA preempts any willing provider statutes which are somewhat similar to the one at issue here. Because these decisions were made without the benefit of the most recent Supreme Court pronouncements, their analyses are not useful here. Indeed, in Texas Pharmacy Association v. Prudential Insurance Co., 105 F.3d 1035, 1039-40 (5th Cir. 1997), the court observed that "[t]here is room to doubt if ERISA's drafters intended that it would preempt any-willing-provider statutes" but felt that a finding of preemption was "compelled" by Supreme Court holdings as to the breadth of that doctrine; the Fifth Circuit felt that a "different result [would] require further guidance from the Supreme Court or further action from Congress." The decisions in Dillingham and DeBuono, both issued after the Texas Pharmacy decision, provide this "further guidance" and make clear that a finding of preemption is no longer, if it ever was, compelled. It is also worth noting that the any willing provider statutes at issue in other cases differ from the Massachusetts Act in significant ways, thus reducing the relevance of such cases to this analysis still further. For example, many of the statutes were held to relate to ERISA plans primarily because they included explicit references to such plans. E.g., CIGNA Healthplan v. Louisiana, 82 F.3d 642 (5th Cir.), cert. denied ___ U.S. ___, 117 S. Ct. 387, 136 L. Ed. 2d 304 (1996); Prudential Ins. Co. v. National Park Med. Center, 964 F. Supp. 1285 (E.D.Ark.1997); Blue Cross and Blue Shield v. Nielsen, 917 F. Supp. 1532 (N.D.Ala. 1996). The Massachusetts Act contains no such reference. It is also clear that ERISA plans were themselves required to comply with the statute at issue in Texas Pharmacy. This is not true of the Massachusetts Act.

The presumption that Congress does not intend to preempt laws which fall within areas of traditional state regulation is applicable here. The Act is an exercise of the state's historic powers to regulate matters of health and safety; it concerns two areas of traditional state governance the health care and insurance industries. The Supreme Court has underscored that "nothing in the language of [ERISA] or the context of its passage indicates that Congress chose to displace general health care regulation, which historically has been a matter of local concern." Travelers, 514 U.S. at 661, 115 S. Ct. at 1680 (citation omitted). The defendants' suggestion that the Act does not involve a traditional area of state regulation because any willing provider statutes are a recent development and do not truly regulate health care is unpersuasive. The appropriate question is not whether the particular statute in question has a long history, but whether the statute falls within an area of traditional state regulation. See DeBuono, ___ U.S. at ___, 117 S. Ct. at 1752 (holding that a recently enacted tax on health care institutions "clearly operates in a field that `has been traditionally occupied by the States' [even though it is not] a regulation of hospitals"); Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724, 741, 105 S. Ct. 2380, 2389-90, 85 L. Ed. 2d 728 (1985) (presumption *66 against preemption applies equally to "traditional" and "innovative" insurance laws). The Act is directed at opening up the carrier-affiliated networks through which many insureds receive pharmaceutical benefits; this clearly concerns the health care and insurance industries.

The presumption that Congress did not intend to supplant this type of law is strengthened by the fact that the goals and effects of the Act are remote from those of ERISA. The Supreme Court explained in Dillingham that "the pre-emption of traditionally state-regulated substantive law in those areas where ERISA has nothing to say would be `unsettling.'" ___ U.S. at ___, 117 S. Ct. at 840 (emphasizing that the statute at issue was "quite remote from the areas with which ERISA is expressly concerned `reporting, disclosure, fiduciary responsibility and the like'") (citations omitted). Here, the Act's focus on how carriers deliver benefits to a clientele which includes ERISA plan beneficiaries is also "quite remote" from any of the concerns expressed in ERISA. Moreover, the Act is one of general applicability it "makes no reference to, [and] indeed functions irrespective of, the existence of an ERISA plan." See Ingersoll-Rand, 498 U.S. at 139, 111 S. Ct. at 483.

The defendants rely heavily on the Supreme Court statement that ERISA preempts "state laws that mandate[ ] employee benefit structures or their administration." Travelers, 514 U.S. at 658, 115 S. Ct. at 1678. However, here there is no question that the Act's mandates are not directly applicable to ERISA plans, that is, ERISA plans are not "carriers" and so may themselves offer pharmacy networks without complying with the Act's requirements. The question is whether the Act is subject to preemption under this standard because it has the indirect effect of preventing ERISA plans like all other purchasers of health care from contracting with carriers for restricted networks which do not comply with the Act's commands.

The parties' disagreement as to the answer to this question stems in part from a dispute over the appropriate interpretation of the term "mandate." The plaintiffs argue that preemption is proper only where a state law "binds" ERISA plans by "requir[ing] that a particular choice be made" by such plans. The plaintiffs conclude that since ERISA plans may continue to use restricted pharmacy networks, the Act is not subject to preemption. In contrast, the defendants claim that "[t]he elimination of only one choice among several triggers preemption." Each proposed interpretation is extreme and somewhat problematic. It seems clear that some laws which restrict ERISA plans to two options could be subject to preemption. For example, a state law providing that employers may make benefit disbursements only in cash or by direct deposit seems an impermissible mandate. On the other hand, surely it is not true that any state law of general applicability which happens to eliminate an option previously available to ERISA plans is automatically preempted. If a state were to require a class of providers to obtain certification in order to practice, ERISA plans would no longer have the option of using uncertified providers in that state. Nonetheless, such a general health care regulation would not appear to be preempted. See Dillingham, ___ U.S. at ___, 117 S. Ct. at 840 (explaining that if ERISA prevented states from taking action concerning medical-care quality standards, "we could scarcely see the end of ERISA's preemptive reach"). Or again, general state contract, zoning or tort legislation can surely affect the options available to ERISA plans without thereby being preempted.

It is critical to examine the nature of the Act's effect on ERISA plans in light of the objectives of ERISA and of ERISA preemption in particular. The primary concern of ERISA was "the mismanagement of funds accumulated to finance employee benefits and the failure to pay employees benefits from accumulated funds." Massachusetts v. Morash, 490 U.S. 107, 115, 109 S. Ct. 1668, 1673, 104 L. Ed. 2d 98 (1989). "The basic thrust of the preemption clause [ ] was to avoid a multiplicity of regulation in order to permit the nationally uniform administration of employee benefit plans." Travelers, 514 U.S. at 657, 115 S. Ct. at 1677-78. The plaintiffs claim that the Act does not interfere *67 with this goal because it "does not affect any of [the] administrative functions and responsibilities" entailed by administration of an ERISA plan. The defendants argue that a carrier's operation of a pharmacy network for an ERISA plan is plan administration and should therefore be protected from state regulation.

The Supreme Court has explained that administering an ERISA plan entails "a host of obligations, such as determining the eligibility of claimants, calculating benefit levels, making disbursements, monitoring the availability of funds for benefit payments, and keeping appropriate records in order to comply with applicable reporting requirements." Fort Halifax Packing Co. v. Coyne, 482 U.S. 1, 9, 107 S. Ct. 2211, 2216, 96 L. Ed. 2d 1 (1987). In FMC Corp. v. Holliday, 498 U.S. 52, 60, 111 S. Ct. 403, 408-09, 112 L. Ed. 2d 356 (1990), the Supreme Court noted the importance of allowing an ERISA plan to develop a uniform administrative scheme with "a set of standard procedures to guide processing of claims and disbursement of benefits."

These cases provide significant guidance as to the Supreme Court's understanding of the administrative practices Congress wished to protect from state interference. The practices which these cases contemplate ERISA plans administering uniformly are different from, and not nearly as extensive as, what the defendants suggest here. The decisions recognize a limited range of administrative functions which are part of operating an employee benefit plan. The Massachusetts Act does not involve any of these areas of plan administration: it in no way interferes with eligibility determinations, benefit calculations, disbursements, fund monitoring or recordkeeping. Nor does the establishment and operation of provider networks seem the type of activity which could become "nationally uniform" absent state interference. Selection of providers is inherently a local matter which involves a variety of regional factors, and it can hardly be expected that it could achieve the same national uniformity as standards for processing claims or making disbursements.

Indeed, it would be odd to conclude that Congress' concern with uniform plan administration is implicated here since selection and operation of provider networks is not a traditional function of ERISA plans, and surely was not a function Congress contemplated in enacting ERISA. The scope of ERISA preemption cannot be extended beyond Congressional contemplation simply because ERISA plans and related entities become involved in additional activities which they wish shielded from state regulation. The Supreme Court provided a convincing illustration of this principle in DeBuono, ___ U.S. ___, 117 S. Ct. 1747, 138 L. Ed. 2d 21. There, the Court held that the fact that an ERISA plan owned and operated its own medical centers did not justify extending the preemptive effect of ERISA to preclude the state from taxing those medical centers. Similarly, the involvement of ERISA plans or fiduciary in activities such as establishment and operation of provider networks ought not extend the scope of ERISA preemption. See Karen A. Jordan, Travelers Insurance: New Support for the Argument to Restrain ERISA Pre-emption, 13 Yale J.Reg. 255, 331 (1996) ("In the emerging integrated delivery systems, some entity must perform the function of deciding which [providers] will comprise the network ... This function is clearly attenuated from functions such as claims administration and therefore should be viewed as constituting ... administration of its business [rather than] administration of an ERISA plan.").

While some of the activities carriers undertake for ERISA plans undoubtedly constitute plan administration and ought be protected from state regulation, it does not follow that everything carriers do for ERISA plans is entitled to the same protection. See Jordan, 13 Yale J.Reg. at 303 ("Because preemption only reaches laws that impermissibly infringe on practices that constitute `administration of the plan,' courts must distinguish between those practices which pertain specifically to the claims management type of functions the [third-party administrator] is performing for an employer, and those functions it is performing for the network.") Professor Jordan emphasizes that where a third-party, such as a carrier, provides administrative services for a plan, it is critical to distinguish between the *68 carrier's administration of the ERISA plan and "its own administration of its business." Id. at 327. She suggests that the appropriate inquiry is "whether the practice effected ... is one of the on-going processes and practices developed to effectuate an employer's provision of benefits ... or whether the practice is more properly characterized as one for the administration of the business of the third-party." Id. The practice at issue here concerns neither eligibility determinations, nor benefit calculations, nor disbursements, nor fund monitoring, nor recordkeeping. Indeed, it touches none of the practices the Supreme Court has identified as part of the administration of ERISA plans. Accordingly, the organization and offering of restricted pharmacy networks should be seen as part of the carrier's own administration rather than its administration of ERISA plans.

Additional support for this interpretation of the scope of plan administration can be found in Travelers, in the Court's cataloguing of earlier decisions which held various state statutes preempted for "mandat[ing] employee benefit structures or their administration." 514 U.S. at 658, 115 S. Ct. at 1678. The Court explained that ERISA preempted the statutes at issue in Shaw because they imposed "mandates affecting coverage" which directly affected the benefit structures which ERISA plans could offer. 514 U.S. at 657, 115 S. Ct. at 1677-78. The law at issue in FMC Corp. v. Holliday interfered with benefit calculations; by prohibiting plans from obtaining subrogation, the law frustrated any attempt at providing uniform national benefits. 514 U.S. at 657-58, 115 S. Ct. at 1677-78. In Alessi v. Raybestos-Manhattan, Inc., 451 U.S. 504, 101 S. Ct. 1895, 68 L. Ed. 2d 402 (1981), ERISA preempted a statute which prohibited plans from using a method of calculating benefits permitted by federal law. 514 U.S. at 658, 115 S. Ct. at 1678. In each of these cases, the Court was concerned with administrative and structural matters central to the administration of ERISA plans themselves. The Massachusetts Act implicates none of these concerns. See also Aetna Life Ins. Co. v. Borges, 869 F.2d 142, 146-47 (2d Cir.1989) ("What triggers ERISA preemption is not just any indirect effect on administrative procedures but rather an effect on the primary administrative functions of benefit plans, such as determining an employee's eligibility for a benefit and the amount of the benefit.").

While the defendants cite much authority for the proposition that ERISA preempts state laws which regulate the administration and structure of ERISA plans, these cases do not establish that the Act concerns plan administration. On the contrary, many of the cases cited by the defendants bolster the plaintiffs' argument as to the limited nature of the administrative practices protected by the preemption clause. For example, the administrative practice at issue in Ingersoll concerned termination of benefits; in Hizer v. GMC, Allison Gas Turbine Div., 888 F. Supp. 1453 (S.D.Ind.1995), the preempted cause of action concerned a claim for improper processing of benefits; in Stone & Webster Engineering v. Ilsley, 690 F.2d 323 (2d Cir. 1982), the statute at issue dictated coverage requirements. These cases all concern practices which are central to plan administration. As the court in Hizer explained, "ERISA preempts laws that affect `the primary administrative functions of benefit plans.'" 888 F. Supp. at 1460. The defendants offer no authority which justifies broadening the conception of plan administration implicit in these cases to include administration of provider networks.

Nor is the defendants' argument that the Act is preempted because it interferes with the ability of carriers to make decisions as ERISA plan administrators and fiduciaries any more compelling. While the cases the defendants cite in this regard do establish that carriers may be ERISA fiduciaries under certain circumstances, they do not support the defendants' claim that the Act impinges on carriers' ability to perform their administrative and fiduciary obligations. Rather, these cases emphasize that not all acts done by an "ERISA fiduciary" or "plan administrator" are done in that capacity. See, e.g., Varity Corp. v. Howe, ___ U.S. ___, ___, 116 S. Ct. 1065, 1072, 134 L. Ed. 2d 130 (1996) (distinguishing between activities done as part of plan management or administration and other business activities). The *69 question once again becomes whether a particular activity involves plan management or administration. The cases the defendants cite do not support a claim, or even suggest, that there is any fiduciary duty involved in the organization and provision of restricted pharmacy networks, or that such an activity is the type of administrative practice with which ERISA is concerned. In Varity, the Supreme Court focused on whether a particular practice is "necessary or appropriate ... to carrying out an important plan purpose" in determining whether the practice falls within the fiduciary duties of an ERISA administrator. ___ U.S. at ___, 116 S. Ct. at 1073. There, the Court found that providing beneficiaries with information about benefits involves plan administration. Id. It is noteworthy that this is a function with which ERISA is expressly concerned. The other cases which the defendants cite involve activities which have long been recognized as part of plan administration. See, e.g., Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 109 S. Ct. 948, 103 L. Ed. 2d 80 (1989) (denial of benefits); Libbey-Owens-Ford v. Blue Cross & Blue Shield, 982 F.2d 1031 (6th Cir.1993) (claim administration); Freeman v. Jacques Orthopaedic & Joint Implant Surgery, 721 F.2d 654 (9th Cir.1983) (eligibility for benefits); Pedre Co. v. Robins, 901 F. Supp. 660 (S.D.N.Y.1995) (investment of plan funds).

The defendants have suggested variously that the Act is subject to preemption because it prevents ERISA plans from contracting with carriers to offer restricted pharmacy networks, and because it interferes with the fiduciary relationship between carriers and ERISA plans. Neither of these suggestions is persuasive. While the preemptive sweep of ERISA is broad, it surely is not so broad as to insulate all vendors with whom ERISA plans deal from the effects of state regulation or to cover all activities assumed by ERISA administrators. The claim that a law regulating the manner in which a third-party vendor offers a service or product is subject to preemption because it affects the options available to ERISA plans relies on logic that is not easily bounded; this reasoning could easily preclude state laws designed to control the quality of health care. The Supreme Court in Travelers emphasized that Congress did not intend ERISA to supersede such legislation. 514 U.S. at 660-61, 115 S. Ct. at 1679-80. The claim that the Act is subject to preemption because it interferes with the fiduciary obligations of carriers fails to recognize the limits of "plan administration." ERISA's concern with preserving a sphere for uniform plan administration was limited to activities involving claims and fund management. Selection of service providers and operation of provider networks has "too tenuous, remote and peripheral" a connection with plan administration to warrant preemption.

Accordingly, because I conclude that the Act does not mandate employee benefit structures or administration, I hold that it does not relate to ERISA plans and thus is not preempted.

The breadth of ERISA's preemption provision is limited by what has become known as the "insurance saving clause." 29 U.S.C. § 1144(b) (2) (A) ("[N]othing in this subchapter shall be construed to exempt or relieve any person from any law of any State which regulates insurance."). The plaintiffs argue that even if the Act were found to "relate to" ERISA plans, it would be saved from preemption by this provision. The defendants respond that the Act is not an insurance regulation and that this provision thus would not protect the Act from the preemptive effect of ERISA.

The conclusion that the Act is not preempted because it does not "relate to" ERISA plans makes it unnecessary to decide whether the Act would be saved from preemption as a law "which regulates insurance." Nonetheless, because of the checkered history of preemption decisions, it seems appropriate to address the issue, at least briefly.

The Supreme Court has analyzed saving clause questions in two ways. First, the Court has asked whether the statute in question fits within "a common-sense understanding of the phrase `regulates insurance.'" Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 50, 107 S. Ct. 1549, 1554, 95 L. Ed. 2d 39 (1987). *70 To satisfy this test, "a law must not just have an impact on the insurance industry, but must be specifically directed toward that industry." Id.

Second, the Court has turned to case law interpreting the phrase "business of insurance" in the McCarran-Ferguson Act. In this context, the Court has considered three criteria for determining whether a practice qualifies as the "business of insurance": "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." Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724, 743, 105 S. Ct. 2380, 2391, 85 L. Ed. 2d 728 (1985) (citing Union Labor Life Ins. Co. v. Pireno, 458 U.S. 119, 129, 102 S. Ct. 3002, 3008-09, 73 L. Ed. 2d 647 (1982)). "[N]one of these criteria is necessarily determinative in itself." Pireno, 458 U.S. at 129, 102 S.Ct. at 3008-09; Ocean State Physicians Health Plan, Inc. v. Blue Cross & Blue Shield, 883 F.2d 1101 (1st Cir.1989).

On a common-sense view, the Massachusetts Act regulates insurance. It appears in the statute books as part of a chapter entitled "Unfair Methods of Competition and Unfair and Deceptive Acts and Practices in the Business of Insurance." Mass.Gen.Laws ch. 176D. The Act is concerned with provision of benefits to "insureds." The carriers are all entities which are commonly recognized as insurers or insurer-related entities which, for a fee, assume the risk entailed in ensuring that their customers receive health care. The sample contracts provided by the defendants in which each agrees to provide health-care benefits in exchange for payment of fixed premiums meet the general definition of an insurance contract under Massachusetts law. Mass.Gen.Laws ch. 175 § 2 ("A contract of insurance is an agreement by which one party for a consideration promises to pay money or its equivalent, or to do an act valuable to the insured, upon the destruction, loss or injury of something in which the other party has an interest."). By restricting the ability of carriers to exclude pharmacies from their networks, the Act affects the terms and conditions by which carriers may provide their insureds with pharmacy benefits.

The defendants argue that the Act does not satisfy this test because it is not "specifically directed" toward the insurance industry. They claim that not all carriers are engaged in the business of insurance, and maintain that the Act regulates pharmacies, which are not engaged in the business of insurance.^[9] The suggestion that the Act is not directed toward the insurance industry because it applies to HMOs is unpersuasive; there is no good reason to distinguish here between entities which accept the risk involved in providing benefits in exchange for *71 fixed premiums. See Cellilli v. Cellilli, 939 F. Supp. 72, 76 (D.Mass.1996) (difference between a health maintenance plan and an insurance contract "is immaterial" to saving clause analysis). Nor is the suggestion that the statute's relation to pharmacies prevents it from being an insurance regulation compelling. In Metropolitan Life, the Court recognized that the protective power of the saving clause extends beyond laws which "regulate only the insurer, or the way in which it may sell insurance." 471 U.S. at 741, 105 S. Ct. at 2390. Metropolitan Life demonstrates that a law may be an insurance regulation even if it has a significant impact on non-insurers. The law at issue there required that insurers provide mental health benefits a requirement that clearly has substantial effects on the providers of such benefits.

Pilot Life's statement that a statute must be "specifically directed" toward the insurance industry suggests a looser inquiry than that contemplated by the defendants. In Pilot Life, the Court held that a state law which was an application of "general principles of [ ] tort and contract law" was not an insurance regulation. In Williams v. Ashland Engineering Co., 45 F.3d 588, 592 (1995), the First Circuit concluded that a law whose "impact on the insurance industry [was] happenstance" was not an insurance regulation. These holdings suggest that this inquiry is intended only to ensure that the regulation is about insurance, not to draw fine lines between various entities involved in handling the risks associated with health care. This remains a test rooted in commonsense understanding.

The next question is whether the Act satisfies the three-part McCarran-Ferguson test. First, the Act has the effect of transferring or spreading a policyholder's risk because, by enabling insureds to fill their prescriptions at additional pharmacies, it transfers some of the cost of treatment which is the risk insured against from the insured to the carrier. Without the Act, insureds who fill prescriptions at mom-and-pop pharmacies which are not part of their carrier's network must bear the cost of the medication; with the Act, there is a far greater chance that an insured may be reimbursed for filling a prescription at a neighborhood pharmacy. Second, the Act is an integral part of the relationship between the insurer and insured because it affects the substantive terms of the policy by broadening the range of providers that an insured may choose. And third, as explained above, the Act is limited to entities "within the insurance industry" since all carriers are in the business of accepting the risk of providing health care coverage for a fee. The Supreme Court has recognized that a statute which "impose[s] requirements only on insurers, with the intent of affecting the relationship between the insurer and the policyholder" satisfies this third prong. Metropolitan Life, 471 U.S. at 743, 105 S. Ct. at 2391.

The defendants argue that Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 99 S. Ct. 1067, 59 L. Ed. 2d 261 (1979) compels a contrary conclusion. That case involved an antitrust challenge to Blue Shield's policy of requiring participating pharmacies to accept a $2 payment from insureds and be reimbursed for the remainder of the cost by Blue Shield; an insured who filled a prescription at a non-participating pharmacy paid the full cost and was entitled to partial reimbursement from Blue Shield. The Court held that contracts between Blue Shield and the participating pharmacies were not the business of insurance because these contracts did not affect the risk that policyholders would be unable to purchase prescription drugs. 440 U.S. at 213, 99 S. Ct. at 1074. Rather, the Court explained that the agreements "serve[d] only to minimize the costs Blue Shield incurs in fulfilling its underwriting obligations." Id. Royal Drug is distinguishable because that case was concerned only with whether an insurer's agreement with participating pharmacies was part of the business of insurance. Because the contents of these contracts did not affect the insured's ability to obtain reimbursement, the Court found that "policyholders are basically unconcerned with arrangements made between Blue Shield and participating pharmacies." Id. at 214, 99 S. Ct. at 1074. Here, in contrast, the concern is with the potential expansion of -pharmacy networks to include previously non-participating pharmacies. By directing insurers to open their networks, the Act should affect *72 the availability of benefits to insureds by increasing their ability to select pharmacies of their choice. This does concern the relationship between insurer and policyholder, and thus may be considered part of the business of insurance. Moreover, it is questionable whether the holding of Royal Drug may be translated into the ERISA preemption context at all. That Court emphasized that "quite different" questions are raised by the McCarran-Ferguson Act's exemption of the business of insurance from antitrust laws and by its preservation of state regulation of the activities of insurance companies. Royal Drug, 440 U.S. at 218 n. 18, 99 S. Ct. at 1077 n. 18 (explaining that the antitrust exemption covers a far more limited area than the area which the McCarran-Ferguson Act preserved for state regulation). The Court in Royal Drug considered only whether the contracts at issue fell within the antitrust exemption, not whether they fell within an area which states remain free to regulate.

Because the Act satisfies the commonsense understanding of a law "which regulates insurance," and because it meets all three parts of the McCarran-Ferguson "business of insurance" test, I conclude that the Act is a regulation of insurance protected from preemption by the saving clause.

Accordingly, because the Act does not "relate to" ERISA plans within the meaning of the preemption clause, and because the Act is protected from preemption by the insurance saving clause, ERISA does not preempt the Act.^[10] The defendant's motion for summary judgment is therefore denied.

It is so ordered.

[1] Osco and the Commonwealth are referred to collectively as the plaintiffs.

[2] A "carrier" is defined as:

an insurer operating pursuant to the provisions of [M.G.L. ch. 175], a hospital service corporation operating pursuant to the provisions of [ch. 176A], a medical service corporation operating pursuant to the provisions of [ch. 176B], a health maintenance organization operating pursuant to the provisions of [ch. 176G], and a preferred provider arrangement operating pursuant to the provisions of [ch. 176I], or a wholly-owned subsidiary or affiliate under common ownership thereof.

Mass.Gen.Laws ch. 176D, § 3B.

[3] An "insured" is defined as "a person whose health care services and benefits are provided by, or indemnified by or otherwise covered by a carrier's group or individual insurance policy, or certificate, agreement or contract and shall include subscribers, enrollees or members." Mass. Gen.Laws ch. 176D, § 3B.

[4] A "restricted pharmacy network" is defined as:

an arrangement for the provision of pharmaceutical drug benefits to insureds which under the terms of a carrier's policy, certificate, contract or agreement of insurance or coverage requires an insured or creates a financial incentive for an insured to obtain prescription drug benefits from one or more participating pharmacies that have entered into, a specific contractual relationship with the carrier pursuant to a competitive bidding process.

Mass.Gen.Laws, ch. 176D, § 3B.

[5] See, e.g., Larry J. Pittman, "Any Willing Provider" Laws and ERISA's Saving Clause: A New Solution for an Old Problem, 64 Tenn.L.Rev. 409, 432-439 (1997) (explaining that states may enact any willing provider laws in order to protect providers, improve the quality of patient care and increase patient freedom of choice); William J. Bahr, Comment, Although Offering More Freedom to Choose, "Any Willing Provider" Legislation is the Wrong Choice, 45 U.Kan.L.Rev. 557, 570 (1997) ("In general, [any willing provider] statutes provide a legal recourse to excluded [providers, and] provide people with more selection and freedom to choose their health provider.") (citations omitted).

[6] The remainder of the HPHC members who have a prescription benefit receive pharmacy services from pharmacies owned by HPHC. The Act does not apply to the provision of services by pharmacies wholly-owned by carriers. Mass. Gen.Laws ch. 176D, § 3B.

[7] The information concerning HPHC's operations is taken from the Affidavit of Ferris W. Taylor, and the analysis of Prudential's business has been gleaned from the Affidavit of Richard Campagna. Prudential has not specified what percent of its business involves restricted networks.

[8] The defendants confine their argument to the question of whether the Act impermissibly interferes with the administration of ERISA plans. They do not rely on any claim that the Act is subject to preemption because of its indirect economic effects on ERISA plans. The defendants have reserved arguments on that issue to avoid wide-ranging discovery concerning the economics of restricted pharmacy networks.

[9] The defendants also emphasize that the Act applies to pharmacy networks offered to beneficiaries of self-insured ERISA plans, and argue that because of this the Act is not directed specifically toward the insurance industry. This argument is unpersuasive in the face of Supreme Court interpretation of the "deemer clause," which limits the reach of the saving clause by providing that an ERISA plan shall not "be deemed to be an insurance company or other insurer." 29 U.S.C. § 1144(b) (2) (B). This clause has been interpreted to exempt self-insured plans from the scope of otherwise applicable insurance regulations which are saved from preemption. FMC Corp. v. Holliday, 498 U.S. 52, 61-64, 111 S. Ct. 403, 409-11, 112 L. Ed. 2d 356 (1990) ("Our interpretation of the deemer clause makes clear that if a plan is insured, a State may regulate it indirectly through regulation of its insurer and its insurer's insurance contracts; if the plan is uninsured, the State may not regulate it."); Metropolitan Life, 471 U.S. at 747, 105 S. Ct. at 2393 ("We are aware that our decision results in a distinction between insured and uninsured plans, leaving the former open to indirect regulation while the latter are not."). Consequently, any decision holding that the Act is saved from preemption by the insurance saving clause would have to recognize that the Act could not apply to the provision of services to self-insured plans. Because the Act would not apply to these uninsured relationships, it is unnecessary to consider whether the Act would lose its status as an insurance regulation because it purports to cover these relationships. The defendants' position is inconsistent with the holding of Metropolitan Life. There, the fact that the statute at issue applied by its terms to self-insured plans did not prevent the Supreme Court from concluding that the statute was limited to the insurance industry. 471 U.S. at 735 n. 14, 743, 105 S. Ct. at 2387 n. 14, 2391. Instead, the deemer clause simply prevented application of the statute to self-insured plans. Id.

[10] In light of the determination that ERISA does not preempt the Act, the question of severability does not arise. It is unnecessary to address the purely hypothetical question of whether if the Act were preempted, it would be preempted in its entirety or severed so as to remain applicable where ERISA concerns are not implicated.