LEDA at Harvard Law
“The Department recognizes that people who may be affected by a demonstration project have a legitimate interest in learning about proposed projects and having input into the decision-making process prior to the time a proposal is submitted to the Department. A process that facilitates public involvement and input promotes sound decision-making .”
- HCFA Federal Register notice of 1994 (emphasis added)
The proper level of judicial review for Section 1115 Medicaid demonstration program waiver (“waiver”) decisions made by the Secretary of the Department of Health and Human Services (“the Secretary”) is an important question of law regarding one of the federal and state governments’ largest social programs. Judicial review of waivers is of particular importance in the area of pharmacy benefits. In the absence of Congressional action to address the dual problems of drug costs and coverage that are driving much of health care spending, several states have turned to Medicaid waivers as a means to implement innovative pharmacy benefits programs and garner federal funds for their finance. Many more states are looking to enact such programs in the future. Pharmaceutical access is an issue of national importance due to the broad need for drugs among different groups of health care consumers. The supply-side impacts of pharmaceutical access also are nationwide: large-scale drug manufacturers sell their standardized products in a national pharmaceutical market. While the need for pharmaceutical access may be universal and the market at least semi-national, the proper contours of pharmacy coverage may take different forms based on the particulars of each state’s population, financial resources, and health care system structure. Thus it is imperative that federal and state health agencies work together to address and provide a check for each other on the issue of drug access. More importantly, debates about the proper policy role of the federal and state governments in providing pharmacy assistance, such as those ongoing in the Medicaid waiver context, should be open to all interested parties. A closed process between CMS and individual state agencies risks omitting important consumer and industry voices from the policy debate.
In the past, courts have granted substantial deference to the Secretary’s waiver determinations. Recent administrative case law both identifies the need for and demands a more reasoned justification for such deference than past courts have provided. Over the two terms from 2000 to 2002, the federal district courts for the District of Columbia issued a set of opinions ruling on the authority of the Secretary of the Department of Health and Human Services (DHHS) to approve state Medicaid waiver requests to expand pharmaceutical benefits to otherwise ineligible residents. As the opinions suggest, confusion existed at the time as to the appropriate level of deference to grant the Secretary’s actions under the Administrative Procedure Act (APA) and the lead case on the question of deference, Chevron v. Natural Resources Defense Council . A mere ten days after the final substantive ruling on the pharmaceutical benefits waiver cases, the Supreme Court issued its decision in United States v. Mead Corporation. The Mead Court distinguished those agency actions due heightened deference under the Court’s prior decision in Chevron from other, more informal agency actions which are due an intermediate level of deference in keeping with their “power to persuade.” In elaborating on the proper standard of review for informal agency actions, the Mead Court set up a new framework for future courts faced with Medicaid waiver decisions. Courts first must ask whether Congress delegated to the Secretary the authority to make waiver determinations that “carry the force of law” and whether the Secretary acted in exercise of that authority before granting Chevron deference. The purpose of this paper is not to challenge the Mead decision, but to apply its holding and legal concepts to the specific case of Section 1115 Medicaid waivers for pharmacy benefits expansion proposals.
Mead outlines what may be identified as four major factors for consideration in determining whether Congress intended for the agency actions to have the force of law. They are:
Under these Mead factors, an argument may be made that Congress did intend for the Secretary’s waiver determinations to have the force of law. Even given the case in favor of Chevron deference, Mead’s general focus on formal process argues against granting Chevron deference to the Secretary’s waiver decisions under the agency’s current review process. A better take on Mead’s open balancing test is that agency expertise cannot entirely substitute for process in determining the proper level of judicial deference with regards to certain agency actions. While the Mead Court notes that process is not determinative of the deference question, the Court emphasizes that “[i]t is fair to assume that Congress contemplates administrative action with the effect of law when it provides for a relatively formal administrative procedure tending to foster the fairness and deliberation that should underlie a pronouncement of such force.” One may read this passage to imply that where Congress has not provided for such formal procedure (as with the waiver program) and an agency seeks to act with the force of law in order to merit judicial deference, courts should require that the agency provide some degree of relatively formal administrative process in keeping with the “force of law” characteristics of the action in question. Agencies who do not afford formal process should be at risk of having their decisions closely scrutinized for a match between expertise and the type of decision in question under an alternative test to Chevron.
This proposition is especially convincing where a court is faced with an agency action regarding experimental programs within a scheme of cooperative federalism. By its very definition, the word “experimental” suggests that agency experience alone cannot suffice in producing a desirable decision. Deliberative processes about an unknown future must also involve a full range of voices with a stake in the range of possible outcomes. A full range of parties interested in pharmacy benefits is not limited to the traditional Medicaid interest group of low-income beneficiary advocates. Industry representatives such as drug manufacturers also have a stake in the outcome of agency policymaking. As this case illustrates, the argument for less deference in absence of agency procedures is not skewed toward any vision of what particular interest will benefit the most from increased process, but rather posits that the endpoint good is participation itself. A program’s inclusion as part of a cooperative federalist regime also indicates that a single level of government process is insufficient to ensure full participation.
The Medicaid Section 1115 waiver program has long failed to provide such a desirable quantum of public participation through formal process at the state and federal levels. In light of Mead, reform of the waiver program should focus on a commitment to process. Both state and federal procedural requirements are necessary to ensure that the full range of voices have an opportunity to participate in the waiver process. Increased process in the granting of waivers will serve beneficiaries and other interested parties by ensuring fairness and deliberation under the public eye. The reviewing agency will benefit as well by increasing its own expertise and insulating its decisions from unwarranted judicial interference in technical matters of health policy.
Lack of Process and Public Participation in the Waiver Program
Medicaid demonstration waivers under Section 1115 are one of the strongest tools available to states in creating and reforming health policy. Waivers allow states to implement innovative health policies through the federal-state Medicaid program where such policies would normally not qualify for inclusion, and thus federal funding, under the federal Medicaid statute.  The Secretary’s approvals of waivers ostensibly fall into the broad category of informal agency actions. Neither the waiver statute nor agency regulations subject waiver determinations to formal adjudication, formal rulemaking, or informal notice-and-comment rulemaking. Nor has the Department of Health and Human Services (“DHHS”) issued regulations otherwise governing its consideration, approval, and oversight of waivers. Due to Congress’ broad delegation of authority to the Secretary to make substantive decisions regarding state waiver requests , coupled with a lack of administrative process requirements, the Secretary enjoys almost complete discretion to approve, amend, or disapprove a state’s waiver request. Courts reviewing waiver determinations generally have recognized this discretion by granting substantial deference to the Secretary’s decisions under ordinary Administrative Procedures Act arbitrary and capricious review , thereby effecting “nearly unreviewable authority” in the Secretary to determine whether a waiver request is consistent with the purposes of the Medicaid statute. Thus, the Department exercises substantial discretion in deciding how much public participation to allow in the development and analysis of waiver requests. The Department in practice has continuously failed to follow even its own unofficial procedural outlines and/or shifted the burden of providing process to state Medicaid agencies. At the same time, the Department has granted the states broad discretion in determining how to meet the federal minimum.
In sum, the waiver program exhibits a systematic federal avoidance of public notice and comment procedures and enormous variation in process at the state level. This pattern of evasion and disparity, ironically accompanied by official recognition of the value of administrative process at both the state and federal level, has produced a questionable use of cooperative federalism whereby important federal-state health policy negotiations take place removed from the public eye. Furthermore, it appears that the federal government may be exceeding its statutory authority by inappropriately granting waivers that effect sweeping Medicaid reform, and/or waiver modifications implemented through program continuations as a quick end-run around the more intensive full waiver review process. These routes to Medicaid reform are questionably appropriate under Congress’ intent in passing the waiver provision and the text of the waiver statute itself.
Recent Jurisprudence on Judicial Review of Informal Agency Actions: The PhRMA Cases
Immediately prior to Mead, the federal district courts of the District of Columbia decided a set of companion cases considering waiver requests to expand pharmaceutical coverage under Vermont’s and Maine’s Medicaid programs. At issue was the Secretary of the Department of Health and Human Services’ interpretation of the Medicaid drug rebate provision as reflected through the Secretary’s approval of the states’ waiver requests. In the first appellate decision on the Vermont waiver, the judiciary broke from its usual position of deference to waiver decisions made by the Secretary by invalidating an approval of the state’s waiver request. The district court of appeals based its decision on a straightforward application of the first prong of Chevron’s analysis and found that Congress had directly spoken on the matter at hand. A subsequent district trial court faced the question of domain left open by Chevron but without the guidance of Mead. In delivering its opinion, the court granted Chevron deference to the Secretary’s approval of a similar pharmaceutical benefits program without making an inquiry into administrative process.
The two opinions reflect the opposite sides of the deference coin. As with the appellate court decision, courts may feel an impulse to retain some function of judicial review over waiver decisions, especially in light of the lack of consistent public process and considering the size and importance of the Medicaid program. The Vermont appellate court alluded to its concern with process in referencing a pre-Mead case holding that some informal agency actions do not qualify for heightened deference. Chevron allows courts to retain review oversight where sufficiently clear substantive law must be construed in the Secretary’s waiver analysis. The drug rebate provision, as a nonwaivable element, provides the courts with the perfect opportunity to reign in waiver discretion. At the same time, courts in the post-Chevron world may be hesitant to question the Secretary’s choice on difficult matters of health policy. Courts calling for deference may recognize that while the external process controls over waiver proposals are weak, the Department of Health and Human Services has constructed a centralized and stringent internal, inter-governmental review process for waiver requests. In addition, DHHS has issued numerous detailed guidance memos and policy statements to assist states in constructing waiver proposals. These agency actions, while heavily weighted towards the informal end of the process spectrum, reflect the department’s careful consideration of Medicaid policy issues. Thus, the federal and state agency efforts seem worthy of deference. Chevron itself, nevertheless, was a case involving a cooperative federalist scheme. That a third party brought suit against the federal agency in Chevron is proof that state agency involvement cannot substitute entirely for public participation.
Waiver Reform Proposals and Health Policy
Lack of transparency is a legitimate critique of the waiver system. How to remedy the system is less clear. One approach is to convert Medicaid into a block grant program under which the waiver program ceases to exist and the states gain the power to unilaterally determine Medicaid policy above certain federal minimums. To the extent that policymakers at both levels continue to believe that there are lessons to be learned from the various experiences of the states under the waiver program and that federal funding conditioned on some degree of federal oversight is desirable, waiver reform is in order. A critical component of waiver reform is the adoption of and adherence to clear procedural requirements. Procedural requirements will open the waiver approval process to the participation of both the state and national public. The Department of Health and Human Services will also benefit from the requirements: process will preserve judicial deference for the Secretary’s informed judgments, deference which may be open to attack under Mead. The Secretary’s waiver decisions are statements of health policy in that they approve of a given state’s approach to a health concern facing the state’s residents. Deference to waiver approvals leaves in place the policy statement of the Secretary. Thus, questions remain after the waiver decision as to whether a particular waiver proposal constitutes “good” health policy.
The following analysis proceeds in four parts. Section One provides background on Medicaid as a whole, the particular provisions at issue in the PhRMA cases, and the PhRMA court decisions. Next, Section Two attempts to answer the deference question vis-à-vis Medicaid waivers in the new world of Mead by looking at both the action’s claim to having the “force of law” and the scope of the Secretary’s authority. Section Three consists of a discussion of the benefits of both increased federal and state procedural requirements. Finally, Section Four considers the policy arguments surrounding state expansion of pharmaceutical benefits through the Medicaid waiver program. These arguments include market theories regarding the prudence of state price controls, federalist concerns with the allocation of health policy-making power between the state and federal governments, and medical/economic considerations in relation to the appropriate focus of limited health care financial resources.
Enacted by Congress in 1965, the Medicaid program is an example of cooperative federalism under which the federal and state governments provide medical care to needy individuals. The states are responsible for the administration of care delivery, while the federal government retains oversight through a program approval process and contributes matching funds for approved state programs. Two categories of people may qualify as Medicaid beneficiaries if they meet income eligibility guidelines: families with dependent children and the disabled. Once enrolled, beneficiaries are entitled to receive both “medically necessary services” and “rehabilitation and other services to help... attain or retain capability for independence or self-care.” Beneficiaries may not be required to pay more than a nominal amount out-of-pocket towards the care that they receive through Medicaid.
The federal Medicaid statute imposes particular requirements on the states in terms of eligibility and covered services, but leaves the states discretion in deciding the bounds and contours of their programs. Each state is responsible for devising a State Plan that specifies which services the state has elected to cover and which persons the plan will protect within the federal requirements. Certain benefits, such as prescription drugs, dental care, hospice care, and prosthetic devices, are optional for the states. Under the plan, a state reimburses health care providers for covered services according to government-established rates. The state must submit its State Plan to the Secretary of the U.S. Department of Health and Human Services (the “Secretary”) for approval. Having received the Secretary’s approval, the state may then implement the plan and receive federal matching funds in an amount equal to a statutorily-determined portion of the state’s medical assistance expenditures under the plan. A state may receive between 50 and 77% of the cost of providing care from the federal government depending on the state’s per capita income.
The significance of Medicaid to the lives of low-income Americans, as well as to state and federal governments, cannot be understated. Nationally, Medicaid covers approximately one-in-ten Americans and often is the only source of health care coverage for low-income children and most nursing home care. Medicaid also is a key contributor to state economies, as it constitutes the single largest source of federal funds to the states at 43 percent of all federal grants-in-aid. After a period of significant improvement and expansion of state Medicaid programs during the 1990’s, state governments in the new millennium are facing falling tax revenues and thus budget shortfalls, which in turn put pressure on state “rainy day” funds. The cost of health care has continued to rise over these two periods of surplus and deficit; when combined with growing Medicaid enrollment during economic downturns, state Medicaid costs have risen even as states’ abilities to fund the program decreases.
Prescription drug coverage is one medical benefit that may be included by a state in its Medicaid State Plan. At present, all fifty states and the District of Columbia have opted to provide a Medicaid prescription drug benefit. Medicaid officials in forty-four states cite prescription drugs as among the three most significant factors driving Medicaid expenditure growth in their states; twenty-five states listed prescription drugs as the number one driver of Medicaid costs in FY 2002. Prescription drugs on average account for about 10 percent of states’ total Medicaid spending and represented 19.7 percent of the annual rate of growth in state total Medicaid expenditures from 1998 to 2000.
While rising prescription drug costs have been of particular concern in recent years, recognition of their potential drain on Medicaid purses has long been codified at the federal level. To help contain prescription drug costs, the federal Medicaid statute has included a provision since 1990 mandating that pharmaceutical companies who wish their drugs to be included in the Medicaid program enter into drug rebate agreements with the Secretary of DHHS on behalf of the states. Under a rebate agreement, a pharmaceutical manufacturer reimburses the federal and state governments for a portion of the governments’ expenditures in providing Medicaid enrollees with the company’s drugs. Companies pay rebates on a quarterly basis. The rebate amount is calculated according to the number unit volume of a manufacturer’s drugs that are dispensed to Medicaid beneficiaries. “Payment” for the rebated drugs, in turn, must have been “made under the State Plan.” In 1993, Congress amended the original drug rebate provision to allow for state use of cost containment measures such as prior authorization, formularies, and categorical exclusion of certain drug types.
While the federal Medicaid statute outlines certain State Plan requirements, states may apply for waivers to implement “experimental, pilot, or demonstration” projects under Section 1115 of the Medicaid Act. Congress originally enacted the waiver provision in 1962 as part of a broad effort to give the states more freedom in implementing welfare programs. According to the Senate report accompanying the waiver provision, Congress passed the provision in recognition that the strictures of federal legislation “’often stand in the way of experimental projects designed to test out new ideas and ways of dealing with the problems of public welfare recipients.’” The focus of waiver programs was to be on improvement of the techniques of administering assistance programs. In 1965, Congress amended the waiver provision upon passage of the Medicaid Act to specify its applicability to state Medicaid programs. The waiver provision includes a special, expedited review process under which a state can seek a “continuation” of an existing waiver program prior to the program’s expiration.
A state’s waiver application must include detailed information about the statutory and regulatory requirements seeking to be waived, as well as analysis of the impact of the waiver on program expenditures, relevant laws, and beneficiaries enrolled in the project. The waiver provision itself limits those requirements to which it may apply. Of interest to Vermont and Maine, the Secretary may not waive any requirements of the federal drug rebate provision. Expansion of Medicaid coverage to a new set of beneficiaries may require a waiver. Also important for the present analysis, expansion of coverage to childless adults requires a waiver. Such expansion to children and parents, however, does not. Additionally, the Secretary may waive the nominal copayment provision. For the purposes of a Section 1115 waiver, the Secretary also may designate costs of the project that would not otherwise qualify as “expenditures under the State plan” as qualifying expenditures for a period set by the Secretary. CMS further requires that a state demonstrate the project’s “budget neutrality.” To meet this requirement, a state must show that the lifetime costs of the waiver program to the federal government will not exceed the amount that the federal government would have paid to the state under the State Plan absent the waiver. 
In reviewing a state’s Section 1115 waiver application, the Secretary of DHHS must determine whether the program is likely to help promote the objectives of Medicaid. The Secretary may consider both efficiency and effectiveness criteria in making this determination. To assist the Secretary, the Center for Medicare and Medicaid Services (CMS), the division of DHHS delegated responsible for the Medicare and Medicaid programs, convenes a technical panel that analyzes the state’s proposal in terms of methodology and design, objectives, and expected costs and returns. Also of importance to CMS is the applicant state’s experience in the relevant area of health policy. If the program has a research component, the panel weighs the potential risks to health and safety of participants. The panel then makes a recommendation of approval, conditioned approval, or rejection of the demonstration program, which the Administrator of CMS incorporates into a decision memorandum to the Secretary. While the states have some latitude in designing their Medicaid State Plans, they generally must still adhere to the federal statutory and regulatory guidelines for state plans. States may, through the waiver provision, essentially opt-out of these federal requirements. Such increased state flexibility in determining policy in areas traditionally left to the states is a core tenant of federalism.
In 1995, Vermont received approval from CMS, then the Health Care Finance Administration (HCFA), for a Section 1115 waiver to implement the state’s pilot program known as the Vermont Health Access Plan (VHAP). VHAP extended pharmacy supplemental benefits to two groups: elderly or disabled beneficiaries with incomes up to 150 percent of the federal poverty level (FPL) and uninsured adults with incomes up to 150 percent FPL who are provided a comprehensive insurance program with a prescription-drug benefit. A 1999 extension of HAP in turn provided “maintenance” drugs for elderly or disabled beneficiaries with incomes between 150 and 175 percent of the FPL. VHAP beneficiaries bear 50% of the cost of prescriptions through a copayment on drug purchases. The Secretary of DHHS in June of 2000 extended HAP’s approval through the end of 2003.
The program at issue in PhRMA I was not the initial VHAP waiver project, but rather an expansion of VHAP proposed by the state in the spring of 2000. Under its expansion waiver request, Vermont sought to provide prescription drug assistance in the form of drug discounts to two new groups. First, those individuals who are covered by Medicare and have incomes above 150 percent of the FPL but who lack prescription drug coverage would be entitled to discounted prescription drugs. In addition, all adults with incomes at or below 300 percent FPL who are either uninsured or enrolled in a benefit program which lacks drug coverage would receive the same benefit. The program went by the title of the Prescription Drug Pilot Project (PDP). The latter category of adults, if implemented, would cover approximately 70,000 new beneficiaries. In essence, the PDP would create two new classes of Medicaid beneficiaries for the limited purpose of prescription drug coverage.
In order to finance the prescription drug discounts, the state devised a rebate-reimbursement program. Medicaid beneficiaries were to make a copayment of 82.5% of a drug’s Medicaid price; the State would then pay the remaining 17.5% to the pharmacy. After paying the pharmacy, the State would bill pharmaceutical manufacturers for a rebate in the amount of 17.5% of the Medicaid price. The PDP rebate figure is linked to Vermont’s Medicaid rebate, in that the PDP rebate must be equal to the estimated average rebate that Vermont receives for all drugs. The effect of the program was to “pass-through” manufacturer rebates to PDP enrollees, who themselves pay for 82.5% of the drug’s retail price, with the State incurring the costs of administration and the opportunity costs of “fronting” the 17.5% until the manufacturers pay their rebates.
PhRMA’s claim against the Secretary of DHHS alleged a violation of Title XIX of the Social Security Act (the Medicaid Act) because the PDP imposed no direct costs on the State of Vermont but rather shifted the entire burden of providing drug discounts to drug manufacturers. The PDP thereby failed to meet the requirement that a Medicaid plan include some “payment under a state plan” of the costs of “medical assistance.” In addition, PhRMA alleged that the plan violated the Medicaid rule against charging Medicaid beneficiaries no more than a “nominal” copayment. PhRMA thus asked the court to find DHHS’s approval to be arbitrary and capricious in violation of the Administrative Procedures Act (APA). The State of Vermont intervened on behalf of the defendant. As both the trial court and appeals court decided the case on the basis of the “payment” requirement, thus obviating the need to reach the nominal copayment issue, this section will consider only the payment issue.
According to the trial court, PhRMA’s briefs could be interpreted as alleging two different grounds for violation. First, PhRMA seemingly made a claim of improper waiver: by granting approval, “the Secretary had waived the [non-waivable] Medicaid statute’s requirements of ‘payment under the State plan.’” PhRMA’s submissions could also be read as alleging improper approval, in the sense that the Secretary had approved a program that “’lacked the requisite payment.’” The second ground challenges the Secretary’s interpretation of the statutory term “payment.” In part supported by the defendants’ agreement that the Secretary’s approval was not based on a waiver of the “payment” requirement, the trial court adopted the second interpretation of PhRMA’s claim.
In early 2001, the Secretary approved Maine’s request for a pharmacy benefits waiver demonstration program. The Healthy Maine Prescription proposal sought to extend Medicaid eligibility for prescription drugs to all individuals living in households with incomes of up to 300 percent of the FPL. Like Vermont’s PDP, the HMP would use the federal Medicaid manufacturer-rebate mechanism to collect rebates and create a revolving fund. The state Medicaid program would then reimburse enrollees for a portion of their prescription drug costs equal to the amount of the rebate minus the cost of administering the pharmacy benefits program. As approved by the Secretary in January of 2001, the HMP was “’essentially identical’” to Vermont’s PDP. However, after the appellate court delivered its ruling in PhRMA I requiring that an acceptable drug rebate waiver program include expenditure of state-only funds, Maine began a policy under which it covered two percent of HMP enrollees’ drug costs with “’State-only’” money. PhRMA challenged Maine’s plan under the same theories as in PhRMA I; in addition, PhRMA claimed that the two percent state-only contribution did not change the legal character of the program pursuant to the PhRMA I appellate decision, as the state was not required to make the payments by “any official federal or state action.”
DEFERENCE TO AGENCY ACTIONS
The Supreme Court in 1984 decided the landmark case of Chevron v. Natural Resources Defense Council. In its decision, the Court outlined a new doctrine of heightened judicial deference to federal agency decisionmaking where Congress has not directly spoken to the issue at hand. Left ambiguous by the Chevron decision was whether all federal agency actions qualify for heightened deference under the second prong of Chevron’s analysis. In other words, are all agency actions analyzed under Chevron in the first instance? The Court recently sought to answer this question of Chevron’s “domain” in the controversial case of United States v. Mead. Under Mead, the Court affirmed that some agency decisions, most notably those that are the product of less formal administrative processes, do not deserve heightened deference. Courts faced with reviewing such informal agency actions should instead grant them deference in keeping with their “power to persuade” under the Skidmore doctrine. Section 1115 Medicaid demonstration program waiver determinations fall into the category of informal agency actions that may qualify for less deference.
The question of the degree of deference due the Secretary’s approvals lies at the heart of the opinions on Vermont’s and Maine’s prescription drug benefit demonstration programs. The Chevron doctrine, first articulated by the Supreme Court in a case concerning air quality standards under the Clean Air Act , currently governs the degree of deference due administrative action. Specifically, the doctrine lays out an analytic framework emphasizing the roles of Congress and administrative agencies in determining policy while reducing the role of the courts.
Chevron analysis as originally conceived consists of two principal steps. First, a court must ask “’whether Congress has directly spoken to the precise question at issue.’” This step of the analysis, known as “Step One,” requires the court to engage in statutory construction independent of the agency’s decision. If Congress has clearly spoken, the inquiry ends and the court itself identifies Congress’ intent through statutory interpretation. The language of a statute takes on the meaning attributed to it by the court until it is overruled by a subsequent decision or Congress provides clarification by amending the statute. Thus the method of statutory interpretation adopted by a given court – as well as the zealousness with which the court undertakes its analysis – can be determinative of the policy issues on a given topic of regulation. Nevertheless, Congress remains the ultimate controller of agency authority through its construction of a statutory scheme and the level of specificity with which it chooses to word a statute. Where a court determines that Congress has not directly spoken, i.e., that the statute is “silent or ambiguous with respect to the specific issue,” the court then conducts the second step of Chevron analysis. The court’s inquiry in “Step Two” is “whether the agency’s answer is based on a permissible construction of the statute.” Here a court will “afford substantial deference to the agency’s interpretation of statutory language.”
In the words of one academic, “Chevron... is the Court's most important decision about the most important issue in modern administrative law--the allocation of power between courts and agencies ‘to say what the law is.’” With its pro-agency presumption, Chevron deference has changed the landscape of administrative law by circumscribing courts’ substantive inquiry into the rationality of agency policy choices. Courts applying Chevron have encountered numerous difficulties in applying what originally appeared to be a simple doctrine with an on-off switch. Chevron’s seemingly rule-based analysis in practice may entail consideration of numerous variables, thereby converting the rule into a standard. Step One of Chevron implicates traditional debates over statutory construction. When are words plain on their face? Conversely, when should plain meaning be circumscribed based on statutory context and legislative intent? Courts may disagree about the extent to which Congressional sources outside of the statute itself do in fact indicate that Congress has “directly spoken” to the relevant issue. As pointed out by numerous jurists and academics, legislative histories are particularly susceptible to multiple interpretations, allowing judges “to justify a broader range of answers and mak[ing] it easier for them to read their own preferences into the statute.” While Chevron calls for judicial deference to agency decisionmaking, the agency is not free to act as it wishes. Agency discretion “is intended to be constrained by procedures, by substantive law, and by expectations of rationality and openness”; whether these constraints function properly is the responsibility of the courts. Thus, courts may vary greatly in the degree of stringency with which they analyze an agency decision that does warrant deference in keeping with the court’s perception of the operating constraints. Under arbitrary and capricious review, a form of judicial review under Section 706 of the Administrative Procedure Act which applies whether or not Chevron deference is due, a court may take a “hard look” at the processes and justifications surrounding an agency decision to determine whether the agency’s interpretation is reasonable. Alternatively, a court may conduct a cursory inquiry and merely hold that an agency interpretation is not unreasonable in light of the statutory text. As the PhRMA cases arose in 2000 to mid-2002, the reviewing courts who invoked Chevron applied a straightforward Chevron two-step analysis to the Secretary’s waiver decisions.
At issue in the drug benefit cases was the Secretary’s interpretation of the phrase, contained in the Medicaid drug rebate provision, “outpatient drugs...for which payment was made under the State plan” (emphasis added). The Secretary interpreted the above phrase through the Medicaid state plan approval process, including the waiver sub-process; hence, this approval process is the proper unit of agency action for application of the Chevron/Mead/Skidmore analysis. In a series of three opinions beginning with the Vermont lower court, the reviewing courts produced three vastly different approaches to the deference question posed by the Secretary’s approvals.
The courts’ disparate treatments of and outcomes produced on this issue serve as prime examples of several basic difficulties with the Chevron doctrine. The Vermont lower court neglected to employ Chevron even though the case posed a clear challenge to the Secretary’s interpretive authority. This gaff raises the questions of when to invoke Chevron and why courts might miss an application of the doctrine or avoid it altogether. Reversing the lower court, the Vermont appellate court turned to legislative purpose in its Chevron Step One analysis. The appellate court was able to find ample evidence to support its holding that Congress had clearly spoken to the meaning of “payment” where the trial court had found no guidance from Congress. The discrepancy between the courts’ conclusions leads one to ask whether legislative history can serve as a reliable basis for finding that Congress has “clearly spoken.” Finally, the Maine lower court applied Chevron to uphold the Secretary’s approval of Maine’s demonstration project. The Maine trial court opinion implicates the question of deference at the core of the present analysis, namely, whether an agency action is within Chevron’s “domain” and thus deserving of deference if Congress has not clearly spoken.
Leading off the series of cases, the U.S. District Court, District of Columbia completed its analysis of Vermont’s PDP without explicit mention of the Chevron doctrine. The court’s formal omission is surprising given the administrative expertise of the D.C. District Court and the prominence of Chevron in recent administrative law jurisprudence.  Traces of Chevron are implicit in the opinion, nevertheless, and may be mapped against Chevron analysis. In what appears to be a quasi-Step One inquiry, the Vermont trial court cited the absence of Congressional “guidance” as to whether Vermont’s “payment/ reimbursement structure entails ‘payment under the state plan.’” The lower court’s opinion provided no indication of how it had come to this conclusion. Instead, the court proceeded to interpret the reasonableness of the Secretary’s determination that Vermont’s advances do constitute “payment” within the meaning of the drug rebate provision “independent of the Secretary’s directive.”
The court then employed what resembles Chevron Step Two permissibility review. In this step, the court found that the state’s advances could indeed be viewed as “payment,” as Vermont incurred opportunity costs in its advancement of state money to pharmacies. Put another way, Vermont could have earned interest on the state PDP money, left the money with taxpayers, or used it for another program during the time between the state’s advances to pharmacies and the drug manufacturers’ rebate payments to the state. These foregone opportunities themselves have a monetary value that the court found sufficient to constitute “payment.” The court did not analyze the Secretary’s decision making process, but rather considered the reasonableness of the Secretary’s interpretation by independently assessing the substance of the agency’s interpretation.
Taken together, the two steps of the trial court’s analysis may be viewed as putting the court in a position of control over the statute in question. First, by perfunctorily citing a lack of congressional guidance and omitting any further inquiry into text or legislative history, the court made quick work of Congress’ legislative role. The court similarly discounted the function of the Secretary through conducting its own substantive analysis and failing to inquire into the procedural formalities leading up to the Secretary’s decision. After disposing of the legislative and executive branches, the court established itself as the endower of statutory meaning. This version of judicial institutional centrality is in stark contrast to Chevron’s emphasis on the importance of Congress and federal agencies in determining national policy. The court’s unmoored opinion leaves the judiciary exposed to charges of improper policymaking. Whether the court’s interpretation is correct as a matter of logic and desirable as a matter of policy is secondary to question of whether courts as institutions should be making such determinations. In a post-Chevron world dedicated to agency primacy, the Supreme Court’s answer is likely to be “no.”
The Appeals Court, in contrast to and without citing the lower court’s omission, immediately employed the Chevron analysis. Here, the appellate court found Step One determinative: Congress had “’directly spoken to the precise question at issue,’ i.e., “whether ‘payment’ includes expenditures that are fully reimbursed by manufacturer rebates.” The D.C. Appellate Court acknowledged that the word “payment” is broad enough to include reimbursed expenditures. Rather than follow the recent judicial trend of relying on a dictionary to interpret statutes , however, the court refused to employ this broad meaning. The court instead considered the payment provision in terms of its context as embodied in statutory purpose and legislative history. Construing the language in this context, the court decided that ‘”payment’ [in the context of the drug rebate provision] means only payments with state or federal funds appropriated for Medicaid expenditures.” The court made this determination on the basis that “absent such payments, pharmaceutical rebates would not contribute to reducing the cost of the taxpayer-funded Medicaid program.” According to the court, “the legislative history makes quite clear that Congress’s purpose in requiring rebates was to do just that [i.e., to reduce the cost of Medicaid].”
In supporting its reading of the drug rebate provision’s legislative history, the court looked to two sources traditionally used in this mode of statutory interpretation: the introductory statement of the provision’s sponsor and the House Report accompanying the bill. The court selectively quoted from sponsoring senator Pryor’s statement, emphasizing portions in which the Senator stated that (1) the rebates are designed to stop pharmaceutical manufacturers from “’bankrupting the coffers of the State Medicaid drug programs,” and (2) “’there is no logical reason why, in an era of severe budget constraints and social needs, the Medicaid Program should be denied access to [the] same generous discounts’ hospital and HMOs receive.” The court also cited the 1990 House Report on the bill similarly stating that “’[T]he bill is framed to achieve significant Medicaid savings.’” Paralleling its cited portions of Senator Pryor’s statement, the court went on to quote the portion of the House Report declaring that ‘”Medicaid, the means-tested entitlement program that purchases basic health care for the poor, should have the benefit of the same discounts on single source drugs that other large public and private purchasers enjoy.’”
The court then combined and summarized these statements to deduce two overlapping reasons for Congress’s adoption of the drug rebate provision. First, Congress meant to “reduce the cost of Medicaid.” Second, Congress intended to “prevent pharmaceutical manufacturers from charging the government and taxpayers above-market prices for Medicaid drugs.” The court concluded that given this context, “’payment’ excludes situations where no government funds are spent.” Otherwise, according to the court, manufacturer rebates would not “’achieve significant Medicaid savings,’ nor prevent pharmaceutical companies from ‘bankrupting the coffers of the State Medicaid drug programs.’” The court thus held that Vermont’s prescription drug demonstration program did not fall within this clear meaning, as manufacturer rebates would fully reimburse the state’s literal payments under the plan, thereby resulting in no savings for the Medicaid program.
Use of legislative history in statutory construction, like that in the PhRMA II appellate decision, is by now a widely-accepted feature of judicial interpretation. In the words of one academic, the “judicial invocation of extrinsic legislative sources to interpret statutes has bloomed like azaleas in April” since the opening of the twentieth century. Arguments against the use of legislative history exist as well, with the most visible and arguably strongest position coming from the new textualist formalism as exemplified by the jurisprudence of Justice Scalia. A full treatment of the pros and cons of the use of legislative history in statutory interpretation is beyond the scope of this paper. Rather, in light of the still widespread use of legislative history and its specific application by the PhRMA appellate court, the following analysis starts from the point of accepting legislative history as a legitimate mode of statutory interpretation.
As an application of the mode, the Vermont PhRMA court’s discovery of statutory meaning from legislative history is convincing when taken at face value. Further inspection of the drug rebate provision’s history, however, brings to light two grounds for critiquing the court’s method if not its conclusion. First, the court sloppily cites portions of the legislative history of a version of the bill which does not include the explicit language at issue in the PhRMA cases. Congress added the phrase “payment” in a subsequent amendment. Interpretations of subsequent amendments must stand on their own in terms of supporting legislative history. In the present case, subsequent legislative history is silent on the meaning of Congress’ later inclusion of “payment,” a fact that bolsters the court’s conclusion. Nevertheless, the court’s failure to even identify a change in the key language at issue and look to the legislative history of the amendment calls into question the court’s adherence to its charge under Step One: to thoroughly investigate what Congress intended. Similar gaffs in the future – even if made in good faith – may open judges to the charge that they are judicially legislating. Under this critique, judges impermissibly choose among interpretations that are either counter-indicated or equally supported by statutory amendments and subsequent legislative history. Such judicial choice cannot be reconciled under the Chevron doctrine, which requires that a court declare the statute ambiguous where multiple interpretations are supportable under statutory text and legislative history.
Even had the 1990 version of the bill included the word “payment,” the court does the bill’s original legislative history a disservice by neglecting to mention several purposes explicitly stated in or implicitly derived from the cited sources. To bypass stated purposes in favor of educing a testifying senator’s purpose from the full text of his testimony risks substituting a court’s voice for that of the senator. A full-text analysis is arguably more appropriate when examining the intent of the full Congress: each member may express varying degrees of agreement with the stated purposes accompanying the bill. The argument still remains that, when emphasizing the testimony of a single, sponsoring senator, his or her stated purposes are a clear and reliable indicator of intent. As with inaccurate use of subsequent legislative history, incomplete use of legislative history will put judges in a questionable institutional role.
Legislative history may be invoked to determine the purpose of a provision and hence the meaning of a particular portion of the statute. The legitimacy of legislative history is questionable as a basis for judicial interpretation of language not included in the bill version considered or enacted by Congress at the time of the cited history. Negative qualification seems especially convincing if the aim of the analysis is to determine if Congress has “directly spoken” to a particular issue. Here, the exact word in question, “payment,” does not appear until a later amendment of the drug rebate provision.
Senator Pryor’s statement and the House Report quoted by the court both accompanied the drug rebate provision upon its 1990 introduction. At that time, the portion of the provision at issue in Vermont’s case read as follows:
In general.-A rebate agreement under this subsection shall require the manufacturer to provide, to each State plan approved under this title, a rebate each calendar quarter (or periodically in accordance with a schedule specified by the Secretary) in an amount specified in subsection (c) for covered outpatient drugs of the manufacturer dispensed under the plan during the quarter (or such other period as the Secretary may specify). Such rebate shall be paid by the manufacturer not later than 30 days after the date of receipt of the information described in paragraph (2) for the period involved. [emphasis added].
Nowhere in this section did the word “payment” appear. Rather, “dispensed” is the extent of the provision’s qualification of “covered outpatient drugs.” The “payment” language did not come into being until several years later, when Congress reworded the above provision as a conforming amendment to accompany the “Federal Savings” portions of OBRA 1993:
(A) In general. A rebate agreement under this subsection shall require the manufacturer to provide, to each State plan approved under this subchapter, a rebate for a rebate period in an amount specified in subsection (c) of this section for covered outpatient drugs of the manufacturer dispensed after December 31, 1990, for which payment was made under the State plan for such period . Such rebate shall be paid by the manufacturer not later than 30 days after the date of receipt of the information described in paragraph (2) for the period involved. [emphasis added]
Congress added the word “payment” in light of the amended provision’s new allowance for state restriction of drug coverage through prior authorization programs and other categorical means. Under a prior authorization program, a state could now deny reimbursement for an otherwise-covered outpatient drug where the recipient did not receive prior authorization. Likewise, the amendments permitted a state to deny or restrict drug coverage on the basis of a drug’s prescription drug program status and/or medical characteristics. The word “payment” thus presumably was necessary to distinguish drugs which had been merely dispensed to Medicaid beneficiaries from those which had been dispensed under conditions properly qualifying them for reimbursement.
One may oppose reliance on subsequent legislative history altogether and instead argue that the legislature’s intent upon the passage of the original provision should control, as the Supreme Court has held that the use of subsequent legislative history is an “’unreliable guide to legislative intent.’” The D.C. Court of Appeals took just this view in reviewing another Medicaid action by the Secretary of DHHS regarding reimbursement levels for a hospital in North Broward County. There the court refused to consider subsequent legislative history of an amended statute in interpreting the word “such.” North Broward’s holding, however, is not so broad as to encompass all uses of subsequent legislative history. To the contrary, the North Broward court indicated that subsequent legislative history is instructive where there is evidence that the particular interpretation urged by the plaintiff was “the focus of attention of Congress,” the conference committee considering the amendment, or “even the author of the report.”
Returning to the drug rebate cases, the PhRMA court’s inaccurate choice of legislative history thus is formally inappropriate. The above statement by the North Broward court suggests that the D.C. Appeals Court in PhRMA should have at least inquired into the existence of subsequent legislative history speaking to the addition of the word “payment.” The court may have sincerely missed the change in text. Alternatively, the court may have avoided such an inquiry due to the length and complicated nature of the subsequent legislative history in question. OBRA legislative histories are notoriously long and convoluted, as they inherently balance costs of numerous unrelated programs against each other; as such, they may exclude entirely or include less than complete discussions of statutory amendments. Looking for purpose amid an OBRA’s cursory and/or incomplete consideration of its myriad provisions, courts may argue, runs the risk of wasting significant judicial resources.
One may respond to this statement by pointing out that administrative efficiency arguments lose strength when a more searching inquiry might result in a different substantive decision by the court. Here, the 1990 phrasing stands as a weaker basis for the Medicaid savings-driven interpretation of the PhRMA court. The 1990 phrase speaks only of dispensed drugs, as opposed to those dispensed drugs for which a state may deny reimbursement. At the time of the legislative history quoted by the PhRMA court, states were more limited in their ability to control Medicaid costs on bases other than the drug rebates themselves. The argument could be made that this broader notion of coverage lacks the same degree of savings drive as that which the PhRMA court read into the drug rebate provision.
The overall focus of both versions of the statute is still Medicaid savings. Thus, the PhRMA court’s interpretation is justified under either version of the provision. Whether or not the subsequent legislative history changes the court’s interpretation, the interpretation is ultimately most supportable when arrived at through a thorough investigation of legislative history. The efficiency argument, for all of its practical appeal, does not give a principle to guide decisions about which legislative history future courts will examine. If courts are to look back at legislative history at all, they should do so evenhandedly.
ii. Risk of Judicial Policymaking through Selectivity
A court may properly derive statutory purpose from various parts of the legislative history. The stated purposes of the legislation ostensibly are the most appropriate place to begin; otherwise, the judiciary risks usurping the role of the lawmaker by reading purposes into language where Congress’s intent to state a purpose is questionable. By looking to Congress’ explicit purposes in determining the meaning of a particular phrase, the court avoids the dangers of judicial biased selection and interpretation, unreliability, and historical distortion that some textualists argue are inherent to the legislative history project. Explicit purposes, due to their exposed character, also may pose less danger that a staffer has manipulated the history by injecting language under the nose of Congress. Assuming that the 1990 legislative history is the appropriate basis for inquiry in the present case, other portions of the history not cited by the court detract from the court’s interpretation of “payment.” These portions either explicitly provide purposes for the rebate provision or imply alternative purposes.
In the present case, the court failed to cite the five stated purposes included in Senator Pryor’s statement. The five purposes are as follows:
The court-derived purposes of reducing the cost of Medicaid and preventing pharmaceutical manufacturers from charging the government and taxpayers above-market prices for Medicaid drugs are consistent with these purposes. However, they are not the only possible purposes to be derived from the senator’s statement.
Taking the next interpretive step and reading Senator Pryor’s stated purposes within their context, there is another theme expressed in his statement: a desire to prevent the creation of a dual-price pharmaceutical system that discriminates against “poor elderly, minority, and other vulnerable low-income populations who rely on the Medicaid program,” limits their “access to needed medications,” and ultimately results in “a system of second-class medical care for Medicaid patients, the Nation’s poorest and sickest citizens.” This theme is consistent with the overarching purposes of the Medicaid program, namely, “to furnish (1) medical assistance on behalf of families with dependent children and of aged, blind, or disabled individuals, whose incomes are insufficient to meet the costs of necessary medical services, and (2) rehabilitation and other services to help such families and individuals attain or retain capability for independence or self-care." The provision does not seek drug rebates from manufacturers simply to reduce the costs of the program, but to reduce costs in this manner to avoid “undesirable measures to contain costs, such as excluding categories of drug products from coverage and increasing beneficiary copayments.” Reaping manufacturer rebates while further restricting enrollee access to medically necessary pharmaceuticals would comport with the court-derived purposes of the rebate provision, as Medicaid would receive equal pricing treatment and overall costs of the program would go down. Such a strategy of limiting beneficiary access to needed drugs would not, however, necessarily be in keeping with the overall tenor of the drug rebate provision and the Medicaid statute: to run an efficient assistance program in light of the goal of providing necessary medical care to needy persons.
The preceding arguments constitute methodological attacks on the Vermont appellate court’s decision. Another route for attacking its decision lies in a substantive challenge to the court’s conclusion that manufacturer drug rebates can only produce Medicaid savings if “payment” means “payments with state or federal funds appropriated for Medicaid expenditures.” As a matter of logic and syntax, the court’s own statement suggests that Congress has not clearly spoken to the issue presented by Vermont PDP and Maine HMP. If Congress had meant payments with state or federal funds appropriated for Medicaid expenditures, it could have said as much. Congress instead chose the phrasing “payment was made under the State plan.”
Furthermore, the court fails to address the possibility that other types of payments that do not include state or federally-appropriated funds may result in Medicaid savings. The argument here is based on preventive medicine and efficient disease management. If the state extends pharmaceutical benefits funded by manufacturer rebates to otherwise ineligible individuals, those individuals are more likely to receive important drugs early on in their treatment and hence less likely to develop costly medical conditions that require expensive hospital care. By replacing hospital care with less expensive drugs, these individuals will be less likely to spend down their savings to the point that they qualify for Medicaid. Reducing the number of persons who financially qualify for Medicaid reduces enrollment and thereby produces Medicaid savings. CMS itself has endorsed this theory as the policy rationale behind its guidance to the states for developing “Pharmacy Plus” programs that extend Medicaid pharmaceutical coverage. Vermont’s Agency of Human Services likewise justified its waiver application for the extended pharmacy benefits program on the basis of cost savings from avoidance of “costly inpatient and long-term care, particularly among the frail elderly.”
The basic premise of short- and long-term health care cost savings from proper pharmaceutical use has long been the justification for pharmacy benefits on the whole. From this perspective, preventing a new class of individuals from becoming eligible for complete Medicaid coverage by presently incurring the cost of limited pharmaceutical coverage for those individuals will produce not only overall Medicaid savings, but efficient preventive health care. Beneficiaries’ needs for expensive tertiary care will decrease. One may argue that, even assuming this theory’s soundness, the drug demonstration programs are overbroad for two reasons. Not all enrollees will develop high cost medical needs if their access to pharmaceuticals is reduced or eliminated. The small savings from preventing full Medicaid enrollment of those individuals who will experience such illnesses may not exceed the cost of extending pharmacy benefits to the entire class. Second, even persons with significant incomes may enroll in the PDP, persons who would not spend down to the level of full Medicaid eligibility absent PDP coverage. Overbroadness is a serious challenge to both the Vermont and Maine programs, which sought to extend benefits regardless of income or health status to a large group of state residents. CMS notes that states must avoid overbroadness in order to meet the budget neutrality requirement for waivers. In guiding states, the CMS Pharmacy Plus program model caps income eligibility at 200 percent of the FPL (as compared to Vermont’s 300 percent) and recommends using sliding scale schedules that vary with an individual’s income to better target and manage the spend-down population.
In the end, the Vermont appellate court’s reading of the payment provision is highly supportable in light of legislative history accompanying the original and amended versions. Portions of the legislative history clearly emphasize the Medicaid savings thrust of the drug rebate provision. The standard of requiring some state-only contribution of funds towards extended pharmaceutical coverage thus is likely to stand. However, the issue of what constitutes “payment” for purposes of the drug rebate provision is far from apparent under this vaguely articulated standard. It appears that Congress may not have spoken so directly after all. The standard announced by the PhRMA I appellate court gives little guidance to courts faced with programs similar to the Vermont PDP, as it simply eliminates one possible scenario: that in which the pharmacy benefit plan expends no government funds and reaps no Medicaid savings. Thus, the court only identified which pharmaceutical benefits arrangements Congress clearly did NOT intend to cover – and left a gaping whole as to what exactly IS covered. The standard implies an easy path open for state plan developers: outline a plan which expends some government funds and which reaps some Medicaid savings – no matter how minimal those funds and savings may be. Maine’s Medicaid agency immediately took advantage of this loophole and amended its program to include a meager 2% direct government outlay. In the end, the burden on the state and restriction on the Secretary’s discretion is minimal.
Three routes exist for the courts and Congress following the Vermont appellate court’s decision. Courts may revisit “payment” in order to narrow judicial interpretation of Congress’ intent to those programs which involve “substantial” Medicaid savings instead of just savings. Congress itself may amend (1) the payment language to clarify the degree conundrum, (2) the waiver provision to explicitly allow waiver of the drug rebate provision, and/or (3) the state plan provision to mandate or permit the type of state pharmacy benefits program proposed by Vermont. Finally, Congress may abstain from acting and courts may defer to the agency’s interpretation, leaving CMS in charge of further defining “payment.”
A subsequent court believing that Congress intended for the states to bear a greater burden may remove a degree of discretion from the Secretary by focusing on a word quoted by the Vermont appellate court but mysteriously left out of its holding. As discussed above, the Vermont court cited portions of the House Report that identified the rebate provision as “framed to achieve significant Medicaid savings.” Given that the court focused on Medicaid savings as a crucial component of its reading of the legislative history, it is not clear why the word “significant” did not appear in the court’s holding. A future court may reasonably latch upon this omission in its own Chevron Step One analysis and supplement the previous court’s holding such that a permissible program must include expenditure of significant government funds and significant Medicaid savings. While permitted under the Chevron doctrine, such further judicial elaboration may not be desirable from the judiciary’s perspective as it toes the line between interpreting and making law. In practice, a standard that includes “significant” might not produce results that differ greatly from a standard without it, as the term implies some quantity less than the Congressional favorite “substantial.” It may, however, give both the state Medicaid agency and CMS pause to consider whether the proposed level has heft or whether additional state funds might be included. Along these lines, the state agency may have more political weight with state legislatures
The PhRMA II trial court faced the payment question without the aid of subsequent congressional amendment (and, with the current political disagreement over pharmacy benefits reform at the federal level, such amendment is unlikely to occur at anytime in the near future ). Armed with the Vermont appellate court’s resolution of Chevron Step One, the Maine trial court thus had to decide whether to take the first route of further defining the statutory term or the third option of granting deference to the Secretary’s interpretation. The same court which failed to find a congressional directive in the PhRMA I case with the same judge writing for the majority decided PhRMA II at the trial level. Given this continuity, it is not surprising that the PhRMA II trial court opted not to look further into legislative history, but rather to defer to the Secretary in approving the extension of benefits. Deference is desirable from the judiciary’s standpoint, as it insulates judges from having to enter into areas of uncertain technical policy. In conducting a deference analysis, the court had to determine whether the Secretary’s interpretation regarding the Maine program was due heightened Chevron deference. This inquiry invokes an existing controversy surrounding the Chevron analysis: whether Chevron even applies to certain types of informal agency actions. The PhRMA II trial court found grounds for deference. By failing to inquire into the amount of process granted by the Secretary in coming to his decision, as opposed to the substance of the decision as written, the court’s rationales for deference are inadequate under current administrative law doctrine.
Recent Supreme Court decisions expanding on the Chevron doctrine have added a new query to the Chevron two-step analysis applied by the PhRMA II trial court: whether the agency action is of a type entitled to Chevron deference. This question is distinct from the Chevron Step Two question of agency reasonableness in light of congressional ambiguity. Commentators phrase the former question in terms of whether the agency action is within Chevron’s “domain.” The justification for recognizing a distinction between agency processes which deserve heightened Chevron deference and those which do not lies in the idea that expertise and accountability have procedural analogs. In other words, the more formal process that occurs, the more likely it is that the agency will produce a well-considered decision based on sound justifications, thereby eliminating the need for extensive judicial oversight. Two identifiable justifications lie behind a refusal to grant more informal actions full Chevron deference. First, such informal actions lack the “legislative-type procedures” that occur in notice-and-comment rulemaking , and thus are less transparent and possibly less well-reasoned. Agencies engaging in informal actions also may not be acting pursuant to delegated lawmaking authority.
Over the nearly two decades since deciding Chevron, the Supreme Court has held that certain agency actions affording more formal process are presumptively within Chevron’s domain. These actions include formal adjudications , formal rulemaking and (informal) notice and comment rulemaking. The proper treatment of less formal actions is murky under current circuit court and Supreme Court jurisprudence, although the Court has provided some relatively straightforward guidelines on this issue. For example, opinion letters, policy statements, agency manuals and enforcement guidelines fall outside of Chevron’s domain. Other informal actions which do not qualify for Chevron deference may still receive deference under a less permissive totality of the circumstances standard as put forth in Skidmore, a case which pre-dated Chevron. According to Skidmore, agency decisions may be granted deference in proportion to their “power to persuade.” For a time following Chevron, doubt existed as to Skidmore’s vitality: did Chevron’s rule of heightened deference eliminate the need for the Skidmore standard? A recent and high profile Supreme Court opinion introduced the Chevron domain standard, held that Skidmore is alive and well, and affirmed the appropriateness of Skidmore analysis in cases where Chevron deference is not warranted. Medicaid waiver demonstration decisions arguably present such a case.
Legal academics have read the Mead opinion as expanding Chevron into a four step analysis. After asking whether Congress has clearly spoken, the court must then define the scope of the agency’s statutory authority. This new Step Two is the point at which the court decides whether to continue the Chevron analysis. If the process provided for by Congress does not meet some threshold of formality, thus qualifying it for Chevron deference, the court will consider the action under the pre-Chevron Skidmore balancing test. Step Three consists of asking whether the agency’s interpretation fell within its statutory authority. Finally, Step Four is equivalent to Chevron Step Two.
In the United States v. Mead, the Court was faced with the question of whether to grant deference to an agency action in the form of a tariff classification ruling letter. Under regulations promulgated by the Secretary of the Treasury pursuant to statutory rulemaking authority, the letters (1) were to be applied only to the transaction of the moment, (2) were not subject to notice and comment before issuance, (3) were subject to modification or revocation without notice, (4) were not required to be published, (5) and could not be relied on by subsequent requestors. Letters could be issued by any one of forty-six Customs offices, as well as the Customs Headquarter Office; most letters contained “little to no reasoning.” It is in this context that the Court articulated a new two-part test. To qualify for Chevron deference, the agency must show that (1) “Congress delegated authority to the agency generally to make rules carrying the force of law,” and (2) the agency acted pursuant to this authority in issuing the interpretation for which deference is claimed. Congress’ intent to authorize an agency action with the force of law may be inferred from “the agency’s generally conferred authority and other statutory circumstances.”
Defining the contours of this standard, the Mead Court identified the products of statutorily-authorized rulemaking or adjudication as warranting Chevron deference. The rationale for deference here is that more formal processes tend to “foster the fairness and deliberation that should underlie a pronouncement [having the force of law].” The Court then further implied by reference to its prior holdings that the products of notice-and-comment rulemaking and formal adjudication would receive Chevron deference. In a move that introduced huge uncertainty into the Chevron standard, the Court went on to state that the absence of notice-and-comment rulemaking is not determinative, as the Court has “sometimes found reasons for Chevron deference even when no such administrative formality was required and none was afforded.” The Court then held that, due to the tariff letters’ circumscribed present binding force, weak precedential value, independent review provision, lax agency practice, shear numerousness and low-level authority of issuing agency representatives, no Chevron deference was due. Instead, the tariff letters were “best treated like ‘interpretations contained in policy statements, agency manuals, and enforcement guidelines.’” The letters were, in other words, “’entitled to respect... but only to the extent that [they had] the ‘power to persuade.’”
With this lead-in, the Court finally turned to Skidmore analysis. To find that an agency action does not fall within Chevron’s domain, stated the Court, “is not... to place [it] outside the pale of any deference whatsoever.” Rather, “Chevron did nothing to eliminate Skidmore’s holding that an agency’s interpretation may merit some deference whatever its form’” based on the “‘specialized experience and broader investigations and information available to the agency,” as well as “the value of uniformity in [the agency’s] understandings of what a national law requires.’” An agency thus may raise a Skidmore claim under circumstances involving a highly detailed regulatory scheme and an evaluating agency that can claim the benefit of specialized experience. The Court outlined several factors relevant in analyzing cases for which Chevron deference is not due, including “the merit of [the ruling’s] writer’s thoroughness, logic and expertness, its fit with prior interpretations, and any other sources of weight.” Rare among the present Court’s decisions, Mead was a seven-to-one majority opinion; Justice Scalia provided the lone dissenting voice.
In digesting Mead, four categories become apparent for identifying agency actions intended by Congress to have the “force of law.” The first two categories reemphasize the importance of formal process in determining the level of deference due to agency actions. First, one must ask whether the agency statutorily may engage in adjudication; if so, Chevron deference is appropriate. The text of Mead does not explicitly limit this category to formal adjudication, thereby leaving open the possibility that the prima facie category includes informal adjudication. However, there are two arguments against the inclusion of informal adjudication as a prima facie case for heightened deference. As noted by analysts of Mead, the Court cites only instances of formal adjudication in its note accompanying the above-quoted language on adjudication. The exclusion of informal adjudication from the text of the APA also weighs against Chevron deference here, as agencies are left with no guidelines as to the minimum amount of process that must be provided in informal adjudications. Including informal adjudications along with formal adjudications, for which the APA does outline specific procedural requirements, in the prima facie category would be inconsistent with the Mead Court’s focus on mandatory procedural protections. Academics writing on the subject interpret the Court as identifying the products of formal adjudication as presumptively warranting heightened deference; to the extent that they view Mead as calling for deference for informal adjudications, they do so by inference. In addition to formal adjudicatory authority, statutory authorization to conduct formal and/or informal notice-and-comment rulemaking also calls for heightened deference. Both the adjudication and rulemaking authority categories arguably include non-explicit statutory grants of such authority, both due to Mead’s reference to implied authority and because the APA provides definitions by which non-explicit congressional authorizations may be identified as either adjudication or rulemaking.
The Mead Court further identifies a third general category of other qualitative and quantitative aspects of Congress’ authorization which, when taken together, may weigh in favor of deference. These factors, cited above, include precedential value, provision for independent review, numerousness of actions, and identity of the agency representative ultimately responsible for the action. Finally, the agency’s practice as it reflects the first three categories constitutes a fourth category which may weigh for or against heightened deference. The agency may, in other words, qualify for deference by voluntarily conducting adjudication or rulemaking, or treating its determinations as having precedential value. The exact balance between categories three and four, relative weighing of subfactors within these categories, and possible existence of other subfactors are all matters left unclear in the Mead opinion: tariff classifications were found so lacking in all of the noted process factors as to create a clear case against heightened Chevron deference.
Mead was decided a mere ten days after the D.C. Circuit issued its decision in PhRMA I. Since Mead, the lower courts have decided several cases involving informal agency actions. As stated above, the Supreme Court itself has not laid an explicit framework for informal adjudications – a category which may, as discussed below, include waiver approvals like those at issue in the PhRMA cases. In sum, the proper amount of deference to grant the Secretary’s decision to grant a Medicaid waiver extension is an open question.
The waiver statute does not explicitly authorize the Secretary to engage in adjudication or rulemaking when making a waiver determination. In fact, the main waiver provision omits any reference to process whatsoever, instead outlining the broad substantive terms governing waiver approval. Congress thus cannot have authorized the agency to engage in formal adjudication or formal rulemaking, as such authorization must be explicit in terms of its provision for a hearing that occurs “on the record” or its functional equivalent. Nor has Congress given courts further guidance in characterizing waiver decisions for the purposes of determining the appropriate level of judicial review. Circuit court decisions reviewing HHS’s approvals of state AFDC and Medicaid waiver applications under Section 1115, in turn, have provided little explicit guidance of their own as to which APA term best describes the waiver approval process. At a minimum, an AFDC waiver approval is reviewable and not “committed to agency discretion by law” based on the program’s “complex and detailed regulations” and the statute’s overall lack of “congressional commitment to the unfettered discretion of the Secretary.” Courts reviewing Medicaid waivers implicitly assume that this holding applies in the Medicaid context as well. A waiver approval is also an “informal agency action” for which a court "is not empowered to substitute its judgment for that of the agency." Beyond identifying waiver determinations as informal agency actions, courts have not characterized waiver decisions as either adjudications or rulemakings.
The Maine trial court delivered the final substantive opinion to date on the drug demonstration programs. In its opinion, the court addressed the Mead question: whether the Secretary’s waiver approval is a form of agency action that falls within Chevron’s domain and thus is due deference within the realm defined by Congress’ statutory language. This question was explicitly left outstanding by the Vermont appellate court. The Maine appellate court avoided the deference question on a technical ground, holding that the program in question was never formally before the Secretary and hence its approval was not ripe for review.  Perhaps signaling its leaning on the issue of Chevron’s domain as applied to the demonstration program approvals, the Vermont appellate court prefaced its Step One analysis by stating that “Of course, not all agency interpretations of statutes warrant Chevron deference.” The court here cited Christiansen, a case in which the Supreme Court limited Chevron’s applicability. According to the Christiansen court, agency interpretations in opinion letters, as well as those in policy statements, agency manuals, and enforcement guidelines, do not merit Chevron deference.
Bypassing the procedural point later relied upon by the Maine appellate court in granting PhRMA’s injunction, the Maine trial court analyzed the Maine program’s 2% payment in light of the Vermont appellate court’s holding. The trial court independently assessed the 2% payment to determine whether it fell within the meaning attributed to the term “payment” by the Vermont appellate court’s prior Step One analysis. The court, after assessing the “facts of this case,” concluded that the Maine 2% payment did fall within the court-interpreted category of “’ payments with state or federal funds appropriated for Medicaid expenditures.’” Specifically, the court found significant the state’s showing that “’the rebates under Maine's HMP are expected to amount to $5,451,494.00 annually..., whereas total state-only appropriations amount to $19,750,292.00 annually,’” as well as that “’the two percent increase in non-DEL payments cost[s] the [S]tate between $500,000.00 and $1 million dollars per year.’” On the deference question, the Maine trial court concluded that the agency process under which the Secretary made his decision constituted “formal agency regulations that apply to all cases and have the force of law.”
The Maine trial court’s characterization of the agency action, delivered with little investigation into the actual process at hand, deserves scrutiny. Subsequent courts facing the question of deference to informal agency actions like waiver determinations must now abide by the approach outlined in Mead. Mead tells reviewing courts that judicial review in this area proceeds in two tiers: first Chevron, then if necessary Skidmore. The initial question in choosing between Chevron and Skidmore under Mead’s first prong is whether Congress has delegated authority to the agency to act with the force of law. A court’s analysis of the first prong should advance through the four categories outlined above to inquire whether (1) Congress has authorized the agency to engage in formal adjudication, (2) the agency has statutory authority to conduct formal or informal rulemaking, (3) other factors suggest that Congress intended the action to have the force of law, and/or (4) agency practice takes on a threshold degree of procedural formality. Under Mead’s second prong, a court must then assess whether the agency action was promulgated pursuant to such authority. An agency action which fails either prong is then subject to Skidmore review.
The reviewing courts in the PhRMA decisions correctly identified the Secretary’s waiver decision as the level at which to apply the Chevron domain question. Section 1315(a) announces the standard that the Secretary must apply in considering whether to grant a waiver:
In the case of any experimental, pilot, or demonstration project which, in the judgment of the Secretary, is likely to assist in promoting the objectives of subchapter I, X, XIV, XVI, or XIX of this chapter, or part A or D of subchapter IV of this chapter, in a State or States--
(1) the Secretary may waive compliance with any of the requirements of section 302, 602, 654, 1202, 1352, 1382, or 1396a of this title, as the case may be, to the extent and for the period he finds necessary to enable such State or States to carry out such project, and
(2)(A) costs of such project which would not otherwise be included as expenditures under section 303, 655, 1203, 1353, 1383, or 1396b of this title, as the case may be, and which are not included as part of the costs of projects under section 1310 of this title, shall, to the extent and for the period prescribed by the Secretary, be regarded as expenditures under the State plan or plans approved under such subchapter, or for administration of such State plan or plans, as may be appropriate, and
(B) costs of such project which would not otherwise be a permissible use of funds under part A of subchapter IV of this chapter and which are not included as part of the costs of projects under section 1310 of this title, shall to the extent and for the period prescribed by the Secretary, be regarded as a permissible use of funds under such part.
Through the waiver process, the Secretary then explicitly interprets the waivable sections of the Medicaid Act and either explicitly or implicitly construes all other nonwaivable provisions. In other words, the Secretary cannot approve a waiver program on the basis of its waivable portions alone, but must find all other portions in keeping with nonwaivable Medicaid requirements. A grant of approval in which the Secretary is silent as to nonwaivable provisions implies that the Secretary has construed the components of the state’s request as being consistent with these provisions. The Secretary also may explicitly interpret the nonwaivable provisions by addressing them directly in the final approval; the Secretary may base such interpretation on the state’s responses to questions posed during the approval process or sui generis . Most importantly for present purposes, Section 1315(a) is silent as to any procedural standards that apply to initial waiver request. The only procedural requirements for Medicaid waivers explicitly laid out by Section 1315 appear later in 1315(f); these minimal requirements, as taken up below, apply only to “continuations” of existing Medicaid waivers.
Congress’ lack of explicit authorization to engage in formal adjudication or formal rulemaking when making waiver decisions eliminates several options on which to base Chevron deference under categories one and two. There still remains the possibility that Congress intended for the Secretary to decide waiver requests pursuant to the Secretary’s general rulemaking authority under the Medicaid statute. If the waiver program as outlined by Congress qualifies under the APA as rulemaking, the Secretary may claim deference based on this initial characterization under Mead and obviate the need for a more in-depth Prong I inquiry. A counter argument exists that the waiver program is an informal adjudication. If Congress intended for the Secretary to conduct informal adjudications in granting waivers, the Secretary cannot substitute his or her general rulemaking powers without violating the will of Congress. An informal adjudication view of waiver process would require a multi-factor Prong I inquiry in keeping with Mead, as Mead does not include informal adjudication as presumptively evidencing Congress’ intent to have the agency act with the force of law. Which argument wins out in the end is not clear.
On the one hand, the determination is an informal adjudication vis-à-vis the state’s application. The APA defines adjudication as “agency process for the formulating of an order”; an order, in turn, “means the whole or a part of a final disposition, whether affirmative, negative, injunctive, or declaratory in form, of an agency in a matter other than rulemaking but including licensing”; licensing “includes agency process respecting the grant, renewal, denial, revocation, suspension, annulment, withdrawal, limitation, amendment, modification, or conditioning of a license”; and finally, a license “includes the whole or a part of an agency... approval [or] statutory exemption...” Tracing this path, a waiver action may be characterized as an adjudication in that it is a final disposition of a licensing action in the form of an approval or statutory exemption. The Medicaid waiver provision includes no hearing requirement that would place a waiver determination under the APA’s requirements for formal adjudication ; therefore, the determination is an informal adjudication. Informal adjudications are an enigma under the APA, as the statute does not directly speak of informal adjudications as it does formal actions and informal rulemaking.
Perhaps due to this uncertainty of what informal adjudication actually is, the Supreme Court has yet to announce the deference standard for informal adjudications. Disagreement exists among academics as to whether informal adjudications deserve Chevron deference. One writer believes that post-Mead, informal adjudications will get deference based on the logic of the Mead opinion emphasizing that Chevron deference is owed when the agency is “making law.” Another writer, while agreeing that Mead’s logic calls for deference for informal adjudications, criticizes the opinion itself, lamenting that the Court has created a “cumbersome, unworkable regime under which courts must draw increasingly fine distinctions using impossibly vague standards.” According to this critic, the concerns in Mead with informal actions can be resolved in a less troublesome manner by treating Chevron Step Two review as hard look review under the APA’s arbitrary and capricious standard.  Both Mead and hard look review focus on the adequacy of the agency’s explanation. By using hard look review instead of Mead’s totality of the circumstances plus vague “other” category, the courts would leave policymaking authority in the executive branch, as the remedy for an inadequate explanation would be a remand to the agency.
Notably, the U.S. District Court for the District of Columbia, the court of decision in the PhRMA cases, is one of the few courts that has considered deference for an informal adjudication. The court wrote its decision in Fontana vs. Caldera less than a month after Mead came down. In contrast to the PhRMA cases, the court had Mead to guide its analysis and found in favor of deference. According to the Fontana court, whether an action is labeled informal adjudication is immaterial to the deference question: the correct query under Mead is whether Congress delegated the authority to make decisions having the force of law. The Fontana court then proceeded to apply a Mead “totality of the circumstances” test to determine whether the case presented some indication of congressional intent comparable to that evidenced by authority to conduct adjudication or rulemaking. As the statute in question made clear that the Secretary’s orders were to be “final and conclusive on all officers of the United States,” the court granted Chevron deference. In sum, if waiver decisions are best characterized as informal adjudications, it appears that they are subject to Mead’s totality of the circumstances test. They may qualify for deference under Mead’s more searching inquiry, but cannot presumptively receive heightened Chevron deference as can formal agency actions or even informal rulemaking. Hence, under Mead, the PhRMA II trial court improperly applied Chevron Step Two. Informal adjudication, however, does not sufficiently capture the nature of a waiver determination.
Waiver decisions, while arguably qualifying as adjudications under APA § 551, appear significantly different from both the tariff classification letters in Mead and the separation from service date determinations in Fontana. The waiver decision is not only a determination with regards to the immediate state application at hand analogous to an approval of an individual private party’s application for statutory exemption, a rate determination, or a ruling on status. In another sense, a Medicaid waiver approval sanctions an extensive state welfare program which itself has rule-like qualities by backing the program with federal dollars. The produce of cooperative federalism, the approved waiver may be taken as a rule or regulation adopted by both the federal and state governments. According to the APA, a rule is “the whole or a part of an agency statement of general or particular applicability and future effect designed to implement, interpret, or prescribe law or policy or describing the organization, procedure, or practice requirements of an agency...” A Medicaid State Plan fits this general description, as it is the statement of the state Medicaid agency on how the agency will administer its Medicaid program. It thereby has a future effect on the legal rights of Medicaid beneficiaries and providers, as well as on the state’s ability to collect federal funds to support its program. Absent an approved state plan, the federal Medicaid agency cannot stand in the state agency’s place to develop and implement a state-level plan. The federal government instead signals its approval by appropriating money to support the state program’s administration.
Under an informal rulemaking view of the waiver process, the state acts as a partner and an arm of the federal government: the state agency conducts the fact-gathering and drafting steps that the federal government would otherwise undertake in notice-and-comment rulemaking. The Administrator of CMS and Secretary of HHS then sign off on the state’s proposal after providing their own comments and receiving the state’s responses, just as they would in approving a final rule developed solely within the federal agency. This dispersion or delegation of functional role is justified on the basis that, as noted above, the state has traditionally been responsible for matters of health within its borders. The state as proper assessor in certain areas of policy is a basic tenant of federalism, under which the states operate as “laboratories of democracy.” Most importantly, the state is the level at which interested parties will participate in public comment sessions regarding health benefits in their own state – benefits which may be of little interest to other states and/or national interests. Another analogy focusing on the importance of centralized review would be to think of the state as an outside consultant hired by the federal government to assist in the rulemaking process. As a formal matter, the Secretary does have general rulemaking authority under the Medicaid Act.
The appropriateness of an analogy between the waiver process and informal notice-and-comment rulemaking is supported by HHS’s self-imposition of procedures for consideration of waiver applications. In 1994, the agency released a notice in the Federal Register outlining procedures that it intended to apply to waiver proposals. These procedures approximate notice-and-comment rulemaking by including provision for state and federal notice and comment, findings, and the development of an administrative record. Included in the record were to be (1) the formal demonstration application from the state, (2) issue papers sent to the state and the state’s responses, (3) Public and Congressional comments sent to the Department and any responses, (4) the Department’s decision memorandum, and (5) the final terns and conditions of the waiver grant. The 1994 notice creates a dual level of procedural responsibility with the states providing some form of public procedure in addition to that at the federal level. However, the procedures do not bind the agency to the same extent as would a statutory direction to undertake informal rulemaking: HHS limited the waiver procedures by stating that the notice “does not create any right or benefit, substantive or procedural, enforceable at law or equity, by any person or entity, against the United States, its agencies or instrumentalities, the States, or any other person.” It is not clear why the agency elected not to invoke its general rulemaking authority to create more binding waiver process requirements. The Secretary’ authority to do so, nevertheless, is plain from the Medicaid statute; thus, under a rulemaking view of waivers, the agency passes Mead Prong I. If the agency actually bound itself to quasi-notice-and-comment rulemaking, waiver approvals granted after the conclusion of the processes outlined in the 1994 notice should, like the products of notice-and-comment rulemaking, receive Chevron deference under Mead Prong II.
Faced with these arguments in favor of categorizing a waiver as an adjudication and as a rulemaking, a waiver decision cannot easily be characterized as either. Thus, without a clear label to attach to waivers, courts must delve deeper into the substance of Congress’ authorization in accordance with Mead. By looking more closely at waivers, courts may be able to determine whether waiver decisions are actions intended to have the force of law.
While Mead leaves open the question of what actions have the “force of law” beyond the products of formal adjudication and formal and informal rulemaking, the qualitative and quantitative factors cited in Mead as weighing against Chevron deference are instructive. Here is the “totality of the circumstances test” that the court references in Fontana. The relevant Mead factors are that the action has circumscribed present binding force, weak precedential value, and provision for independent review. Also pertinent are the number of decisions produced and the nature of the agency representative who actually issues and holds responsibility for the decision; these factors also come into play in the agency practice category. Mead uses strong language to limit the force of law category to small numbers of decisions which are issued by a limited number of centralized agency representatives with significant authority within the agency.
Circumscribed present binding force refers to the range of persons legally affected by the particular decision at hand. The Mead Court provided two factors that may be grouped under the heading of circumscribed present binding force. They are that the agency action: (a) “respond[s] to transactions of the moment,” and (b) may be modified or revoked without notice to any person except the applicant without an opportunity to comment. Like tariff classifications, waiver requests face case-by-case analysis in which the Secretary is not bound to approve state plans that follow the same format as previously-approved plans. The Secretary is free to approve, request modification of, or deny approval of any waiver request that he or she finds, in his or her “judgment,” does not promote the objectives of the Medicaid statute. This discretion to approve or deny ostensibly allows the Secretary to approve a project in one state while denying an analogous project in another. The Secretary may base a determination on any number of factors such as differences in population number, demographics, available providers, relative costs of health care, or just about any other characteristic weighing on the implementation of a health program.
It is questionable under the waiver provision whether the Secretary may modify or revoke a decision without notice to any person but the applicant. As a preliminary matter, it appears that the Secretary has substantial discretion to modify the terms of a waiver once he or she has given approval: the text of Section 1315(a) authorizes the Secretary to waive compliance “to the extent and for the period he finds necessary to enable such State or States to carry out [a waiver demonstration project].” The provision does not require the Secretary to abide by his or her past determinations going forward. Nor does the provision mention notice, either to the state or otherwise, upon a subsequent decision by the Secretary that the waiver no longer promotes the objectives of the Medicaid program or that a waiver is no longer required for such promotion. However, as a product of cooperative federalism, the waiver approval itself arguably acts more like a contract between bargaining equals than a government tariff rate imposed on an applicant. Thus it appears in practice that the Secretary may not unilaterally change the terms of a waiver decision outside of a negotiation process with the state. This is not to say that the Secretary may not revoke a program approval for failing to comply with the terms of its approval, but merely that the Secretary may not be able to change his or her mind on the prudence of the waiver program midstream without involving at least the state. On the subject of third party notification, CMS does impose a non-legally binding obligation on state agencies to conduct some form of public hearing regarding an initial waiver proposal. Once the state and federal agencies agree to a modification, however, there is no requirement that other persons be informed of that change before a waiver is approved: the results of the “closed-door state-federal waiver negotiations” may be adopted without further notice to or comment from third parties. Nor do states have a duty to inform potential and current Medicaid beneficiaries of changes in program services or eligibility. In sum, the Secretary apparently may modify or revoke a waiver decision without notice to third parties, thereby circumscribing the decision’s present binding force. This degree of independence argues against Chevron deference.
Whereas circumscribed present binding force looks at a single action’s scope at a solitary point in time, weak precedential value connotes the impact that the decision will have on future persons involved in similar cases. Under Mead, weak precedential value includes that (a) future persons may not rely on the decision, and (b) the agency is not bound in any way to apply the principles of that decision in future cases. The Mead Court did not clarify whether the degree of precedential value must be statutorily specified or whether it may be evidenced by agency practice, thereby blurring the bounds between this inquiry and category four. As stated above, the waiver provision grants the Secretary the authority to exercise his or her “judgment” in making waiver determinations. This language implies that determinations should be made on a case-by-case basis according to the particulars of the state and the proposed program at hand. No state is guaranteed approval because it follows the logic of the questions-and-answers exchange between CMS and an approved prior applicant in designing its current proposal. With such individual consideration, it appears that waivers do not have significant precedential value.
Nevertheless, in contrast to the Secretary of the Treasury’s warning against reliance in tariff classification system, CMS makes no explicit statement against reliance and does view past applications as instructive in making its waiver determinations. CMS creates guidelines for state proposals, in part based on what it views as successful waiver programs in other states. While guidelines are not legally binding, they may be relevant to the Mead inquiry in that the CMS treats them as binding by committing to give expedited treatment to and/or approve proposals that follow the guidelines. Furthermore, absent or prior to the promulgation of guidelines, CMS in fact approves nearly identical programs in different states as evidenced by the Vermont and Maine pharmacy waiver approvals. A recent study of waivers by the U.S. General Accounting Office states that several states have waivers under consideration that are similar to the handful of approved waivers, and that review of these requests “will likely be influenced by prior decisions and precedents.” The agency’s actual endowment of past waivers with precedential value argues in favor of a finding that waivers have the “force of law” under the totality of the circumstances test.
Ironically, CMS’s practice may weigh against deference. The agency’s abundant reliance on past waiver programs in the evaluation of new proposals and willingness to approve copycat programs raises concerns that the agency is going beyond its statutory authority to approve demonstration programs. Congress originally passed the waiver provision in 1965 as a means for states to try out experimental programs of limited scope. An added benefit of demonstration programs on top of their informational value for the states would be their service as models for future Medicaid reform at the federal level. If the agency is exceeding its statutory authority, the waiver will be analyzed under Skidmore’s “power to persuade” test following Mead (see below).
The existence of a higher level of independent review which treats the initial agency action as having little weight also argues against designation as having the force of law. In Mead, tariff ruling letters could be appealed to the Court of International Trade, which treated the decisions on the same level as other Secretarial rulings considered to be unimportant beyond the individual case. Waiver approvals, in contrast, are not subject to independent review. Under the CMS regulations governing State plan approvals, a state may file a request for administrative review with the Administrator of Medicaid regarding the Administrator’s “action on plan material” ; logically, this right of internal review should extend to the waiver context as the waiver program stands to become part of the State plan upon approval. Following review by the Administrator, the Secretary’s determination is a final administrative decision which may then be appealed within the federal court system as evidenced by the PhRMA cases. Appeal usually is pursuant to the APA § 706, which charges a court to “’hold unlawful and set aside action, findings and conclusions found to be... in excess of statutory jurisdiction, authority, or limitations, or short of statutory right,’” as well as to set aside agency actions that are “’arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’” Again in contrast to tariff classification letters, waivers appear to warrant Chevron deference.
In Mead, tariff ruling letters were issued on the scale of 10,000 to 15,000 per year by forty-six different point-of-entry Customs offices, as well as the Customs Headquarters. The Mead Court emphasized that most tariff letters were terse and lacked reasoning. In combination, these concerns appear to add up to a great apprehension: that subordinate employees with little expertise, no public accountability, and little contact with each other were issuing the letters containing poorly reasoned determinations. To give these persons the power to make law to which courts should defer would fly in the face of the principles of administrative law, in that arbitrary and unjustified decisions were almost guaranteed by the system.
Initial waiver approvals and continuations stand in stark contrast to this avalanche of tariff letters. A recent report by the U.S. General Accounting Office states that “only a handful of demonstrations have been approved to date” ; another report puts the number of statewide Medicaid waiver approvals in effect through FY2001 at only seventeen. Seventeen states are developing or considering waivers in FY2003 under the new expedited procedures available through the Health Insurance Flexibility and Accountability Initiative, while eighteen states are developing pharmacy waivers. Considering that these two groups overlap, and even given the existence of waiver development in other areas, the total number of waiver applications is low when compared to tariff classification letters. More importantly, tight high level administrative oversight exists for the agency’s waiver review process. A state must submit a highly detailed report to HHS, which then refers the report to the Centers for Medicare and Medicaid (CMS). Next, a technical panel conducts an in-depth review of the report and scores the report on multiple factors. The technical panel then makes a recommendation of approval, conditioned approval, or rejection; the Office of Research and Development within CMS then incorporates the panel’s recommendation into a decision memorandum for submission to the Administrator of CMS. Finally, the Administrator makes the decision to approve or not and must directly receive the Secretary’s final approval. On the topic of the ultimate decisionmaker, academics have suggested that an agency’s choice of how to internally delegate decisionmaking power should be a factor in determining whether the agency’s actions merit Chevron deference under Mead. Those agencies who, as in Mead, allow lesser agency staff to issue decisions cannot claim that such decisions have the force of law. In the present case, the highest agency actor must ultimately make the waiver decision. Chevron deference appears due under this highly-centralized system charged with responsibility over a small number of waiver requests.
A challenge to heightened deference here may be raised if one views Mead’s emphasis on numbers and level of authority as being a proxy for careful agency consideration. In other words, the Mead Court may be read as being concerned with agency overload and lack of oversight, and the concomitant effects on quality of decisionmaking. The same overload concern may apply to waivers when the size of each waiver request is taken into consideration. In comparison to a tariff classification, which ostensibly requires a quick economic analysis of a single product, a waiver proposal includes myriad proposals and subproposals which the agency must evaluate individually and in the context of the whole request to determine their consistency with the Medicaid statute’s objectives. As noted above, the Secretary implicitly interprets the non-waivable provisions of the Medicaid statute in granting a waiver; the Secretary must also take into account the many ways in which the detail of a given waiver proposal might conflict with Medicaid’s objectives. Multiplying the number of factors and subfactors to be weighed times the number of waiver requests submitted may raise the total number of agency decisions to the overload threshold. Furthermore, many of these sub-issues will be considered by lesser agency staff such as the project officer assigned to each waiver request; whether the Secretary actually considers each issue, rather than simply signing off and deferring to the judgment of the lower agency representative, is questionable. With Mead as the only point of comparison, waivers still appear to be on the safe side of overload and poor oversight: to attain the same level of decisionmaking as in Mead, each waiver determination would have to include in the neighborhood of 27,000 factors for consideration , and a the existence of a few project teams within the central agency office hardly compares to a scattering of forty-six regional offices.
Raw comparison aside, there are strong arguments on either side of this Mead factor when applied to waivers. Health policy analysts criticize CMS for what they view as substandard, perfunctory analysis by an agency focused on increasing the speed and number of approvals. These critics specifically cite the pressures exerted by the past two White House administrations in terms of expediting the waiver process. Notably, HHS issued the 1994 Federal Register notice in response to concerns about the agency’s expedited approval of several controversial waiver programs in the early 1990s. Others cite that the agency, with its main levels of waiver review and detailed state negotiation process, exhibits a high level of care and oversight. Without more empirical data on the actual review of specific waivers, it is difficult to conclude whether numerousness and location of decisional authority weigh in favor or against Chevron deference.
Whether the agency actually engages in the decisional process with an apparent intent to make law (regardless of statutory authority to do so) is relevant to the force of law analysis. Here, Mead lists several telling factors. Agency behavior that “makes clear that a [decision]’s binding character... stops sort of third parties” and that regards the decision as “conclusive only as between [the agency] and the [party to which the decision was issued]” evidences no intent to engage in lawmaking. The agency may also clearly express its intent to not make law by warning parties “against assuming any right of detrimental reliance.” On the other side, a court should look at the amount of process that the agency attributes to each determination. If the agency voluntarily undertakes notice-and-comment rulemaking, the court may impute an agency desire to make law. As noted above, third parties are not directly bound by the Secretary’s waiver determination: each state must seek an individual decision on its waiver request. CMS does not warn states against detrimental reliance, but rather holds successful state projects out as models for other states and publishes guidance notices outlining formats for acceptable waiver programs. These factors all weigh in favor of heightened deference.
A significant argument exists, nevertheless, that agency practice rules against heightened deference for Medicaid waivers: CMS has systematically failed to ensure a consistent level of process and public access regarding waiver applications at both the federal and state levels. According to one recent study of waivers, pressure has built over the years to “open up” the waiver process due to a waiver’s impact on both a state’s Medicaid program operation and millions of dollars in state and federal spending. DHHS responded to this pressure by publishing the 1994 notice cited above, essentially outlining a quasi-notice-and-comment rulemaking process for consideration of waiver applications. DHHS’s self-imposed process for waiver approvals in the 1994 notice does not have the force of law. Thus, the agency is free to abide by the process or not without facing legal sanction based directly on the notice.
The same study, which covers waiver proposals through FY2001, notes that the Department has not abided by the 1994 notice. The General Accounting Office waiver study voices a stronger concern that DHHS is failing to consistently follow the agency’s own stated policy to provide persons who may be affected by a waiver an opportunity to learn about and comment on a proposal. For example, the Department no longer publishes monthly notices; the last available Federal Register notice dates from 1998. Nor does the Department routinely make waiver documents available to the public, but instead refers inquiries to the states. CMS also has not created a system that ensures that it receives public comments prior to making a waiver determination, although public comment and meetings with federal officials have occurred in several cases where the public was made aware of a waiver proposal. The Department’s primary route of providing public input into the waiver process consists of delegating this responsibility to the states: the Department does consistently require state certification that the state has provided an opportunity for public comment before submission of a proposal to federal officials. Under this arrangement, states have significant discretion as to how to meet the public comment requirement.
Critics of CMS’s delegation note that the state public process options sanctioned by CMS vary tremendously in the amount of access that they provide to the public. In a recent letter from the Director of the Center for Medicaid & State Operations to state Medicaid directors, the agency gives the following examples of acceptable public input processes:
Comparing only the last two options highlights how great the difference in access between acceptable public processes can be. Furthermore, individual states implement the options in various manners and to varying degrees. At the low end of the scale, several states effectively provide little to no public input whatsoever, possibly due to lack of resources. Other states have scaled back the amount of process that they originally proposed to give to waivers. Advocates concerned with a state’s discretion in public process note that such flexibility may be politically applied to shield unpopular proposals and trumpet popular ones.  The new HIFA initiative, described below, further reduces the procedural burden on the states. Where once states had to describe their public processes in detail to CMS, now they need only check a box certifying that they followed a public process in which beneficiaries and other interested parties had an opportunity to comment. Finally, and perhaps most importantly, the federal-state agency set of negotiations for determining the final set of terms and conditions is not open to the public and there is no public record for tracking the exchanges as they happen.
Congress has recognized the need for more formal process at the state and federal levels. In October of 2002, Senator Baucus and 23 co-sponsors introduced a Medicare reform bill that included a section providing for more process in the waiver program. The section mandates a system of state-level notice and opportunity to comment, a reporting system under which states inform CMS as to the details of their chosen process route, and extensive federal notice, comment, information disclosure, and statement of bases requirements. Whether the bill is successful remains to be seen; linked as it is to other Medicare reform proposals and given the Congress’ current inability to pass Medicare reform bills , the prognosis is not good. Thus, CMS is likely to be left holding the responsibility for implementing waiver procedures, either through a legally binding rulemaking pursuant to the Secretary’s general rulemaking authority or by more stringent enforcement of the 1994 notice.
Congress and the federal agency may explicitly or implicitly create the conditions that give an agency action the force of law and hence make the action deserving of heightened deference. The primary ways in which Congress may do so include (1) authorizing the agency to conduct formal adjudication, formal rulemaking, or informal rulemaking, and/or (2) providing certain statutory characteristics to the action itself that mirror these categories. The relevant characteristics include wide present binding force, strong precedential value, lack of independent review, and limited issuance by a highly centralized and responsible ultimate decisionmaker. In turn, the agency has the ability to endow its actions with the force of law by acting in a manner which tracks these characteristics even absent congressional specification. Finally, to warrant Chevron deference under the Mead standard, the action must have been promulgated under the agency’s delegated authority.
In the case of waivers, Congress has not explicitly endowed the Secretary with the authority to engage in formal adjudication or formal rulemaking. The Secretary may undertake informal rulemaking in issuing waivers, but has not chosen to do so in a manner approximating notice-and-comment rulemaking. Without clear indicators as to usual procedural categories which warrant Chevron deference, one must turn to Mead’s other factors in assessing whether waiver approvals – in the form of either informal adjudications or very informal rulemaking (i.e., less than notice-and-comment) – are intended to have the force of law. Three out of the four Mead characteristics suggest that waivers are due heightened deference. The agency treats waivers as having some precedential value, there is no statutory provision for independent review of an agency waiver decision, and the agency issues only a handful of waivers each year subject to the approval of the Secretary at the advice of the CMS Administrator. Weighing against Chevron deference, waivers have circumscribed present binding force. In addition, two of the three Mead factors supporting Chevron deference can be read the other way. First, waivers’ precedential value is not statutorily required and so the agency may choose not to follow prior determinations subject only to arbitrary and capricious review. Second, waiver requests implicate myriad sub-issues which presumably undergo evaluation by inferior agency staff on whom higher agency representatives rely without conducting their own independent analysis. Mead does not provide guidance as to how courts should balance these competing characteristics; thus one may look to agency practice for a possible tiebreaker. Here, while agency practice evidences some intent to issue actions having the force of law, the agency’s delegation of discretionary procedural authority to the states and near complete abdication of responsibility for any form of public participation at the federal level strongly weigh against heightened deference. Mead’s focus on procedure suggests that, absent clear indications otherwise, agency actions like waivers that issue without the benefit of sufficient process do not merit Chevron deference under the force of law inquiry.
The above discussion of precedential value touched upon a concern that CMS is acting beyond the authority granted to the Secretary under the waiver provision. This concern implicates the very question to be answered under Prong II of Mead: whether the agency action at issue was promulgated pursuant to the congressionally delegated authority identified under Prong I. In 1962, at the time of the initial passage of Section 1115, the Senate gave guidance that "’a demonstration project usually cannot be statewide in operation’ and that the Secretary was to ‘selectively’ approve state proposals .” Critics of the waiver program’s implementation hold that the Secretary’s practice of granting waivers for increasingly long-term and large-scale programs exceeds the agency’s authority to approve “experimental” waivers. According to one legal academic, the “widespread health reform” achieved under Section 1115 waivers “represents a broad extension of the statute's reach” in a way that may be beyond what is legally permissible. Proponents of the Secretary’s use of waiver authority in a manner that give states significant flexibility in health policy matters may argue that the waiver provision still confers great discretion on the Secretary as evidenced by the phrases “in the judgment of the Secretary” and “to the extent and for the period he finds necessary.” This wide discretion, however, must be exercised within the prior limit that the program must fit the description of “experimental, pilot, or demonstration.” In sum, courts should find against Chevron deference where the scope of the program under consideration exceeds these descriptive terms. Both the Maine and Vermont programs were state-wide and represented temporal extensions of the same concept of pharmacy benefits for expansion populations that had been previously implemented through other waivers. A reviewing court thus may find that the Secretary exceeded his statutory authority in approving the states’ pharmacy benefits programs.
The scope of the Secretary’s waiver authority has prompted much debate within Congress. On the one hand, to allow an administrative agency to go so far beyond its initial grant of authority in reshaping the Medicaid program is to displace Congress in its role as Medicaid reformer in violation of Chevron’s call for congressional primacy in policymaking. Some members of Congress view such displacement as an improper allocation of power among the branches of government. To whit, Representatives Dingell, Waxman, and Brown’s comment that “[t]he 1115 authority was not intended to be a blanket circumvention of statutory or regulatory guidelines or Congressional intent” and call for careful Congressional oversight of waiver determinations. A broadening of federal agency authority to grant waivers within a cooperative federalist scheme, however, paradoxically shifts more power to the states to determine the content of Medicaid. Members of Congress who believe in the new federalism may thus support a broadening of the Secretary’s authority that ostensibly empowers the states such as that seen in President Bush’s 2001 Health Insurance Flexibility and Accountability Demonstration Initiative (HIFA). The main thrust of HIFA is to give states more flexibility and control in their Medicaid programs. A further discussion of the Secretary’s authority, federalism, and Medicaid reform is taken up below. Suffice to say, the current broad wording of the waiver provision does little to limit the Secretary’s discretion to increase the scope of Medicaid waivers beyond that originally intended by the 1962 and 1965 Congresses. Critics of a strategy that uses the waiver program to increase states’ power under Medicaid should seek reform of the waiver provision to further define “experimental, pilot, or demonstration project.” The agency’s current waiver practice of giving past programs precedential value where both the initial and subsequent proposals are reasonably characterized as experimental does qualify waiver determinations for heightened deference.
The text of the waiver provision brings to light another area of concern with regards to the scope of the Secretary’s waiver authority: whether the Secretary is improperly approving expansions of existing waiver programs through expedited procedures. Two issues arise with respect to this question. First, expedited review may consist of a reduced level of process which risks conflicting with Congress’ express view of when expedited process is appropriate in the waiver context. Second, by choosing to forego more formal procedures in favor of expedited review, the Secretary is not acting pursuant to the office’s general rulemaking authority. The Secretary thus has failed to invoke the implicit authority to produce actions having the force of law as discussed above. Under either reading, Chevron deference is inappropriate.
Expansions of existing waivers to new groups of beneficiaries require full waiver review. In contrast, Congress made explicit provision for a more circumscribed process for waiver continuations , which are essentially temporal extensions of existing waiver projects. Under 1315(e), the chief executive officer of a state operating a Medicaid waiver program approved under 1315(a) may make a written request for an extension of the project for up to three years. Subsection 1315(e) also governs the substance of the continuation, stating that “the extension of a waiver project under this subsection shall be on the same terms and conditions... that applied to the project before its extension under this subsection.” This “same terms and conditions” provision is substantively conditioned by only one other provision: if federal budget neutrality was a condition of the initial approval, the Secretary is charged with taking any steps necessary to ensure the program’s continuing federal budget neutrality over the period of time covered by the extension. Thus, the Secretary may vary the terms and conditions of the initial waiver that impact federal budget neutrality. Nowhere does Subsection 1315(e) make allowance for substantive program modifications beyond those required for continuing federal budget neutrality or changes proposed by the state itself.
Section 1315(f) is a procedural provision outlining the steps that the federal agency must take in approving a 1315(e) continuation request. Review by the Secretary is of “the waiver project.” This choice of wording indicates that the Secretary must review the terms and conditions of the initial approved waiver program and decide whether the approved program should move forward in its existing form given the requirements of 1315(a). In other words, the Secretary conducts a retrospective review of whether (1) the program accomplished what it set out to do and can continue to do so in the future, or (2) based on its years of operation, the program has the potential to accomplish its goals and simply needs more time. There is no indication in the statutory language that the Secretary is authorized to review any new terms and conditions proposed by the state in its extension application; otherwise, the statute would direct the Secretary to review “the extension request.” Under the statute’s text, the Secretary serves as the initiator and source of new terms and conditions, as supported by the statute’s charge to “inform the State of any proposed changes in the terms and conditions of the waiver project.” Ostensibly, the scope of the Secretary’s proposed changes is limited to those outlined in Section 1315(e), i.e., terms and conditions that ensure federal budget neutrality. Once informed, the Secretary and the State engage in negotiation over the proposed terms and conditions. Section 1315(a), as stated above, constitutes the substantive standard to be applied to all waiver requests. In contrast, 1315(e) makes special provision for continuation waivers and 1315(f) defines the process for approval of continuations.
In recent years at the behest of the White House (both the Clinton and second Bush administrations), CMS has been progressively speeding its review the waiver process. Reductions have included reductions in the amount of information that states must provide to CMS regarding state-level public process and in the time allotted for responses between the state and federal agencies, as well as the introduction of form-type applications. Expedited review of initial waivers under 1315(a) may begin to resemble waiver continuation review under 1315(f). To the extent that this bleeding together of processes occurs, the agency may be exceeding its authority. Congress, in creating a specific expedited process for approval of the less-involved waiver continuations, implicitly held the Secretary to greater process regarding review of expansion programs. The expedited process for initial waiver requests thus may impermissibly cut back on the amount of process to be accorded to states and argue against Chevron deference.
The Medicaid statute gives the Secretary general rulemaking authority to “make and publish such rules and regulations, not inconsistent with this chapter, as may be necessary to the efficient administration of the functions with which each is charged under this chapter.” As discussed above, this authority may serve as grounds for Chevron deference to waiver determinations under Prong I of Mead. Possessing general rulemaking authority, while potentially necessary for Chevron deference, is not a sufficient basis on its own to justify heightened deference. Rather, the Secretary must act pursuant to that authority under Mead’s Prong II. The Secretary has never exercised general rulemaking authority in the context of Medicaid waivers. Without an affirmative move by the Secretary to self-impose the rulemaking process, the Secretary cannot claim heightened deference under Mead.
Under a Mead force-of-law and scope-of-authority analyses as applied above, it is likely that a subsequent court reviewing the waiver programs like those in the PhRMA cases will find against Chevron deference. The court must then engage in a Skidmore analysis to determine the decision’s power to persuade. Relevant to Medicaid waiver cases, Skidmore claims may be raised where an agency with significant expertise acts pursuant to a highly detailed regulatory regime. Here the factors for consideration are “the merit of [the ruling’s] writer’s thoroughness, logic and expertness, its fit with prior interpretations, and any other sources of weight ” (emphasis added). Skidmore, like Mead, thus is a multi-factor test with an open category through which judges may express their views on what makes a sound agency decision. In the context of Medicaid demonstration waivers, the Skidmore open category should include the amount of public process afforded at both the state and federal levels.
Several aspects of the Medicaid program and CMS weigh in favor of Skidmore deference. The Medicaid statute, while having undergone substantial amendment over the past decades, has existed in its current cooperative federalist form since the statute’s enactment in 1965. Through detailed federal statutory requirements and even more specific agency regulations, the longstanding program covers nearly every aspect of health care policy from the general concept of medically necessary care to the minutiae of reimbursement for health services. Huge manuals published by CMS and the states occupy shelf space on health providers’ and regulators’ bookcases across the country; these manuals serve as reference guides on a daily basis on such topics as coverage, hospitals, outpatient physical therapy, state operations, home health, and skilled nursing facilities. HCFA, now CMS, has had primary authority for overseeing the Medicaid program since 1977. The agency’s responsibilities include “assur[ing] that the Medicaid, Medicare and SCHIP programs are properly run by [the agency’s] contractors and state agents, establish[ing] policies for paying health care providers, conduct[ing] research on the effectiveness of various methods of health care management, treatment, and financing, and assess[ing] the quality of health care facilities and services and taking enforcement actions as appropriate.” Several standing committees provide information to the Secretary in the areas of physician practice, Medicare coverage, Medicare education, competitive pricing, ambulatory classification groups, and special negotiated rulemaking. The breadth and depth of CMS’s experience in health policy matters under this detailed regime can hardly be questioned.
CMS also has specialized experience with regards to the specific issue in the PhRMA cases. Vermont’s pharmacy benefits program did not present with a novel policy: the state directly based its proposal on the federal Medicaid drug rebate program, a statutory scheme which had been in existence for nearly ten years by the time the D.C. Circuit decided PhRMA I. Additionally, the Vermont PDP sought to build on an existing waiver program by expanding the class of individuals eligible for pharmacy benefits under the states’ already-approved pharmacy benefits waiver. The Maine HMP mirrored the Vermont program. As CMS had a decade of experience with drug manufacturer rebates at the time of the Secretary’s waiver approval, the decision appears to merit some deference under Skidmore. Significant interaction between CMS and the state agencies with responsibility for Medicaid further weighs in favor of Skidmore deference. Ostensibly, although it cannot be gleaned from the PhRMA record or documents available through the agencies, the state and federal agencies engaged in some informal development discussions prior to the waiver submissions. A paper trail of questions posed by CMS and responses from the Office of Vermont Health Access also serves as evidence of the federal agency’s thoroughness and logic with respects to the PDP. Since the PDP was an extension of Vermont’s pre-existing Medicaid pharmacy waiver program, the question-and-answer exchange regarding this prior program may be counted in favor of the agency’s careful consideration of the PDP.
A parsing of “expertise” shows that Skidmore deference is less warranted, absent public procedural protections, in cases involving a cooperative federalist program focused on experimental policies developed by the states and modified by a federal agency. First, the notions of expertise and experiments are to some degree theoretically inconsistent. Experimental programs by definition present a level of unknowing and risk that expertise cannot address; as noted by one writer, “any given reform plan will likely unfold in unpredictable and complex fashions.” CMS has an interest in approving and tracking the progress of a demonstration project for the very reason that the agency wishes to improve its expertise in a given area where no clear data can be gathered absent the project. The same can be said for the state agency. Second, from the viewpoint of cooperative federalism, one agency’s expertise at a particular level of government is insufficient to answer the full range of questions that will arise with respects to a regulatory program. While CMS may have had experience with the federal drug rebate program and Vermont’s pre-existing pharmacy waiver program, it cannot be said to have had direct experience with extending Medicaid pharmacy coverage to a large group of state residents whose household income is significantly above the poverty level. Vermont, in like manner, cannot entirely step into CMS’s shoes as experts. The state lacks the federal agency’s expertise regarding the impact of drug rebates on the national pharmaceutical market and the effect of multiple different state programs on the Medicaid program as a whole. Third, neither agency can claim expertise matching that of third parties in several areas important to pharmacy benefits and pricing. From the supply side, drug manufacturers have considerably more experience in the business aspects of drug development, pricing, and markets than do the agencies. Pharmaceutical consumers, a.k.a. potential beneficiaries, also have better information than do agencies about what drug prices their individual household budgets can sustain. Similar third party expertise is likely to arise in areas other than pharmacy benefits.
An agency decisionmaking process that excludes these crucial third party voices can hardly be said to be thorough. Related to the exclusion of third parties (which would ostensibly increase the development time for proposals), indications that the waiver development process occurs rapidly at the state level also argue against a finding of thoroughness. A development and review process that fails to take into account the relevant concerns of third parties is logical only within an incomplete sphere. Nor is it sufficient for a single agency at a particular level of government to serve as the sole repository of expertise and/or the only access point for public comments in a cooperative federalist scheme. For these reasons, courts analyzing Medicaid waiver determinations in the future should look to the amount of process and degree of public participation afforded by the agencies before granting deference under Skidmore.
A CALL FOR PROCESS
The shortcomings of the Secretary’s waiver decisions as brought to light by the preceding Mead and Skidmore analyses suggest that significant changes in the waiver program should take place before courts defer to agency waiver determinations. The two main areas for reform are the scope of the Secretary’s authority and the amount of process that CMS and the states grant to waiver determinations. As the waiver statute, legislative history and related case law give little guidance on how to precisely define an “experimental, pilot, or demonstration program,” the question of scope of authority will be left for a future time when Congress and/or the courts speak to the issue. The agency itself has spoken on process by issuing the 1994 notice; the Supreme Court in Mead again has emphasized the importance of process in determining the degree of deference owed to agency actions. Hence, the issue of process is ripe for consideration. In addressing the proper amount of process in waiver determinations, the following section attempts to answer why both state and federal process are desirable, and why more stringent requirements are necessary at both levels.
The basic arguments in favor of process – that process enhances quality decisionmaking, democratic participation, accountability etc. – are well-supported tenants of administrative law. Assuming that more process stands to improve the waiver approval process along these lines, an additional question arises in this particular instance: why both federal and state process? CMS has waffled on this question, indicating at different times that both are necessary in the Medicaid context (as in the 1994 notice) and alternatively that state process alone is sufficient (by its adherence to only the state requirements of the 1994 notice). A preliminary answer is that Medicaid is a cooperative federalist program and thus, as both federal and state agencies hold significant responsibility under the statutory scheme, both should be held to procedural norms. A fleshing out of this argument and several more arguments in favor of process at both levels of government follow.
a. Waivers and Medicaid Reform
At a basic level, as a provision within a larger cooperative federalist scheme, the waiver program necessarily implicates a national concern: the proper role of the federal government in setting minimum standards for Medicaid. Federal oversight plays an important role in assuring some uniformity of health care quality and access for beneficiaries which may be lost when comprehensive reform takes place under state waiver programs. It is difficult to imagine what is gained in this debate from shutting out other states and interested parties during the negotiation of an individual state’s waiver program. Those parties who believe in a strong federal role may wish to encourage a permissive review process and/or argue in favor of the state under review, while others may wish to participate to bolster the federal agency’s case. Regardless of their preferences, these parties should be given a voice where questions of the federal government’s role in health policy are implicated.
The need for a federal forum also comes about because of the inter-state distributional effects of Medicaid. While states have traditionally held responsibility for health-related matters within their borders , the Medicaid program extends federal financial support for state medical assistance programs. For one state to implement a program qualifying for federal funds means that the viability of a program in another state may be threatened in the next congressional year, as Congress may decide to scale back certain services that are federally reimbursable under Medicaid in response to growth in another service area. Each state – as well as beneficiaries, providers, and other interested parties in individual states – thus has an interest in what Medicaid programs operate in other states. CMS requires as a condition of approval that states demonstrate the “budget neutrality” of their waiver programs, i.e., that the program will not cost the federal government any more money than if the waiver were not in place. Budget neutrality prevents the waiver program from aggravating the allocational dilemma described above.
Another national distributional concern, however, arises even given the budget neutrality requirement. Different states may have varying levels of concern about the legislative appropriateness and political impacts of waiver approvals at the federal level. States who view the current federal statutory requirements as correctly setting the national minimum standards for Medicaid may see extensive waiver approval as amending the Medicaid statute through an impermissible sidestep of the legislative process. A more desirable and institutionally sound route to Medicaid reform, according to this view, would be to have the entire Congress representing the various states work out the details in a multi-party deliberative setting. At the same time, advocates of the current federal scheme may worry that generous and rash waiver decisions will send an inaccurate signal for Medicaid reform to Congress. Here the concern is that the waiver programs of individual states will put the core mandatory federal Medicaid program at risk for other states. Congress, looking to CMS’s waiver approvals as indicators of the direction in which Medicaid policy should be going, may restrict certain other classes of mandatory beneficiaries or services in order to enact and fund national requirements mirroring the waiver programs. Other states and out-of-state interested parties thus have an interest in presenting their concerns with a particular state’s waiver proposal through a federal agency process where all voices – in-state and out-of-state – have an opportunity to comment in the same forum.
CMS often looks to past waiver applications and approvals to guide its individual decisionmaking, as well as to formulate guidelines that apply to categories of waivers. Thus, states who intend to submit a certain type of proposal in the future have an interest in how the agency treats proposals currently under review. A state may wish to support the approval of another state’s analogous waiver request to improve its own future position with the agency. Conversely, a state may see value in opposing the approval of a proposal that it views as casting a negative light on its proposal in the agency’s eye. Nor are states the only entities with concerns about precedent in waiver approvals: the same concern extends to all interested parties who may each be said to have an interest in creating a positive track record before the agency. For example, as will be taken up below, drug companies may be concerned about approval of Vermont’s pharmacy benefits program on two different levels. They may disagree with the prudence of extending the Medicaid rebate provision to new beneficiaries as to the individual state as a matter of law and of economic principle. In other words, drug manufacturers may wish to argue before CMS that the Medicaid statute does not allow such extension (the argument in the PhRMA cases) and state price controls of any sort will disrupt the market function for pharmaceuticals. At the same time, manufacturers may not be so worried about the market impact of granting rebates to Vermont, but with the effects of approving the subsequent stream of other states’ waiver requests for identical programs. The pharmaceutical market may be able to sustain a pharmacy benefits program in a tiny state like Vermont. Whether the market can sustain analogous programs in Maine, Maryland, New Hampshire and any other new applicants is a doubtful prospect which drug manufacturers may wish to bring to the agency’s attention prior to initiating litigation.
Arising from the precedential value of waivers, a fairness argument from a state’s perspective says that similar waiver programs should be treated similarly. Under a fairness view, where CMS denies one state’s waiver request and approves a subsequent similar application, the former state is due a public explanation of the agency’s bases for unequal treatment. Such an explanation will not only serve to clarify the record for administrative purposes, but has the additional policy benefit of giving the first state an indication of how to rework its own prior rejected request.
In contrast to physician services, the markets for which are traditionally viewed as local , pharmaceuticals are a national market. Several large branded manufacturers and a group of smaller generics sell standardized drugs to wholesalers and retailers by negotiating discounts from their average wholesale price (AWP). While pharmacy benefits managers (PBMs) may serve as more local price intermediaries in these negotiations, the prices that they ultimately receive are tied to the national AWP. What happens in one state pharmacy program in terms of price thus will likely affect the price of drugs across the nation. A recent report by the Government Accounting Office noted this ripple effect in its consideration of extending federal drug rebates to the Medicare program:
Mandating that federal prices for outpatient prescription drugs be extended to a large group of purchasers, such as Medicare beneficiaries, could lower the prices they pay but raise prices for others. Such price changes could occur because drug manufacturers would be required to charge beneficiaries and federal purchasers the same prices. To protect their revenues, manufacturers could raise prices for federal purchasers. Furthermore, because federal prices are generally based on prices paid by nonfederal purchasers, manufacturers would have to raise prices to these purchasers in order to raise the federal prices. In particular, large private purchasers that tend to pay lower prices, such as health maintenance organizations (HMO) and other insurers, could see their prices rise.
Analogous concerns arise when the government purchaser mandating the rebate is a state government leveraging tens of thousands of consumers (a projected 70,000 in Vermont’s case and 225,000 in Maine ).
The national character of the pharmaceutical market suggests that public access to the federal program overseeing approval of state drug programs under Medicaid is highly desirable. Creating a federal waiver process accessible to the public would allow the full range of interested parties to engage in the important debate about affordable prescription drugs before an agency with accumulated expertise in the area of pharmacy benefits. These parties may include other states representing large numbers of privately insured residents who might face increased drug prices because of the waivers, beneficiaries from states who do not have such programs in place, and drug manufacturers themselves – whose interests in the Secretary’s waiver approvals were recognized by both the Vermont and Maine courts under the standing doctrine , as well as CMS itself in its questions to Vermont’s health agency.
Inclusion of drug manufacturers via federal level process is especially important given that their interests may be over looked at the state level, where most comments are likely to come from beneficiary advocates and direct care providers. CMS itself made such an omission in its letter to State Medicaid Directors by specifying that states may provide sufficient opportunity for public input on waivers by making “presentations to enrollees, families, private providers and public interest representatives,” or “forming a special commission with broad representation including recipients, families, private providers, and public interest representatives.” One may argue that this omission at the state level is appropriate and that pharmaceutical manufacturers hardly need representation at the federal agency level given the strength of their lobbying efforts in Congress. However, to the extent that Congress has lost oversight control of the waiver process, may wish to defer to the policy expertise of the Secretary of DHHS or touts state flexibility in Medicaid policy, drug manufacturers deserve another point of access to the policymaking process.
That waiver decisions in one state may impact the market for certain health care services in another market is not an argument restricted to pharmaceuticals. In addition to national health care markets, regional markets exist for health care services such as physician and hospital care. Hospital service populations do not stop at state borders; physicians who practice close to state lines are likely to treat patients who travel from out-of-state. Furthermore, it cannot be assumed that neighboring states are as friendly towards each other on health care matters as are Vermont, Maine, and New Hampshire. From these fluid markets and the potential interstate impacts of waiver decisions, it becomes apparent that some degree of federal mediation should be required.
Another fairness argument, this time from the position of non-state interested parties (beneficiaries, providers, etc.), focuses on the flexibility of states in deciding how much process to provide at the state level. The argument is as follows: where all state requests must ultimately go through the federal approval process, it is unfair that some interested parties, i.e., those living in states with more generous views of public participation, have significantly greater access to the federal agency via their states’ processes. The options for public participation sanctioned by CMS vary widely in the degree of notice, opportunity to comment, and state agency explanation that they afford interested parties. A simple newspaper announcement with no further explicit procedural requirements apparently is acceptable; at the opposite end of the spectrum, a state may opt to provide full notice-and-comment rulemaking procedures in keeping with its state statute governing administrative procedures. While notions of federalism indicate that states should have a say in deciding how much process to grant during waiver development at the state level, the variation resulting from states’ choice calls for implementation of a federal process which equalizes access to the final arbiter across the states.
b. Financial Resources and Efficiency
A focus on budgets and institutional capacity brings to light another argument in favor of supplementing state process with federal procedural requirements. Given a system of state flexibility in choosing procedures, budgetary concerns become prominent. Currently, states are experiencing enormous budgetary difficulties and thus are looking for places to cut corners. Administrative process, being a politically discrete area of spending, may be an easy target for budget cuts. In these lean times, a federal process can serve as an important back-up to scaled down state processes. Moreover, CMS has substantial experience in the national level rulemaking arena as evidenced by the long list of Medicaid implementing regulations coming from its offices. CMS also has the benefit of being privy to the various implemented state public processes through the state certification requirement in the 1994 notice and may adopt what it views as the most successful in engaging interested parties. Thus, the federal agency is potentially more capable and efficient than the individual states in conducting waiver processes for state-based, out-of-state, and national interested parties.
The acceptable options listed by CMS in its letter to state Medicaid directors range from notices in newspapers to formal notice and comment under a state’s administrative procedures statute. A recent Senate bill calling for improvement in the process for the “development and implementation” of waivers adopts an approach similar to the more stringent requirement of notice and comment. In outlining a state’s procedural responsibilities, the bill mandates notice to a medical care advisory committee (“committee”) established by the state plus any other individual or organization who requests such notice, as well as a meeting of the committee to discuss the proposed plan and any modifications prior to the State’s publication of notice. The bill also requires that the state subsequently publish for written comment, in accordance with the state’s procedure for issuing a regulation, a notice of the waiver proposal that includes (1) information on how the public may submit comments to the state, (2) a statement of the state’s projections of “the likely effect and impact of the proposal on any [eligible beneficiaries,]” and (3) a statement of the state’s projections as to the “likely effect and impact of the proposal on any providers or suppliers of items or services for which payment may be made” under the Medicaid Act and the state’s assumptions underlying the projection. At the same time, the state must post the proposal and any modifications on the state’s internet site and directly provide notice to the committee and requesting individuals. Following publication of the required notice, the state must hold at least one open meeting of the commission for consideration of the waiver proposal and modifications. At least two public hearings on the proposal and any modifications also must occur before submission to CMS. The state must keep a record of all public comments submitted in response to the notice or at meetings and hearings. Finally, the state must provide CMS with a detailed description of the public process used to develop the proposal along with copies of all required notices, dates or all meetings and hearings, and a summary of the public comments received in response to notices or at hearings and meetings. If the state fails to comply with these procedural requirements, the Secretary must return the waiver proposal to the state without action.
Calling for the states to provide this level of process presumably arises from two conclusions. First, the bill’s framers must believe that process at the state level is desirable. The framers must hold the further belief that a high degree of mandated process at the state level is preferable to state flexibility within a large range of alternatives. A basic premise of federalism backs the first belief: that states are more accessible and accountable on matters within their traditional realm of operation. Concomitantly, interested parties will seek to participate at the nearest, i.e., state, level on matters that hit close to home. A forum that focuses on traditionally state matters will serve them best, as it will be less congested and distracted by competing federal concerns. Health has long been within the province of the states and, in absence of federal initiative, states have led innovations in health policy throughout recent years. State health agencies and health sections of the state legislatures have longstanding relationships with advocacy groups in their respective states, as such groups keep the government institutions attuned to the needs of the Medicaid beneficiary class. For example, the Office of Vermont Health Access and Vermont Legal Aid worked together during the development period to address issues arising from the pharmacy benefits program.
A concern with legislative distraction and congestion additionally supports mandated procedures led by a government body with specific responsibility for health policy generally and waivers specifically. In other words, a health agency may best hear the concerns of interested parties and incorporate them into the waiver proposal. Agency specialization and its ability to produce responsive processes argue against leaving state waiver process to the governor and/or legislature who, like Congress, are likely to have many other issues on their minds. Finally, as discussed above, several of CMS’s options (especially the presentation and special committee routes) risk leaving important voices out of the development process.
Senate Bill 3018 codifies both state and federal procedural requirements. The bill takes its lead from processes specifically recognized as valuable by the federal agency itself in the 1994 notice. In a recent letter responding to Senator Max Baucus, Secretary Thompson addressed the senator’s concerns with public access to waiver information by listing the ways that CMS has begun to make the waiver process more open. Based on the above reasons, and in light of these multiple acknowledgments that more process at the state and federal levels is a good thing for waivers, the agency should stand by its 1994 commitment to process by formalizing the stated requirements in a binding agency rule.
Ultimately, whether the Secretary’s approvals of the Maine and Vermont pharmacy benefit programs receive judicial deference under Mead Prong I should turn on the actual amount of process afforded in each case and not on a theoretical comparison to what the Secretary has said the agency will do or what measures Congress should enact. The court’s inquiry should take into account process at the state as well as federal agency level. CMS relies on states in large part to develop health policy, and neither level of government should be able to avoid public scrutiny by hiding behind the other. Furthermore, some degree of public participation at the federal level is desirable where state policy proposals involve matters of national significance. Drug coverage is such an area, as the market for drugs is made up of national suppliers who devise their pricing policies based on activities in the many states. At the same time, state level process is important in that it is closer to the ground and may draw out voices not heard in a federal debate. These voices may include those persons who are unaware of or intimidated by federal agency processes; they may also represent parties who eschew federal involvement in state policy. State and federal officials could not be reached for further elaboration on Vermont’s and Maine’s state processes at the time of this writing.
Legal questions about statutory interpretation and the proper degree of deference to give administrative agencies obscure several basic policy questions arising from the Vermont and Maine cases. The Vermont appellate court refrained from addressing policy issues, stating that “[o]ur task... is neither to evaluate the PDP’s policy justification nor to determine whether the program best serves the pharmaceutical needs of the poor.” By passing on these questions, however, the court struck down a program which many states view as desirable policy (evidenced by the number of states seeking approval for similar programs). In avoiding policy by referencing congressional intent, the court left it to Congress to amend the statute if Congress ultimately disagrees with the policy outcome occasioned by the court. That the courts will face other pharmacy waiver proposals in the future is a given considering the sizable gap left by the Vermont appellate court’s holding. A look into the policy questions thus is appropriate in thinking about the future of drug rebate waiver programs in the legislatures and courts.
Three main policy questions arise from the Vermont and Maine cases. First, is extension of Medicaid-based drug manufacturer rebates to a large percentage of a state’s population a desirable way to solve the dual problems of high pharmaceutical prices and lack of insurance coverage for pharmaceuticals? Next, is a “carve-out” of pharmacy benefits desirable from a health policy perspective? The final question raises a highly contentious issue in both the political and legal worlds: is the Medicaid waiver process, both in theory and practice, an appropriate method of implementing health policy? The Chevron doctrine is structured, although questionably so applied, to avoid having the judiciary decide just these difficult policy issues. Administrative process, nevertheless, is crucial to answering both questions of proper health policy and government structure.
In responding to the question of drug prices and coverage, one must look at state efforts to create a new, limited purpose class of Medicaid beneficiaries through programs like Maine HMP and Vermont PDP, as well as to leverage drug manufacturer rebates outside of the Medicaid program. The latter strategy may be seen in the Maine Rx case currently before the Supreme Court. Both policy choices seek to affect the same end, namely, to use large-scale government programs to reduce pharmaceutical costs to state residents. Due to the present focus on the Medicaid waiver approval process, this paper will address only the first type of program.
An argument in favor of Vermont’s and Maine’s programs derives from the basic premise of federalism that there is benefit in states acting as “laboratories of democracy,” especially in areas of traditional state expertise like health. Experimentation per se is not the only good arising from this position. A view of states as laboratories sees them as especially capable of choosing among competing policy ideas for the good of their own constituencies, as evidenced by the Medicaid statute’s grant of significant discretion to states to “identify[ ] eligibility rules and benefit levels that reflect their idiosyncratic preferences and capacities.” In addition, states-as-laboratories provide models of successful policies for other states. The states-as-laboratories notion is appropriate in the context of health care reform for several reasons. These rationale include that: (1) states are major regulators, payers, and providers of care, and (2) lack of reform is an impediment to politics at the state level because “’public perceptions are structurally biased to underestimate health care costs,’” thereby “alienating citizens from elected officials.” One may question whether the states in practice are able to function as policy laboratories in the face of modern interest group influence on state and congressional political processes. However, practical realities do little to detract from the theoretical strength of the federalism argument in analyzing why the pharmacy benefits programs are desirable as a health policy matter removed from their political context.
Viewing the Vermont and Maine drug demonstration programs through the federalism lens leads to the position that the state policymakers who created the programs are well-, if not best, situated to assess the pharmacy needs of their residents (such as affordable drugs and drug coverage) and translate those needs into policy proposals. The pharmacy programs thus should be implemented because they are the most likely to succeed in lowering drug costs to an applicant state’s residents. More importantly, Vermont and Maine policymakers should have the opportunity to test their programs through implementation to benefit policymakers in other states who are looking to reduce pharmaceutical costs to residents. Value exists in a program that “fails” as well as with a program that “succeeds” in that future policymakers, both in- and out-of-state, have a data point to use in forming new policy. Studies on pharmaceutical price controls and government provision of goods otherwise available on the market often rely on empirical data from other countries’ national level price controls in assessing the desirability of such policies. In an area of such high theoretical contention, additional data from the Vermont and Maine programs stand to save valuable time and effort dedicated to abstract argument about the advisability of individual state level price policies in the United State.
A counter to the pro-drug demonstration program position arises within the federalism framework as well. In analyzing the PhRMA cases, the question arises as to whether a state’s use of the waiver process to enhance its access to an already-existent federal program is in keeping with the theoretical bases for cooperative federalism. Cooperative federalism regulatory regimes like that in Medicaid have become increasingly common since Congress enacted the first such regulations in the 1960s and 1970s. Under cooperative federalism, the federal level sets uniform standards through statutes and/or agency regulations; states then have the discretion to implement, supplement and sometimes receive exemption from the federal requirements. The basic rationale for cooperative federalism is that in certain areas of policy, “the benefits of allowing for diversity in a federal regulatory program[ ] outweigh[s] the benefits of demanding uniformity in all situations.” In other words, states may be able to positively tailor the federal regime to local conditions. Here Congress drafted the waiver provision to enable states to conduct “short-term experimental projects with innovative policies.” Under Vermont’s and Maine’s programs, the states are not necessarily seeking to implement, supplement, or seek exemption from federal requirements in the name of program innovation, diversity, and local tailoring. Rather, the states may be viewed as attempting to mimic a federal program by simply extending its applicability to a greater class of beneficiaries. The fact that the two states facing challenges on their programs essentially adopted the same program within a short time of each other also defies the logic of cooperative federalism. Given the short time frame, there is only a weak argument that the first state had enough experience with its program to deem it a success and thus for it the serve as a model pharmacy policy for the second state. Nor may these states argue, consistent with the rationale for cooperative federalism, that they enacted their programs based on the success of the federal drug rebate program. The success of the federal program does not necessarily speak to its success as applied and expanded by individual states; furthermore, if there were such an inference to be drawn, it is arguably the province of Congress to so conclude.
A strong market argument exists against state-compelled drug rebates: such rebates distort the competitive market and keep drug prices at an artificially high level for either the whole market or certain segments. As stated in the section on process, drug manufacturer pricing for any given purchaser is highly dependent on the level of discount off of AWP that the manufacturer must give another purchaser, as well as the number of enrollees that the other purchaser represents. If a state may compel rebates for a large portion of its population, a manufacturer is likely to respond by raising prices to other groups of consumers. Competitive market proponents argue against state price controls such as the drug rebate programs on the basis that private interests like managed care organizations already negotiate for volume-based discounts from manufacturers. These private sector transactions produce accurate and efficient pricing of pharmaceuticals, as the best discounts go to those purchasers who can move the most drugs to consumers in an administratively- and medically-cost effective manner. State price controls dampen negotiations, in that manufacturers will be less likely to offer deep discounts reflecting a private insurer’s true efficiencies if they must offer the same discounts to a state’s expanded Medicaid population. Market advocates thus hold that policymakers can make drugs more affordable by encouraging private individuals to enroll in competing private health plans to better match pharmaceutical delivery efficiency with price.
Strictly banning the Vermont and Maine programs in the name of the market, however, misses an important way in which government programs may play a market role. Health plan competition may not be robust in certain areas and/or over discrete periods of time, thereby leaving room for government intervention in the short term. Limited government programs may serve an important function in signaling a market failure. The waiver program, with its supposed commitment to limited-term, experimental projects, seems an appropriate vehicle for such involvement. One may argue that, in keeping with the above argument on data collection, the waiver program really serves as a steppingstone to long-term Medicaid policy. Thus, use of the program to implement short-term market corrective measures is improper. This argument is myopic, however, as it takes as a baseline that long-term price controls will not succeed in bringing affordable drugs to state residents. If the short-term experiments show positive results, there is less reason to argue against the implementation of price controls on a broader level. Another critique is that states are trying to effect large-scale reform through the waiver process. While the argument has force, it also makes a crucial assumption: that the Secretary will in fact approve arguably inappropriate waiver requests aimed at broad reform. The argument loses force if one views the Secretary as complying with his or her statutory authority – compliance that can be supported by proper judicial review.
Vermont and Maine may make an appeal for an exception to the market rule based on the lack of managed care competition in their respective states. Generally speaking, population density is a key factor to the success of managed care in a given insurance market: a large population provides a sufficient base over which managed care companies can spread health care costs, as well as an ample number of providers with which to construct a comprehensive network. States with significant rural areas, such as Vermont and Maine, thus are less likely to be adequately served by managed care insurers than are states with sizable urban and suburban populations. Lack of managed care competition may be due to lack of entry or substantial exit. A recent report on withdrawal of HMOs from rural areas of California cites that the reasons for managed care exit from rural markets include “a residential population that is relatively expensive to insure, the inherent difficulty of distributing the risks and costs of health coverage to a smaller population base, shortages of health care providers, expensive medical practices that increase costs, and concerns over reimbursement rates for care paid by health-care purchasers.” While industry reports vary, most accounts show the HMO market penetration of Vermont and Maine at levels significantly below their New England counterparts. Vermont in particular shows a penetration rate of more than forty percentage points below top-ranking Massachusetts. Thus, while state pharmaceutical programs employing price controls may not be a prudent long-term solution to drug access for every state, they may provide important interim coverage in areas where private insurers have failed or have yet to venture.
Market proponents may still argue against state price control programs on the basis that such programs inhibit private insurer entry of companies who are willing to take a stab at small rural markets. Until those companies make their intentions known, a hard-line market view does little to address the present needs of working class rural families. In the meantime, state pharmacy discount programs may act as an important stop-gap measure that can be removed once earnest entrants come on the scene. The possibility also exists that state discount programs will actually induce private insurance entry. The Vermont pharmacy benefits program would have exempted persons who qualify for private coverage from the state plan. Faced with a state discount program, pharmaceutical manufacturers may have the incentive to make deals with managed care companies to reexamine the economics of rural markets in exchange for better drug discount rates for the companies’ full enrollee rosters. New managed care entrants may decide that they are capable of providing a better all around package of benefits than that of the state, thereby giving persons incentives to shift out of the limited state program and onto more comprehensive private coverage. In sum, a temporary extended state pharmacy benefit program may encourage important bottom-line negotiations between private companies who traditionally have overlooked, avoided, or decided against entering less desirable markets.
The Vermont and Maine programs also raise a broader health policy question of the appropriateness of the current focus on prescription drug coverage apart from access to health care and insurance generally. Without the political will to enact broad, systematic health care reform measures, legislators have taken a piecemeal approach to health policy that at times focuses on specific types of health care needs or treatments. Prescription drug coverage is a current favorite of both state and federal legislators. A focus on prescription drug coverage, however, runs the risk of undesirably separating drugs from the context of comprehensive, continuous health care. Furthermore, emphasis on drugs may be misplaced from treatment and cost effectiveness perspectives. Policymakers thus would be well-served to question the prudence of a focus on prescription drug coverage.
Following the failure of President Clinton’s 1993 policy campaign for comprehensive national health care reform, health care reform at the federal level has become increasingly fragmentary and piecemeal. Recent efforts to enact a Medicare prescription drug benefit evidence the primacy now granted pharmaceutical issues by both Congress and the White House. A political focus on drug coverage seems appropriate in the current climate of rising health care expenditures driven to a large degree by pharmaceutical costs. Waiver expansion programs adopted to address the lack of adequate pharmaceutical insurance in essence create a pool of limited purpose beneficiaries who receive drug coverage but only partial coverage of other health care goods and services. This limited focus at the federal and state levels extracts drugs from, and thus risks elevating them above, the full context of comprehensive health insurance and continuous care. Indeed, CMS recognized this risk by recommending in its guidance on Pharmacy Plus waivers that a state include provision for basic health care coverage to accompany pharmaceutical benefits; CMS further enforced its recommendation by including basic primary care in a list of check boxes on the agency’s Pharmacy Plus waiver proposal form.
A shift in the direction of drug primacy without an attendant commitment to coverage of low cost, behavioral preventive care may saddle an already strapped health care system with uncontrollable costs. In recent years, drugs have accounted for a significant percentage of increases in Medicaid costs , as well as health care spending on the whole. Several studies have documented the ability of proper pharmaceutical use to reduce costly hospital stays. However, as one article notes, “there is little evidence that the huge increase in spending [on pharmaceuticals] is dramatically improving the health of Americans.” Cost effectiveness studies of pharmaceuticals, like many outcomes-related measures, are still in their infancy and significant discussion exists on proper methods of measuring cost effectiveness. Furthermore, reports on pharmaceutical contribution to reduced hospitalization rates alone fail to measure the cost effectiveness of pharmaceuticals relative to other types of health care services. Low-cost preventive measures other than drug treatments can be “more cost-effective than many curative services and be a good use of health care funds.” By focusing on certain drugs or classes of drugs, they may also fail to capture the range of cost-effectiveness among pharmaceuticals. One recent and high profile study of pharmaceuticals stresses that, due to the enormous variation in the quality of pharmaceutical innovation, “efforts to understand both the benefits and costs of new drugs need to be supported by an account of pharmaceuticals that distinguishes among levels of innovation.” The report goes on to note that the large scale reached by certain brand manufacturers has brought about a “line extension” strategy – under which companies create “new” products by using repackaged old active ingredients – to attract new patients, sustain or raise the prices of popular drugs, and reap high returns on investment. Through line extensions, large drug manufacturers can generate the billions of dollars in extra revenue that they need to meet Wall Street growth targets. This source of revenue can make up for lack of revenue generated by new active ingredients in a market that averages one such product launched per company per year.
To list critiques of pharmaceutical market strategy and clinical effectiveness is not to condemn the use of drugs or all aspects of their pricing. The pharmaceutical industry faces exceptionally high research and development costs which justify a substantial degree of the high retail price of drugs. Properly prescribed and monitored drugs are an important and often crucial component of comprehensive medical care. Unreasonably high drug prices and lack of pharmaceutical insurance are real problems for many individuals and families who need these important products to maintain and/or improve their health. Noting concerns in pharmaceutical markets and areas of medical uncertainty regarding drug treatments is only to caution against an overly optimistic view of a rosy health care future where all ills can be easily and economically cured through a pill. Policymakers should base their decisions on reliable, comparative studies of treatment cost-effectiveness. In formulating comprehensive health policy, decisionmakers should avoid separating and privileging drug coverage in ways that may incentivize beneficiaries to forego other types of low-cost coverage and care in favor of drugs where the relative benefits of such drugs may be in question. Nor should policies lead patients to risk adverse drug reactions by taking multiple drugs without sufficient physician oversight covered by comprehensive insurance. Instead, policymakers should consider drug coverage as an integral part – but only a part – of access to comprehensive health care geared towards low-cost prevention.
The Section 1115 waiver option has been described as “[p]erhaps the most powerful health policy tool that the executive branch possesses” in terms of its capacity to restructure the Medicaid program. A part of Medicaid since the program’s inception in 1965, the Section 1115 waiver program has encountered two primary critiques. From a cost perspective, waiver critics allege that demonstration waivers have contributed significantly to increases in federal Medicaid costs. Legal and political critics in turn focus on the program’s allocation of power between the federal government and states. Believers in the “New Federalism,” who seek to decrease the role of the federal government in policy matters and concomitantly increase the power of the states, challenge that the cooperative federalist scheme of Medicaid gives too much oversight to a federal agency. These commentators seek either to (1) give states greater ability to implement waiver programs by reducing the review function of DHHS/CMS, or (2) revamp Medicaid to remove federal waiver review altogether. However, movements towards state “flexibility” are at odds with Congress’ intent in enacting the waiver provision and potentially violate the separation of powers.
The Medicaid program’s funding structure, whereby states may receive federal matching funds for qualifying state Medicaid expenditures, has led to a strategy among states known as “Medicaid maximization.” Under this approach to state cost containment, a state may shift the cost of traditionally state-funded programs to the federal government. States achieve this outcome by reconfiguring the state programs such that they are compatible with Medicaid standards for federal financial participation. Through the waiver system, states are even more capable of accomplishing such a shift. Rather than having to fit their programs into the federal mandatory provisions to qualify for federal funding and potentially sacrificing important components of the programs, states may seek exemption from the federal requirements altogether. Thus, an expanded Medicaid pharmacy benefits can replace already existing state pharmacy programs that provide drugs at a discounted price to residents with higher incomes than traditional Medicaid beneficiaries. State fiscal stress increases the incentives to shift state costs to the federal level. With the states experiencing severe budget constraints in the early years of the 21st century, it is not surprising that states have increasingly sought out Medicaid waivers to fund beneficiary expansions. The states may be enacting Medicaid coverage expansions out of altruism, but such use of waivers to shift costs is arguably in violation of the federal government’s budget-setting powers. In the words of Senator Baucus, sponsor of a 2002 bill to formalize the waiver review process, “[w]here Congress has set limits on the use of federal dollars, waivers should not be used as a back door way to get around those limits.”
Congress enacted the Medicaid program in 1965 to ensure care for persons “whose incomes are insufficient to meet the costs of necessary medical services.” While states have the discretion to go beyond the federal eligibility requirements and expand the class of needy residents served by Medicaid , excessive expansion through the waiver program faces a critique from the position of fairness. Although Congress has yet to articulate an upper bound on permissible expansion of the Medicaid beneficiary class, it is conceivable that Congress had some such boundary in mind when it first enacted the legislation; otherwise, Congress would have dispensed with or written weaker language describing eligibility. The wording of the Medicaid statute calls to mind a level of economic deprivation below some ability to choose among non-government methods of paying for health care (albeit accompanied by economic hardship). To the extent one believes that Congress intended for Medicaid to serve the poorest and most health-compromised citizens, expansion of the program to those who can afford health care if they make responsible consumer choices (such as those with incomes from 200-300% of FPL) may seem unfair from a moral perspective.
Fairness is not only a matter of how one feels about who merits Medicaid benefits from a financial need perspective; violations of our notions of fairness can cause real distributional impacts due to limited government financial resources. As noted by analysts of Oregon’s waiver program, expansion of any particular state Medicaid program in terms of both services and income eligibility risks taking away some degree of care from the state’s most needy residents who are already enrolled in Medicaid. Perceptions of fairness also impact political realities. In the current federal political climate of reducing both federal and state welfare rolls , expansion of an already costly welfare program at the states’ initiative rather than the federal government’s puts Medicaid in the line of fire for reform at the federal level. Politicians who fear an exponential growth of federal outlays with regards to the new, questionably needy beneficiaries may react by seeking to restructure the federal program in a way that ultimately harms those at the more extreme end of need. They may do so by redefining federal eligibility to include a relatively low cap on allowable household income and/or changing the finance structure of Medicaid altogether. Such reform proposals already are on the table, as seen in the Bush administration’s 2003 Medicaid reform proposal (see below). In sum, the possible negative impacts of Medicaid expansion through waivers on the original state and federal Medicaid beneficiary classes suggests that debates about such expansion should include a full range of in-state and out-of-state voices – not just those of the applicant state and CMS.
As discussed above, there is a significant argument that the Secretary is exceeding his or her substantive authority to approve experimental programs under the waiver provision. The Secretary thus may be going beyond the specific powers delegated to his or her office by Congress. This action by the executive branch is a potential breach of the separation of powers between the legislature and the executive branches Agency action beyond authority delegated by Congress cannot be sanctioned as a proper mode of for determining health policy under the Chevron doctrine given the doctrine’s focus on maintaining the separation of powers.
The Secretary is not the only executive branch entity which is potentially violating the separation of powers. Apart from making individual waiver approvals of questionable scope, the Secretary also acts as an arm of the White House by implementing waiver initiatives aimed at increasing states’ flexibility under the waiver program. The most recent waiver initiative permits broad changes in a state’s Medicaid program to be accomplished through the waiver provision. In addition, the initiative runs counter to the experimental and innovation thrusts of the waiver provision by providing templates for state proposals and applications. The initiative formalizes a move away from the 1994 guidance by providing waiver review with expedited process to states that submit proposals under the initiative, thereby running further afoot from the above Mead analysis.
Subsequent to the Vermont and Maine decisions, the Bush administration in August of 2001 announced a new waiver initiative, known as the Health Insurance Flexibility and Accountability Demonstration Initiative. No clearly delineated federal guidance on Medicaid waivers existed before this time. The new initiative put forth several goals for the waiver program, including the promotion of coverage expansions given existing Medicaid resources; encouraging premium assistance programs; allowance of broad state flexibility in setting Medicaid rules; and improvement in accountability regarding the impact of waiver initiatives on reducing the number of low-income uninsured individuals. DHHS concomitantly issued guidance on the types of Medicaid waivers that would be approved under HIFA; state proposals that follow the guidelines will be given expedited review, described by CMS as “efficient and priority review.” The guidance divides Medicaid beneficiaries into three categories – mandatory, optional, and expansion – and grants states respectively increasing flexibility to design rules for each group. Relevant to the present discussion, the expansion group includes those persons who states cannot cover without a Medicaid waiver, such as childless adults under age 65. HIFA generally encourages states to use waivers to expand coverage, especially for persons with incomes below 200 percent of the FPL. States enjoy broad flexibility to design rules governing expansion group coverage. In terms of the benefits granted to expansion beneficiaries and cost sharing that applies to this group, DHHS will grant states “virtually unlimited” flexibility. At a minimum, states must provide expansion beneficiaries with “basic primary care,” vaguely defined as including physician services but otherwise open for interpretation. HIFA guidance does not set caps on premiums, co-payments, and other cost sharing that a state may impose on expansion groups. Under current federal law, these amounts are still governed by the general Medicaid provision that beneficiaries cannot be required to make a copayment of, or participate in another form of cost-sharing for, more than a “nominal” amount. Other waiver proposals will continue to be considered outside of HIFA.
The original waiver process employs an in-depth review process that involves various federal agencies within DHHS and OMB. HIFA is meant to respond to criticism from states and interest groups that this review process in practice has been too slow and has required too many rounds of negotiations. Such critiques have continued to be made even after a similar initiative in the Clinton administration focused on speeding waiver considerations at the federal level. Many of the critiques, as well as the approaches behind the new HIFA initiative, have come from the National Governors Association. Like waiver continuations, HIFA waiver proposals arguably are to receive expedited process under “efficient and priority review.” As in the 1994 notice, state applicants must certify that they have used a public process to allow for public comment on the proposed HIFA program. The current guidance leaves open whether parties will have an opportunity to comment at the federal level. This omission is in contrast to the 1994 notice, which provided that notice publication in the Federal Register would “indicate that the Department accepts written comments regarding all demonstration project proposals.” In practice, CMS has not provided notice of the submission of HIFA applications, other 1115 waivers, or waiver amendments, nor has CMS established a federal level public comment period. According to one study, it is unclear whether the HIFA review process will include the same question-and-response exchange between state and federal agency officials as does the general waiver program. The concerns with public process, thoroughness and quality discussed in the prior section on Mead thus are potentially augmented under HIFA.
HIFA’s focus on expedited review takes a different tone than that during the Clinton administration. Rather than viewing cumbersome federal agency review as needing streamlining as did the previous administration , the Bush administration seems to be taking the position that federal agency review is not necessary for implementation of successful Medicaid programs. This position is evidenced by the current administration’s proposal, less than two years after HIFA, calling for the conversion of Medicaid into a quasi-block grant program. Most importantly for the present analysis, the Bush proposal would eliminate the waiver program. Thus it appears that HIFA may be an attempt to do through executive prerogative what the President couldn’t convince Congress to enact in 2001. In the alternative, HIFA may be viewed as a stepping stone to congressional amendment, in that the initiative’s two year history gives Congress an example of what can be accomplished by abandoning much of the waiver process. President Reagan made a similar proposal in 1981 calling for a cap on federal Medicaid expenditures, which Congress ultimately rejected.
DHHS describes the 2003 Bush plan as “giv[ing] states the upfront investment and flexibility to design health care programs that best meet the needs of their citizens and expand coverage to more people, including the mentally ill, chronically ill, those with HIV/AIDS and those with substance abuse problems.” However, current opponents of the Bush plan view the shift in power to the states as putting optional beneficiaries at risk for losing benefits, having to pay additional costs, and/or facing waiting lists. According to policy analysts, the proposal uses a “bait and switch” approach by providing a participating state with an upfront injection of funds to help it meet Medicaid costs during the current state fiscal crisis in return for the state’s commitment to convert Medicaid and State Children’s Health Insurance Program (SCHIP) into a capped, consolidated block grant. While the Bush plan does not directly affect the core mandatory beneficiary class under Medicaid, the proposal does signal unwillingness at the federal level to support ever-increasing state expansion of Medicaid eligibility. The proposal, in essence, puts in place a federal cap on the definition of “needy.” The Bush plan may be in part a negative reaction to the costly waiver program and not just a move towards states rights under the New Federalism, as is suggested by the program’s strict future caps on federal spending. If so, the proposal fails to address why a complete shift in power to the states above a certain threshold is necessary, rather than a revamping of the waiver process. Health-related welfare policies at the state level implicate national concerns, while notions of good health care management indicate the continued desirability of some degree of federal oversight relating to health standards in addition to expenditures. Therefore, reform of the Medicaid program should retain the dual policymaking responsibilities under the current cooperative federalist structure and reinforce important procedural controls on waiver approval to ensure broad participation at both the state and federal levels.
Courts may reign in agency actions through several means. As seen in the PhRMA cases, those who do so by holding that Congress has clearly spoken on a particular issue run the usual challenges of statutory interpretation and familiar risks of judicial legislating under Chevron Step One. Ultimately, once courts have sufficiently delineated Congress’ legislative intent, they will have to decide whether to limit agency discretion at the deference stage. The Supreme Court’s current jurisprudence on deference as articulated in Mead leaves a substantial role for the courts in characterizing and scrutinizing particular forms of agency actions.
Whether or not the Mead doctrine holds up over time, courts in the near future must consider the deference question within an existing analytical framework that emphasizes the roles of Congress in crafting statutory directives and of procedural protections. These two factors operate to ensure that agencies act within their properly delegated authority and make sound, quality decisions pursuant to that authority. Thus, courts should not grant heightened deference where an action which, for all intensive purposes, resembles an agency action normally requiring more formal process occurs with minimal public participation to the exclusion of interested parties at either level of a cooperative federalist regime. An appeal for deference based solely on agency expertise does not warrant passive approval where the very proposal in question, like a demonstration waiver, is highly experimental in nature. Nor should courts grant deference to agency decisions which blatantly sidestep the legislative process by going beyond any reasonable interpretation of the scope of the agency’s authority. Courts should instead cabin agency discretion where the actual form of an action is inconsistent with the amount of process granted and where outside voices may valuably supplement agency expertise. In addition, courts should hold agencies to their delegated authority through close scrutiny of statutory language and congressional intent.
The entire burden of restraining agencies need not, and should not, be borne by the judiciary. Congress can take a lead role in responding to critics of the waiver process by amending the waiver provision itself or the Medicaid statute more broadly. Pharmacy coverage is a vital area of health policy with significant implications for national as well as state interests. When such an important issue is under consideration at the agency level, a full range of participants should have their say. Congress can secure public participation by strengthening the state and federal procedural requirements under the current waiver regime. Furthermore, Congress should legislate in this contentious area, as refraining to do so in light of Executive Branch initiatives creates enormous inconsistencies among statutory text, purpose and enforcement that detract from a regulatory regime’s legitimacy. Congressional inaction also effectively shifts control of agencies to the Executive Branch in a manner contrary to Chevron’s lesson of congressional legislative centrality. While the battle over Medicaid reform may be a difficult one, little value comes from permitting Congress to avoid its legislative duties by looking the other way while agencies exceed their statutory authority. Only with proper congressional delineation and oversight of agency authority coupled with tailored judicial review will agencies produce quality actions in keeping with their mandates.
 States may seek several different forms of waivers under the Medicaid program. These waivers include Section 1115 demonstration waivers, Section 1915(b) “Freedom of Choice” waivers, and Section 1915(c) “Home and Community-Based Service” waivers. Of the three types, Section 1115 waivers have the greatest potential to change core components of the Medicaid program. For a brief description of the various waiver programs, as well as critiques of current proposals to reform Medicaid, see JOHNS HOPKINS AIDS SERVICE, MEDICAID OVERVIEW: REFORM & MANAGED CARE, available at www.hopkins-aids.edu/manage/medicaid_2.html (last visited Apr. 12, 2003).
 See John Bentivoglio et al, State Controls on Drug Costs: An Out-of-Control Experiment in Federalism ? 10 BNA’S HEALTH LAW REPORTER 41, 1609 (2001).
 See Judith M. Rosenberg and David T. Zaring, Managing Medicaid Waivers: Section 1115 and State Health Care Reform , 32 HARV. J. ON LEGIS . 545 (1995).
 Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp.2d 1 (2001), rev’d , Pharmaceutical Research and Manufacturers of America v. Thompson 251 F.2d 219 (2001); Pharmaceutical Research and Manufacturers of America v. Thompson, 191 F.Supp.2d 48 (2002), rev’d , Pharmaceutical Research and Manufacturers of America v. Thompson, 2002 WL 31863531 (D.C. Cir. 2002).
 5 U.S.C. §§ 551 to 559, 701 to 706, 1305, 3105, 3344, 4301, 5335, 5372, and 7521,
 Chevron v. Natural Resources Defense Council, 467 U.S. 837 (1984).
 United States v. Mead, 533 U.S. 218 (2001).
 Id. at 235 (citing Skidmore v. Swift & Co., 323 U.S. 134, 140 (1940)).
 Id. at 226-27.
 Mead itself speaks in terms of congressional authorization. However, as pointed out by Justice Scalia in his dissent, authorization may be prescriptive or discretionary, a difference which opens the door for incoherencies in outcome under the majority’s standard. See United States v. Mead, 533 U.S. 218, 243 (Scalia, J., dissenting: “[a]gencies [who are authorized but not required to undertake informal rulemaking] are free to give guidance through rulemaking, but they may proceed to administer their statute case-by-case, ‘making law’ as they implement their program... Is it likely – or indeed even plausible – that Congress meant, when such an agency chooses rulemaking, to accord the administrators of that agency, and their successors, the flexibility of interpreting the ambiguous statute now one way, and later another; but, when such an agency chooses case-by-case administration, to eliminate all future agency discretion by having that same ambiguity resolved authoritatively (and forever) by the courts? Surely that makes no sense.”) Rather than discard Mead, which is the current state of the law by a large majority, I leave aside the implications of Scalia’s dissent for the present purposes.
 These factors are an approximation of those laid out by Thomas Merrill, who identifies three considerations given weight in Mead: “(1) whether Congress has prescribed relatively formal procedures; (2) whether Congress has authorized the agency to adopt rules or precedents that generalize to more than a single case; and (3) whether Congress has authorized the agency to prescribe legal norms that apply uniformly throughout its jurisdiction.” According to Merrill, “the number and correct characterization of the factors invoked in the majority opinion is open to debate.” Thomas W. Merrill, The Mead Doctrine: Rules and Standards, Meta-Rules and Meta-Standards , 54 ADMIN. L. REV . 807, 814 (2002) [hereinafter Merrill Meta-Rules ].
 This position has been stated by at least one circuit court following Mead in the context of Medicaid home-and-community based waivers under 42 U.S.C. § 1396n(c), see Bryson v. Shumway, 308 F.3d 79 (1st Cir. 2002) (deferring to the expertise of the Secretary of HHS under Mead and Skidmore, even though “[b]ecause the approval process did not utilize formal procedures, it may not be entitled to Chevron deference, but there remains the deference owed agencies due to their ‘specialized experience.’”)
 United States v. Mead, 533 U.S. 218, 230 (2001).
 CINDY MANN, KAISER COMMISSION ON MEDICAID AND THE UNINSURED, SECTION 1115 WAIVERS IN MEDICAID AND THE STATE CHILDREN’S HEALTH INSURANCE PROGRAM: AN OVERVIEW (2001).
 See id.
 42 U.S.C. § 1315(a) (“In the case of any experimental, pilot, or demonstration project which, in the judgment of the Secretary, is likely to assist in promoting the objectives of subchapter I, X, XIV, XVI, or XIX of this chapter, or part A or D of subchapter IV of this chapter, in a State or States-- (1) the Secretary may waive compliance with any of the requirements of section 302, 602, 654, 1202, 1352, 1382, or 1396a of this title, as the case may be, to the extent and for the period he finds necessary to enable such State or States to carry out such project...”).
 Arbitrary and capricious review is the baseline inquiry into agency decisions required by §706 of the Administrative Procedures Act. For applications of arbitrary and capricious review to Section 1115 waiver decisions, see C.K. v. New Jersey Department of Health and Human Services, 92 F.3d 171 (3rd Cir. 1996); Beno v. Shalala, 30 F.3d 1057 (9th Cir. 1994); Crane v. Mathews, 417 F. Supp. 532, 539 (N.D. Ga. 1976) ; Aguayo v. Richardson, 473 F.2d 1090, 1103 (2d Cir. 1973) . While the majority of cases applying the arbitrary and capricious standard to Section 1115 waivers deal with the Aid to Families with Dependent Children (AFDC) welfare program, the same standard applies to Section 1115 waivers in the context of Medicaid as a single provision allows for waivers of both substantive statutes, 42 U.S.C. § 1315(a).
 See MANN , supra note 14.
 See UNITED STATES GENERAL ACCOUNTING OFFICE, REPORT TO THE COMMITTEE ON FINANCE, U.S. SENATE, MEDICAID AND SCHIP: RECENT APPROVALS OF DEMONSTRATION WAIVER PROJECTS RAISE CONCERNS, GAO-02-817 (July 2002) (“GAO Waiver Concerns”)
 See PhRMA v. Thompson, 251 F.3d 219, 224 (2001) (citing Christensen v. Harris County, 529 U.S. 576 (2000) in stating that “not all agency interpretations of statutes warrant Chevron deference... ‘Interpretations such as those in opinion letters – like interpretations contained in policy statements, agency manuals, and enforcement guidelines, all of which lack the force of law – do not warrant Chevron-style deference.’”)
 See Rosenberg, supra note 3.
 Such as those available on CMS’s website, www.hhs.gov/medicaid/waivers , especially “Guidelines for States Interested in Applying for a HIFA Demonstration.” For more information on the HIFA demonstration waiver initiative, see “(Im)proper Executive Initiative? Greater State Flexibility,” below).
 42 U.S.C. § 1396.
 42 U.S.C. § 1396o, “Use of enrollment fees, premiums, deductions, cost sharing, and similar charges.”
 42 U.S.C. § 1396a, “State plans for medical assistance.”
 VERNON SMITH, ET AL, KAISER COMMISSION ON MEDICAID AND THE UNINSURED, MEDICAID SPENDING GROWTH: RESULTS FROM A 2002 SURVEY (2002), available at www.kff.org/content/2002/4064/4064.pdf.
 42 U.S.C. § 1396a.
 42 U.S.C. § 1396a.
 42 U.S.C. § 1396b(a)(1).
 SMITH , supra note 26.
 Id .
 See Leighton Ku and Matthew Broaddus, “Why Are States’ Medicaid Expenditures Rising?” Center for Budget and Policy Priorities (Jan 2003), available at www.cbpp.org/1-13-03health.htm www.cbpp.org/1-13-03health.htm. See also Smith, supra note 26.
 See SMITH , supra note 26.
 KAISER COMMISSION ON MEDICAID AND THE UNINSURED, MEDICAID AND THE PRESCRIPTION DRUG BENEFIT: COST CONTAINMENT STRATEGIES AND STATE EXPERIENCES (2002) [hereinafter MEDICAID PRESCRIPTION DRUG BENEFIT ], available at www.kff.org/content/2002/20020213/4063.pdf.
 SMITH , supra note 26, at 16.
 Id . at 10.
 42 U.S.C. § 1396r-8(b), Grants to States for Medical Assistance Programs, Payment for covered outpatient drugs.
 42 U.S.C. § 1396r-8(b)(1)(A).
 See Omnibus Reconciliation Act, Pub. L. No. 103-66, § 13602, Additional Federal Savings Through Modifications to Drug Rebate Program (1993).
 42 U.S.C. § 1315.
 See Rosenberg, supra note 3.
 S. REP. NO . 1589, 87th Cong., 2d Sess. 19 (1962) (as cited in Rosenberg, supra note 3).
 S. REP. NO . 1589, 87th Cong., 2d Sess. 19, at 20 (1962) (as quoted in Rosenberg, supra note 3).
 42 U.S.C. §1315(e).
 Rosenberg, supra note 3, at 551.
 42 U.S.C. § 1315(a)(1).
 42 U.S.C. § 1315(a)(1), while allowing for waiver of § 1396a, does not include as waivable § 1396r-8 (as cited by Pharmaceutical Research and Manufacturers of America v. Thompson, 251 F.3d 219, 222 (2001))
 Mann, supra note 14.
 Id .
 42 U.S.C. § 1315(a)(2)(A).
 MANN , supra note 14.
 58 Fed. Reg. 49249, 49250 (1994).
 42 U.S.C. § 1315(a).
 Rosenberg, supra note 3, at 551 (citing Allen Dobson et al., The Role of Federal Waivers in the Health Policy Process , HEALTH AFF . 72 (1992)).
 See Medtronic, Inc. v. Lohr, 518 U.S. 470, 485 (1996).
 Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp.2d 1, 5 (D.D.C. 2001) (“PhRMA I trial”),
 Id . at 5.
 Id . at 6.
 Id at 14.
 Id at 3.
 Id .
 Id . at 13.
 Id . at 13
 Pharmaceutical Research and Manufacturers of America v. Thompson, 191 F.Supp.2d 48, 52 (D.D.C. 2002) (“PhRMA II trial”)
 Id . at 53.
 Id . at 56.
 467 U.S. 837 (1984).
 As academics have termed the question of Chevron’s scope, see, e.g., Thomas W. Merrill and Kristin E. Hickman, Chevron’s Domain , 89 GEO. L.J. 833 (2001) [hereinafter Merrill Chevron’s Domain ].
 Three out of the four opinions implicate deference to administrative action directly. The fourth, the Maine appeal, avoided the question of deference by deciding the case on a procedural ground, i.e., that the Secretary had not in fact considered and approved the program that was before the court. See Pharmaceutical Research and Manufacturers of America v. Thompson, 313 F.3d 600, 604 (D.C. Cir. 2002).
 Pharmaceutical Research and Manufacturers of America v. Thompson, 251 F.3d 219,224 (D.C. Cir. 2001) (“PhRMA I appellate”).
 See Chevron, 467 U.S. at 843 (noting that “[t]he judiciary is the final authority on issues of statutory construction and must reject administrative constructions which are contrary to clear congressional intent.”); see also PETER STRAUSS, ADMINISTRATIVE JUSTICE IN THE UNITED STATES 370 (2002) (stating that Chevron’s first step “requires the reviewing court independently to determine the extent to which the statute settles the question of meaning with respect to the issue before it.”)
 PhRMA I appellate, 251 F.3d at 224 (2001) (citing Chevron at 843).
 Merrill Meta-Rules , supra note 11, at 810 (quoting Cass Sunstein, Law and Administration After Chevron, 90 Colum. L. Rev. 2071, 2074-75 (1990) ).
 See Merrill Chevron’s Domain , supra note 71, at n.16.
 See, e.g ., the disparate conclusions of the PhRMA I trial court, 135 F.Supp.2d at 13 (2001) (noting the “[a]bsence of guidance from Congress”) and the PhRMA I appellate court, 251 F.3d at 224 (2001) (finding that Congress had “directly spoken to the precise question at issue’” by reference to the statutory text’s “context – the statute’s purpose and legislative history”).
 WILLIAM ESKRIDGE, DYNAMIC STATUTORY INTERPRETATION , 224 (Harvard University Press 1994). See also Strauss, supra note 75, at 352 (noting that “as a descriptive proposition, it is hard to speak of a singular, precise intended meaning in the work product of several hundred very political individuals – the Congress.”)
 For a discussion the varying degrees of judicial scrutiny under Chevron, including the application of hard look review in Chevron Step Two, see Ronald M. Levin, The Anatomy of Chevron: Step Two Reconsidered, 72 Chi.-Kent L. Rev. 1253 (1997) .
 42 U.S.C. § 1396r-8(b)(1)(A)
 Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp.2d 1 (D.D.C. 2001).
 Peter Strauss has suggested that one may view a court’s failure to apply Chevron analysis as an instance in which the court was able to reach a conclusion on the question presented at Step One. See Strauss, supra note 75, at 370. However, this rationale does not apply to the Vermont trial court opinion, as the court found the “payment” passage ambiguous with regards to Vermont’s proposed waiver program.
 PhRMA I trial, 135 F. Supp.2d at 13
 Id . at 14
 Id .
 Pharmaceutical Research and Manufacturers of America v. Thompson, 251 F.3d 219, 224 (D.C. Cir. 2001) (“PhRMA I appellate”).
 Id .
 See Note, Looking It Up: Dictionaries and Statutory Interpretation , 107 HARV. L. REV. 1437, 1454 (1994), as cited in Strauss, supra note 75, at354 .
 PhRMA I appellate, 251 F.3d at 224
 Id .
 Id . at 225
 Id .
 Id .
 Id .
 Id .
 Id .
 Id .
 Id .
 Eskridge, supra note 81, at 207.
 Id . at 226.
 “Still” here is used to qualify the method’s widespread adoption due to a recent trend towards text and formal tools of textual interpretation. According to Peter Strauss, five factors driving this change have arisen in recent decades. They are: (1) “a general return of conservative values to American politics and, correspondingly, judging,” (2) the dominance of statutes among forms of law at the national level effecting a shift in the courts from common law functions to statutory interpretation, (3) a change in perception of legislatures to emphasize their arbitrariness and irrational/political bases for decisions, (4) changes in the legislative process that have made legislative materials less likely to reflect the views of legislators themselves, and (5) manipulation of legislative histories by savvy staff persons who have taken note of the courts’ reliance on such materials to effect judicial outcomes that were not politically feasible. See Strauss, supra note75.
 Omnibus Reconciliation Act, Pub. L. No. PL 103-66 (1993) (“OBRA 1993”).
 See, e.g., Chapman v. United States, 500 U.S. 453, 464 n. 4 (1991) (as cited in North Broward Hospital District vs. Shalala, 172 F.3d 90, 98 (1999), involving application of the Chevron analysis to an interpretation of the Medicaid statute by the Secretary of DHHS).
 S. 3029, 101st Cong.(1990) , later codified at 42 U.S.C. § 1396r-8 until amended in 1993, see Omnibus Reconciliation Act, Pub. L. No. PL 103-66 (1993).
 Omnibus Reconciliation Act, Pub. L. No. 130-66, § 13602(a)(2)(A)(i)(II) (1993).
 See id., at § 13602 (1993).
 See id , at § 13602(a)(1) (1993). Under the 1993 amendments, states could now restrict coverage for outpatient drugs according to the following language:
(d) LIMITATIONS ON COVERAGE OF DRUGS.--
(1) PERMISSIBLE RESTRICTIONS.--(A) A State may subject to prior authorization any covered outpatient drug. Any such prior authorization program shall comply with the requirements of paragraph (5).
(B) A State may exclude or otherwise restrict coverage of a covered outpatient drug if--
(i) the prescribed use is not for a medically accepted indication (as defined in subsection (k)(6));
(ii) the drug is contained in the list referred to in paragraph (2);
(iii) the drug is subject to such restrictions pursuant to an agreement between a manufacturer and a State authorized by the Secretary under subsection (a)(1) or in effect pursuant to subsection (a)(4); or
(iv) the State has excluded coverage of the drug from its formulary established in accordance with paragraph (4).
(2) LIST OF DRUGS SUBJECT TO RESTRICTION.--The following drugs or classes of drugs, or their medical uses, may be excluded from coverage or otherwise restricted:
(A) Agents when used for anorexia, weight loss, or weight gain.
(B) Agents when used to promote fertility.
(D) Agents when used for the symptomatic relief of cough and colds.
(E) Agents when used to promote smoking cessation.
(F) Prescription vitamins and mineral products, except prenatal vitamins and fluoride preparations.
(G) Nonprescription drugs.
(H) Covered outpatient drugs which the manufacturer seeks to require as a condition of sale that associated tests or monitoring services be purchased exclusively from the manufacturer or its designee.
 Chapman v. United States, 500 U.S. 453, 464 n.4 (1991) (as cited in North Broward Hospital District v. Shalala, 172 F.3d 90 (D.D.C. 1990)); see also United States v. Price, 361 U.S. 313 (1960).
 North Broward Hospital District v. Shalala, 172 F.3d 90 (D.D.C. 1999).
 Id . at 99.
 For discussions of these dangers, fitting under the headings of critiques from formalism, reliability, and historicism, respectively, see Eskridge, supra note 81, at 221.
 Strauss notes the danger of legislative history textual manipulation by staffers. See Strauss, supra note 75, at 353.
 136 Cong. Rec. 24 (daily ed. Sept. 12, 1990) (statement of Sen. Pryor).
 Id .
 See Senator Pryor statement, supra note 119.
 CENTERS FOR MEDICARE AND MEDICAID SERVICES, PHARMACY PLUS SECTION 1115 WAIVER RESEARCH AND DEMONSTRATION PROJECTS TECHNICAL GUIDANCE AND FACT SHEET , available at www.cms.hhs.gov/medicaid/1115/RXFACTSHEET41202.pdf [hereinafter CMS PHARMACY PLUS ]. CMS states that “It should be noted, however, that the purpose of the demonstration is to provide a subsidized pharmacy benefit that is intended to assist individuals in maintaining their healthy status and avoid spending down to Medicaid income and asset eligibility levels. The coverage must be successful in diverting Medicaid eligibility in order for the demonstration to be budget neutral.”
 VT Amendment Cover Letter. From Eileen I. Elliott, Commissioner of the Agency of Human Services, State of Vermont, to Joan Peterson, Project Officer, CMS State Operations (Sept. 5, 2002), available at www.dsw.state.vt.us/districts/ovha/OVHA5.HTM.
 In fact, prescription drug manufacturers have begun offering states disease management programs in treatment areas for which the manufacturers produce drugs. See Melody Petersen, Drug Makers Expand Their Medicaid Role , N.Y. TIMES, April 23, 2003, at C1. Critics of the programs say that the disease management programs are part of a strategy to convince states to drop efforts seeking price reductions.
 See CMS PHARMACY PLUS , supra note 123.
 Several bills calling for explicit allowance of state pharmacy benefit programs such as Vermont’s have made their way into the Congress. See e.g ., S. 3142, 107th Cong. (2002).
 See John Bentivoglio et al, State Controls on Drug Costs: An Out-of-Control Experiment in Federalism ? 10 BNA’S HEALTH LAW REPORTER 41, 1609 (2001).
 See Pharmaceutical Research and Manufacturers of America v. Thompson, 191 F.Supp.2d 48, 64 (D.D.C. 2002).
 See, e.g., Merrill Chevron’s Domain , supra note 71.
 David Barron, Lecture on Administrative Law delivered at Harvard Law School (Fall 2002).
 See, e.g., United States v. Mead, 533 U.S. 218, 230 (noting that more formal process tends to “foster [ ] fairness and deliberation...”)
 See William S. Jordan, Judicial Review of Informal Statutory Interpretations: The Answer is Chevron Step Two, Not Mead . 54 ADMIN. L. REV . 719 (2002) (citing the work of Professor Robert Anthony) [hereinafter Jordan Chevron Two-Step ].
 See id.
 Under 5 U.S.C. §§ 554 and 556/557.
 Under 5 U.S.C. §§ 553 and 556/557.
 Under 5 U.S.C. § 553.
 See Christensen v. Harris County, 529 U.S. 576, 587 (2000).
 United States v. Mead, 533 U.S. 218 (2001).
 Skidmore v. Swift & Co., 323 U.S. 134 (1944).
 See Merrill Meta-Rules , supra note 11, at 810.
 See William S. Jordan III, United States v. Mead: Complicating the Delegation Dance . 31 ENVTL. L. REP . 11425 (2001) [hereinafter Jordon Delegation Dance ].
 United States v. Mead, 533 U.S. 218 (2001).
 Id .
 Id .
 Id. at 226-227.
 Id . at 229.
 Id. (noting that “[w]e have recognized a very good indicator of delegation meriting Chevron treatment in express congressional authorizations to engage in the process of rulemaking or adjudication that produces regulations or rulings for which deference is claimed.”)
 Id . at 230.
 Id .
 Id . at 231.
 Id . at 232-234 (stating that “[t]he [statutory] reference to binding classifications does not, however, bespeak the legislative type of activity that would naturally bind more than the parties to the ruling [after the initial determination of the case at hand]”; “any precedential claim of a classification ruling is counterbalanced by the provision for independent review”; “[i]t is difficult... to see in the agency practice itself any indication that Customs ever set out with a lawmaking pretense...Customs does not engage in notice-and-comment practice when issuing [tariff classifications] and their treatment by the agency makes it clear that a letter’s binding character as a ruling stops short of third parties; Customs has regarded a classification as conclusive only as between itself and the importer to whom it was issued... and even then only until Customs has given advance notice of intended change”; and “[a]ny suggestion that rulings intended to have the force of law are being churned out at a rate of 10,000 a year at an agency’s 46 scattered offices is simply self-refuting.”)
 Id. at 234 (quoting Christensen, 529 U.S. 576, 587).
 Christensen v. Harris County, 529 U.S. 576, 587 (2000).
 Mead, 533 U.S. at 235 (2001).
 Id .
 Id .
 Id .
 Id .
 Justice Scalia argued that the majority’s holding inappropriately returns to a vague balancing test which “collapses” the Chevron doctrine’s emphasis on Congressional intent to have agencies exercise discretion within statutory ambiguities. See id. at 240.
 See Jordan Delegation Dance , supra note 143 (citing United States v. Mead, 533 U.S. 218, 230 n. 12 (2001)).
 See id. ; see also William R. Andersen, Informal Agency Advice: Graphing the Critical Analysis , 54 ADMIN. L. REV . 595 (2002) [hereinafter Informal Agency Advice].
 5 U.S.C. § 551.
 Mead was decided June 18, 2001; PhRMA I June 8, 2001.
 See e.g ., Bryson v. Shumway, 308 F.3d 79 (1st Cir. 2002); see also Fontana v. Caldera, 160 F.Supp.2d 122 (D.D.C. 2001) (citing that the Supreme Court has not spoken on the issue of Chevron deference for informal adjudications and holding that, under Mead’s totality of the circumstances standard for informal agency actions, determinations of the Army Board for Correction of Military Records pursuant to 10 U.S.C. § 1552 are due Chevron deference based on the statutory finality and conclusiveness of such decisions “on all officers of the United States”).
 42 U.S.C. § 1315(a).
 See e.g. , United States v. Florida East Coast Railway Co., 410 U.S. 224 (1973) (interpreting APA § 553(c)’s requirement that “on the record” hearings must be conducted in accordance with the formal procedures outlined in APA §§ 556 and 557); see also Seacoast Anti-Pollution League v. Costle, 572 F.2d 872 (1st Cir 1978) (considering the possibility of informal adjudication through interpretation of APA §§ 554, 556, and 557).
 See Beno v. Shalala, 30 F.3d 1057, 1067 (9th Cir. 1994)
 See e.g., the PhRMA cases.
 C.K. v. New Jersey Department of Health and Human Services, 92 F.3d 171 (3rd Cir. 1996).
 See Pharmaceutical Research and Manufacturers of America v. 251 F.3d 219, 224 (D.C. Cir. 2001).
 The Maine appellate court found that the Secretary’s decision on version of the state’s program including the 2% payment was not properly before the court. The court based its finding on the facts that the state had added the 2% payment after the Secretary’s approval and that the Secretary had issued no subsequent decisions approving the 2% payment as “payment under the State plan.” Having passed on the 2% payment version of the program, the PhRMA II appellate court then reversed the lower court’s decision in favor of the defendant based on the PhRMA I court’s holding. See Pharmaceutical Research and Manufacturers of America v. Thompson, 313 F.3d 600 (D.C. Cir. 2002).
 PhRMA I appellate, 251 F.3d 219, 224 (2001).
 Pharmaceutical Research and Manufacturers of America v. Thompson, 191 F.2d 48, 63 (quoting the PhRMA I appellate decision).
 Id .
 Pharmaceutical Research and Manufacturers of America v. Thompson, 191 F.Supp.2d 48, 65 (2002).
 Section 1315(a) contains no requirement that the Secretary consider any comments in making his or her decision, or a requirement for a statement of basis and purpose, but leaves determinations to the Secretary’s unqualified “judgment.” See 42 U.S.C. 1315(a).
 42 U.S.C. 1302.
 5 U.S.C. § 551.
 See 42 U.S.C. 1315. The APA outlines additional procedural requirements for actions that are required by statute to be “on the record,” see 5 U.S.C. § 553.
 See Informal Agency Advice, supra note 163.
 Jordan Chevron Step-Two, supra note 134;see also United States v. Mead 533 U.S. 218, 240 (Scalia, J. dissenting).
 Jordan Chevron Step-Two , supra note 134, at 727.
 Id .
 See Fontana v. Caldera, 160 F.Supp.2d 122 (D.D.C. 2001).
 See id. at 128.
 Id. at 128-129.
 5 U.S.C. § 551.
 See Medtronic, Inc. v. Lohr, , 518 U.S. 470, 485 (1996).
 New State Ice Co. v. LiebMann, 285 U.S. 262, 311 (1932) (Brandeis, J. dissenting), as cited in Fernando R. Laguarda, Federalism Myth: States as Laboratories of Health Care Reform . 82 GEO. L.J . 159 (1993).
 CMS justified its assignment of public notice and comment procedures to the states on the basis that “states are considered to be a more appropriate forum [than the federal agency] for public input.” See GAO Waiver Concerns, supra note 19, at 25.
 The use of contracted consultants has been addressed in United Steelworkers of America, AFL-CIO-CLC v. Marshall, 647 F.2d 1189, cert denied 453 U.S. 913 (1981) (holding that use of expert consultants is permissible absent specific proof that the statutorily-empowered agency decisionmaker has “failed to confront personally the essential evidence and arguments in setting the final standard.”)
 42 U.S.C. § 1302 (providing that “[t]he Secretary of the Treasury, the Secretary of Labor, and the Secretary of Health and Human Services, respectively, shall make and publish such rules and regulations, not inconsistent with this chapter, as may be necessary to the efficient administration of the functions with which each is charged under this chapter.”)
 Medicaid Program; Demonstration Proposals Pursuant to Section 1115(a) of the Social Security Act; Policies and Procedures. 59 Fed. Reg. 49249 (Sep. 27, 1994).
 See id.
 MANN , supra note 14 (citing DHHS Federal Register notice of Sept. 27, 1994).
 “Legal Effect,” 59 Fed. Reg. 49249 (Sep. 27, 1994).
 The existence of two possible views of the waiver process may come about in part due to an ambiguity in the APA itself: the APA does not differentiate between private parties and other government bodies seeking federal government approval. That the APA is silent on this distinction is not surprising, as Congress drafted the original statute in 1946 long before the rise of cooperative federalist regulatory schemes in the 1970s. Yet courts repeatedly employ the judicial review provisions of the APA and precedent developed through cases involving private parties to review federal agency decisions vis-à-vis states; in fact, Chevron itself involved a cooperative federalist regulatory scheme.
 United States v. Mead, 533 U.S. 218, 233 (2001) (“Any suggestion that rulings intended to have the force of law are being churned out at a rate of 10,000 a year at an agency’s 46 scattered offices is simply self-refuting.”)
 Id . at 223.
 42 U.S.C. § 1315(a).
 42 U.S.C. § 1315(a).
 The state notice requirement is one of the few components of the 1994 notice which CMS has routinely applied since it promulgated the notice. See Mann, supra note 14. However, CMS recently scaled back the amount of information that states must provide on how they specifically complied with the state notice requirement under the 2001 HIFA initiative. Where once states had to provide a narrative description of the process, they need only check a box certifying that they followed a public process which permitted interested stakeholders, including beneficiaries, to comment on a proposed waiver request. See GAO WAIVER CONCERNS , supra note 19, at 25.
 See MANN , supra note 14, at 26.
 See id .
 See e.g ., Fontana v. Caldera, 160 F.Supp.2d 122 (D.D.C. 2001) (basing its holding of Chevron deference on a statutory charge that the agency action be “final and conclusive on all officers of the United States.”)
 See e.g., CENTERS FOR MEDICARE & MEDICAID SERVICES, PHARMACY PLUS SECTION 1115 WAIVER RESEARCH AND DEMONSTRATION PROJECTS TECHNICAL GUIDANCE AND FACT SHEET , available at www.cms.hhs.gov/medicaid/1115/RXFACTSHEET41202.pdf .
 CMS gives submissions under the HIFA guidelines “efficient and priority review,” implying that those which closely follow the guidelines enjoy a presumption of approval. See CENTERS FOR MEDICARE & MEDICAID SERVICES, GUIDELINES FOR STATES INTERESTED IN APPLYING FOR A HIFA DEMONSTRATION , available at www.cms.hhs.gov/hifa/hifagde.asp [hereinafter HIFA GUIDELINES ].
 The PhRMA II appellate court noted that the Maine state health agency represented to CMS that it modeled the Maine program after the Vermont program. Pharmaceutical Research and Manufacturers of America v. Thompson, 313 F.3d 600, 602 (2002).
 GAO WAIVER CONCERNS , supra note 19.
 See Rosenberg, supra note 3, at 551 (quoting the waiver provision’s legislative history).
 See id .
 United States v. Mead, 533 U.S. 218, 232 (2001).
 42 C.F.R. § 430.18.
 5 U.S.C. § 706(2)(C) and (2)(A), as cited in Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp. 2d 1, 9 (D.D.C. 2001).
 United States v. Mead, 533 U.S. 218, 233 (2001).
 GAO WAIVER CONCERNS , supra note 19.
 MANN , supra note 14, at 13.
 SMITH, supra note 26, at 5.
 See Rosenberg, supra note 3.
 See David J. Barron and Elena Kagan, Chevron’s Nondelegation Doctrine , 2001 SUP. CT. REV. 201 (2001).
 This number is a rough estimate derived by multiplying the low end estimate of 10,000 tariff classifications issued per year times 46 issuing bodies, then dividing that number by the total number of waivers so far approved by CMS. In actuality, a more accurate number would divide the first number by the number of waivers submitted each year to better match the actual magnitude of issues before the agency (presumably all tariff requests result in a tariff classification, whereas all waiver requests will not necessarily produce waiver approvals).
 See GAO WAIVER CONCERNS, supra note 19; see also Rosenberg, supra note 3 (citing the Clinton Administration’s commitment to expedited review).
 See GAO WAIVER CONCERNS , supra note 19.
 See Elizabeth Andersen, Administering Health Care: Lessons from the Health Care Finance Administration’s Waiver Policy-Making , 10 J.L. & POL . 215, 225-228 (1994) [hereinafter Andersen HCFA Waivers ].
 See United States v. Mead, 533 U.S. 218, 233 (2001).
 See id .
 See id .
 Agency practice is a less straightforward inquiry with regards to waiver continuation approvals. First, the agency is tied to the amount of process dictated by statute; thus, it seemingly cannot indicate an intent to make law by voluntarily providing more process. The agency may not be so tightly bound, however. Subsection 1315(f) does outline process steps, but limits those steps to responses that must occur within given timelines. The section does not foreclose the possibility of the Secretary giving notice to other parties and soliciting comments within the statutory timeframe for response. Likewise, nothing in the statute appears to prevent a state from involving other parties in the continuation request process. Studies on the waiver program refer to waivers as a general category and do not separately discuss waiver continuations. See e.g., Mann, supra note 14. Due to this lack of data on waiver continuations, it is not currently possible to comment on agency practice with regards to process in this area.
 MANN, supra note 14, at 12.
 Id .
 See GAO WAIVER CONCERNS , supra note 19.
 Id .
 See id.
 See MANN, supra note 14, at 12.
 See GAO WAIVER CONCERNS , supra note 19.
 Form letter from Dennis G. Smith, Director, Center for Medicaid & State Operations, to “State Medicaid Director” (May 3, 2002) (further noting that “[CMS] reached a broad audience in the development of the report to the President on the New Freedom Initiative. The Administration employed a number of methods for soliciting public input including a very successful National Listening Session. Such efforts strengthen the public policy decision-making, identify innovative strategies, and improve coordination of the many different participants in the health care and social services systems that Medicaid waivers often affect.”) available at www. cms.hhs.gov/states/letters/smd50302.asp
 See GAO WAIVER CONCERNS, supra note 19 (citing instances in which states failed to provide written copies of waivers upon request or withheld waiver proposals until after CMS approval).
 See letter from Senators Patty Murray and Maria Cantwell and Representatives George Nethercutt, Adam Smith, and Jay Inslee to Tom Scully, Administrator, Centers for Medicaid and Medicare Services (May 8, 2002) (expressing disappointment that Washington state no longer proposes a requirement for legislative approval of waivers and requesting that CMS maintain such a requirement due to the waiver proposals’ “broad impact on children and their families.”)
 See letter from Joel D. Ferber, Attorney, Legal Services of Eastern Missouri, Inc., to Governor Bob Holden, February 19, 2002 (noting that state was negotiating a controversial decision to cut benefits without public hearing but did hold a public hearing for a waiver that sought an increase in benefits), available at www.healthlaw.org/pubs/waivers/Missouri.Letter.doc
 See GAO WAIVER CONCERNS , supra note 19.
 MANN, supra note 14.
 S. 3018, 107th Cong. § 706 (2002).
 The journal of one of the country’s major health care organizations forecasts that looking forward from 2002, “Medicare Will Get Much Attention, Little Action.” Under this heading, the article notes that during the five preceding years, Congress “has been working to find a path that enough can agree on to reform Medicare.” The author attributes Congress’ inaction to a basic policy disagreement between congressional Republicans, who would like to expand Medicare beneficiary choice among private-sector health plans and congressional Democrats who wish to retain Medicare’s current structure but improve benefits. See Steve Cole, No Ordinary Year , PERMANENTE JOURNAL , Vol. 6, No. 3 (2002), available at www.kaiserpermanente.org/medicine/permjournal/sum02/update.html .
 Rosenberg, supra note 3, at 551.
 Id . The word “experimental,” although not directly footnoted in the Rosenberg piece, comes from both the Senate’s 1962 discussion of the waiver bill, S. Rep. No. 1589, 87th Cong., 2nd Sess., 1962 U.S.C.C.A.N. 1943 (1962) and the waiver provision itself which applies to “any experimental, pilot, or demonstration program,” 42 U.S.C. § 1315(a). Criticism of the Secretary’s assumed authority has arisen most recently with regards to the Health Insurance Flexibility and Accountability Demonstration Initiative (HIFA) which seeks to expand state flexibility under the waiver program, see NATIONAL HEALTH LAW PROGRAM, WHAT IS HIFA AND WHY SHOULD WE BE CONCERNED (2002), available at www.nhelp.org/waiver.shtml.
 Rosenberg, supra note 3, at 551.
 42 U.S.C. § 1315(a).
 42 U.S.C. § 1315(a).
 Letter from Representatives John D. Dingell, Committee on Energy and Commerce; Henry R. Waxman, Committee on Government Reform; and Sherrod Brown, Committee on Energy and Commerce, representing the House Committee on Energy and Commerce Democrats, to Tommy G. Thompson, Secretary of the U.S. Department of Health and Human Services (April 26, 2001).
 CENTERS FOR MEDICARE & MEDICAID SERVICES, HEALTH INSURANCE FLEXIBILITY AND ACCOUNTABILITY DEMONSTRATION INITIATIVE: INTRODUCTION , available at www.cms.hhs.gov/hifa/default.asp [hereinafter CMS HIFA INTRODUCTION ] (last visited Apr. 28, 2003).
 See MANN, supra note 14.
 42 U.S.C. § 1315(e)(6).
 42 U.S.C. § 1315(e)(7). Subsection 1315(e)(6) also makes reference to the constraints of 1315(e)(4). 1315(e)(4) provides for the term of extension upon approval of the request. Beyond this temporal change, 1315(e)(4) does not make any mention of modifications in the substance of the program. The term “federal budget neutrality” is used in this article to distinguish (1) the explicit charge of 1315(e)(7) to ensure that Federal expenditures (on the state’s Medicaid plan) towards the project will not exceed Federal expenditures (on the state’s Medicaid plan) in its absence from (2) a general charge of budget neutrality with regards to both state and federal Medicaid expenditures.
 42 U.S.C.A. 1315. The text of Subsection 1315(f) is as follows:
An application by the chief executive officer of a State for an extension of a waiver project the State is operating under an extension under subsection (e) (in this subsection referred to as the "waiver project") shall be submitted and approved or disapproved in accordance with the following:
(1) The application for an extension of the waiver project shall be submitted to the Secretary at least 120 days prior to the expiration of the current period of the waiver project.
(2) Not later than 45 days after the date such application is received by the Secretary, the Secretary shall notify the State if the Secretary intends to review the terms and conditions of the waiver project. A failure to provide such notification shall be deemed to be an approval of the application.
(3) Not later than 45 days after the date a notification is made in accordance with paragraph (2), the Secretary shall inform the State of proposed changes in the terms and conditions of the waiver project. A failure to provide such information shall be deemed to be an approval of the application.
(4) During the 30-day period that begins on the date information described in paragraph (3) is provided to a State, the Secretary shall negotiate revised terms and conditions of the waiver project with the State.
(5)(A) Not later than 120 days after the date an application for an extension of the waiver project is submitted to the Secretary (or such later date agreed to by the chief executive officer of the State), the Secretary shall--
(i) approve the application subject to such modifications in the terms and conditions--
(I) as have been agreed to by the Secretary and the State; or
(II) in the absence of such agreement, as are determined by the Secretary to be reasonable, consistent with the overall objectives of the waiver project, and not in violation of applicable law; or
(ii) disapprove the application.
(B) A failure by the Secretary to approve or disapprove an application submitted under this subsection in accordance with the requirements of subparagraph (A) shall be deemed to be an approval of the application subject to such modifications in the terms and conditions as have been agreed to (if any) by the Secretary and the State.
(6) An approval of an application for an extension of a waiver project under this subsection shall be for a period not to exceed 3 years.
(7) An extension of a waiver project under this subsection shall be subject to the final reporting and evaluation requirements of paragraphs (4) and (5) of subsection (e) (taking into account the extension under this subsection with respect to any timing requirements imposed under those paragraphs).”
 42 U.S.C. § 1315(f)(3).
 42 U.S.C. § 1315(f)(4).
 See Rosenberg, supra note 3, at 549 (noting that HHS under the Clinton Administration “vowed to limit the administrative constraints on the states and to reduce the processing time for waiver requests.”)
 See id .; see also CMS HIFA INTRODUCTION , note 254; see also MANN, supra note 14.
 42 U.S.C. § 1302.
 See MANN, supra note 14..
 See supra note 11 on complication introduced by Justice Scalia’s dissent, United States v. Mead, 533 U.S. 218, 243-44 (2001).
 United States v. Mead, 533 U.S. 218, 235 (2001).
 For a list of the paper publications available from CMS, see www.cms.hhs.gov/manuals/cmstoc.asp.
 Electronic copies of the agencies’ question and answer letters regarding Vermont’s PDP are available from the Office of Vermont Health Access’s homepage at www.dsw.state.vt.us/districts/ovha/ovha27.htm#pdp , (last visited Apr. 22, 2003).
 Electronic copies of the agencies’ question and answer letters regarding Vermont’s VHAP are available from the CMS website at www.cms.hhs.gov/medicaid/1115/vt1115hap.asp , (last visited Apr. 27, 2003).
 Rosenberg, supra note 3, at 552.
 See NATIONAL HEALTH LAW PROGRAM, SECTION 1115 Q&A , available at www.napas.org/I-3/I-3-d/Section %201115%20Q.htm (last visited Apr. 16, 2003).
 This connection between waiver applications and the role of federal government has been noted by several senators concerned with the public accessibility of the waiver process. See letter from Senator Max Baucus, Chairman, United States Senate Committee on Finance, to Tommy G. Thompson, Secretary, Department of Health and Human Services (Apr. 23, 2002).
 See Andersen HCFA Waivers , supra note 228, at 230.
 See Medtronic, Inc. v. Lohr, 518 U.S. 470, 485 (1996).
 See 59 Fed. Reg. 49249, 29450 (1994), as cited in Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp.2d 1, 5 (D.D.C. 2001).
 While not specifically endorsing the current Medicaid statute, several U.S. senators have expressed concern that the Secretary is exceeding his or her waiver authority in circumvention of statutory guidelines and Congressional intent. See letter from Representatives John D. Dingell, Committee on Energy and Commerce; Henry R. Waxman, Committee on Government Reform; and Sherrod Brown, Committee on Energy and Commerce, representing the House Committee on Energy and Commerce Democrats, to Tommy G. Thompson, Secretary of the U.S. Department of Health and Human Services (April 26, 2001). In addition, a bill to clarify specific limits on the Secretary’s authority was introduced in 2001. See S. 3018, 107th Congress, § 706 (2001).
 See Bentivoglio, supra note 128, at 1609.
 The market for physicians’ services is local in the geographic consumer access sense as embodied in antitrust analysis of physician markets, see U.S. DEPARTMENT OF JUSTICE & FEDERAL TRADE COMMISSION, STATEMENTS OF ENFORCEMENT POLICY AND ANALYTICAL PRINCIPLES RELATING TO HEALTH CARE AND ANTITRUST (1994) (reprinted in 67 ANTITRUST & TRADE REG. REP . (BNA) No. 1682, at S-1). Physicians’ services are also a local market due to the fact that local prices will reflect local variations in the cost of providing care, which in turn reflect factors such as the cost of living, see INDIAN HEALTH SERVICE, U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES, LNF STUDY PART II TECHNICAL REPORT (commenting on what factors to consider in designing a mainstream health benefits package for American Indians and Alaskan Natives), available at www.ihs.gov/NonMedicalPrograms/Lnf/PIIMethods.htm.
 For a more extensive explanation of drug pricing, see CONGRESSIONAL BUDGET OFFICE, HOW THE MEDICAID REBATE ON PRESCRIPTION DRUGS AFFECTS PRICING IN THE PHARMACEUTICAL MARKET (Jan. 1996) [hereinafter CBO MEDICAID DRUG REBATE ]
 Furthermore, many PBMs are large regional or national companies themselves who thus can maintain the large geographic market for pharmaceuticals.
 UNITED STATES GENERAL ACCOUNTING OFFICE, REPORT TO CONGRESSIONAL REQUESTERS, PRESCRIPTION DRUGS: EXPANDING ACCESS TO FEDERAL PRICES COULD CAUSE OTHER PRICE CHANGES . GAO/HEHS-00-118 (2000) [hereinafter GAO EXPANDING ACCESS ]
 Pharmaceutical Manufacturers of America v. United States, 135 F.Supp.2d 1, 5 (D.D.C. 2001).
 Pharmaceutical Manufacturers of America v. Thompson, 191 F.Supp.2d 48, 54 (D.D.C. 2002).
 See Pharmaceutical Manufacturers of America v. United States, 135 F.Supp.2d 1, 11-13 (2001) and Pharmaceutical Manufacturers of America v. Thompson, 191 F.Supp.2d 48, 61 (2002) (finding that PhRMA and its members had standing to challenge the Secretary’s interpretation of the drug rebate clause).
 See Vermont Pharmacy Rebate Program Amendment Request: Questions & Answers (Sep. 11, 2000) (in which CMS asked “[w]hat reaction does the State anticipate from the pharmaceutical industry it this waiver is approved?”)
 Smith, supra note 239.
 One writer has suggested that in the area of health policy, strong interest groups have so dominated the political process at the federal level as to take away the ability of the states to serve as policy laboratories. See Laguarda, supra note 192.
 The three states have joined forces to form a prescription drug purchasing collective known as the Tri-State Prescription Drug Purchasing Coalition. See MEDICAID PRESCRIPTION DRUG BENEFIT , supra note 34. In addition, Massachusetts, Connecticut, New York, Rhode Island, and Pennsylvania have joined the three northern New England states to form the Northeast Legislative Association on Prescription Drug Prices. The Association will week to pool groups other than Medicaid. See id.
 Smith, supra note 239.
 S. 3018, 107th Congress, § 706(a)(1) (2001) (amending Section 1115 by inserting an additional set of procedural requirements to the general substantive grant of waiver authority).
 As noted by CMS, see GAO WAIVER CONCERNS , supra note 19.
 Bentivoglio, supra note 128.
 Letter from the Office of Vermont Health Access to HCFA (full cite not available, as material was removed from CMS’s website during this writing).
 Letter from Tommy Thompson, Secretary of Health and Human Services, to Senator Max Baucus (May 6, 2002) (stating that the CMS website “already provides some material regarding individual state waivers”).
 Pharmaceutical Research and Manufacturers of America v. Thompson, 251 F.3d 219, 223 (D.C. Cir. 2001).
 For the circuit court opinion reversing the lower court’s grant of an injunction to stop the Maine Rx program, see Pharmaceutical Research and Manufacturers of America v. Concannon, 249 F.3d 66 (1st Cir. 2001), cert. granted , 536 U.S. 956 (Jun. 28, 2002).
 New State Ice Co. v. LiebMann, 285 U.S. 262, 311 (1932) (Brandeis, J. dissenting) as cited in Laguarda, supra note 192.
 See Medtronic, Inc. v. Lohr, 518, U.S. 470, 485 (1996) (emphasizing the “historic primacy of state regulation of matters of health and safety”).
 Andersen HCFA Waivers , supra note 228, at 221.
 See Rosenberg, supra note 3, at 551 (1995) (stating that “as more states experiment with Medicaid delivery systems, the country will develop a complete record of health care reform alternatives. Both other states and the federal government will refer to this record during their own efforts to design Medicaid reform plans.”)
 See Laguarda, supra note 192, at 160.
 See id . (arguing that the Supreme Court’s choice to overlook the role of modern interest groups in the legislative process has resulted in a misconception of federalism under which states are capable of acting as laboratories of democracy, especially in the area of health care reform).
 See Rosenberg, supra note 3, at 551.
 See Jerry Stanton, Lesson for the United States from Foreign Price Controls on Pharmaceuticals , 16 CONN. J. INT’L L . 149, 160-64 (2000); see also PATRICIA M. DANZON, PHARMACEUTICAL PRICE REGULATION 15-29 (Ann Petty ed., 1997) (as cited in Christopher R. Stambaugh, State Price Control Laws are the Wrong Prescription for the Problem of Unaffordable Drugs , 12 FORDHAM INTELL. PROPR. MEDIA & ENT. L.J. 897, 899 (2002)).
 For a history of cooperative federalism and its rise in political popularity post-1970, see Philip J. Weiser, Towards a Constitutional Architecture for Cooperative Federalism , 79 N.C. L. REV . 663, 668-673 (2001).
 See Philip J. Weiser, Federal Common, Cooperative Federalism, and the Enforcement of the Telecom Act , 76 N.Y.U. L. REV . 1692, 1696 (2001).
 See id. at 1698
 See id. at 1699.
 Andersen HCFA Waivers , supra note 228, at 224.
 See Christopher R. Stambaugh, State Price Control Laws are the Wrong Prescription for the Problem of Unaffordable Drugs , 12 FORDHAM INTELL. PROPR. MEDIA & ENT. L.J. 897 (2002).
 See GAO EXPANDING ACCESS , supra note 285 (studying the potential effects of opening the drug rebates available to federal departments and agencies to the Medicare program); see also CBO MEDICAID DRUG REBATE , supra note 282.
 For a discussion of manufacturer-purchaser price negotiations and a critique of government price controls, see Stambaugh, supra note 314. For a general description of such price negotiations, see JOHN HANSEN, UNITED STATES GENERAL ACCOUNTING OFFICE, UNITED STATES PRESCRIPTION DRUG PRICING AND REIMBURSEMENT POLICIES , available at pharmacos.udra.org/F3/g10/docs/tse/USA.pdf. (last visited Apr. 27, 2003).
 See Hansen, supra note 316.
 See Stambaugh, supra note 314, at 917.
 See Uwe Reinhardt, Perspectives on the Pharmaceutical Industry, HEALTH AFFAIRS 136-149 (2001); see also Stambaugh, supra note 314, at 900.
 See THE NATIONAL ADVISORY COMMITTEE ON RURAL HEALTH AND HUMAN SERVICES, U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES, KEY RURAL ISSUES FOR MEDICARE REFORM , available at http://ruralcommittee.hrsa.gov/reform_nac.htm .; see also Peter J. Hammer and William B. Sage, Antitrust, Health Care Quality, and the Courts , 102 COLUM.L.REV . 545, 642 (2002) (noting that “managed care plans are more likely to prosper in large metropolitan markets”).
 See THE NATIONAL ADVISORY COMMITTEE ON RURAL HEALTH AND HUMAN SERVICES, U.S. DEPARTMENT OF HEALTH & HUMAN SERVICES, KEY RURAL ISSUES FOR MEDICARE REFORM , available at http://ruralcommittee.hrsa.gov/reform_nac.htm.
 LEGISLATIVE ANALYST’S OFFICE, STATE OF CALIFORNIA, HMOS AND RURAL CALIFORNIA (2002), available at www.lao.ca.gov/2002/hmos_rural_ca/8-02_hmos_rural_ca.html
 With the surprising exception of New Hampshire, which enjoys a strong HMO penetration rate. Managed care market penetration rates as measured by HMO enrollment for 2001 are as follows: Maine – 29.2%; Vermont – 4.2%; Massachusetts 44.7; Connecticut 39.7% ; Rhode Island 34.6%; New Hampshire 38.1% (as reported by InterStudy, Competitive Edge 12.1, Part II: HMO Industry Report (St. Paul, Minnesota, 2002)).
 Vermont’s program applied to income-eligible (a) Medicare-covered individuals without drug coverage, and (b) other adults “’who do not have a benefit program that includes drug coverage.’” Pharmaceutical Research and Manufacturers of America v. United States, 135 F.Supp.2d 1, 5 (2001).
 See Scott D. Litman, Health Care Reform for the 21st Century: the Need for a Federal and State Partnership , 7 CORNELL J.L. & PUB. POL’Y 871 (1998).
 See Bentivoglio, supra note 128, at 1609.
 From 1993-1998, drug expenditures on average increased 12.4% per year, while average overall health care expenditures rose 5% per year, see GAO EXPANDING ACCESS , supra note . From 1992 to 2000, the number of retail prescriptions in the U.S. grew by an average of 6% per year, while the average cost of a retail prescription rose by 48% between 1990 and 2000, see MEDICAID PRESCRIPTION DRUG BENEFIT, supra note 34.
 See MANN , supra note 14.
 See Maine’s draft application submitted to CMS August 15, 2002, PHARMACY PLUS: A DEMONSTRATION PROGRAM UNDER SECTION 1115 DRAFT (2002), available at www.cms.hhs.gov/medicaid/1115/meapp.pdf.
 See SMITH , supra note 26; see also Leighton Ku and Matthew Broaddus, “Why Are States’ Medicaid Expenditures Rising?” Center for Budget and Policy Priorities (Jan 2003), available at www.cbpp.org/1-13-03health.htm www.cbpp.org/1-13-03health.htm.
 F. Lichtenberg, The Effect of Pharmaceutical Utilization and Innovation on Hospitalization and Mortality , National Bureau of Economic Research, Paper No. 5418 (1996) (as cited in ROBERT GOLDBERG, TEN MYTHS ABOUT THE MARKET FOR PRESCRIPTION DRUGS , National Center for Policy Analysis, Policy Report No. 230 (1999), available at www.ncpa.org/studies/s230/s230.html).
 Bitter Medicine: Pills, Profit and the Public Health , ABC NEWS (May 2002), available at abcnews.go.com/onair/ABCNEWSSpecials/pharmaceuticals_020529_pjr_feature.html (last visited Jan. 18, 2003).
 For example, the Journal of Pharmacoepidemiology is dedicated solely to the issues surrounding assessment of drug therapies.
 COLUMBIA UNIVERSITY HEALTH SCIENCES, GUIDE TO CLINICAL PREVENTIVE SERVICES: COST-EFFECTIVENESS AND CLINICAL PREVENTIVE SERVICES , available at cpmcnet.columbia.edu/texts/gcps/gcps0010.html (last visited Apr. 28, 2003).
 NATIONAL INSTITUTE FOR HEALTH CARE MANAGEMENT, CHANGING PATTERNS OF PHARMACEUTICAL INNOVATION (2002).
 Id .
 See Stambaugh, supra note 314.
 MANN , supra note 14.
 TERESA A. COUGHLIN ET AL., MEDICAID SINCE 1980: COSTS COVERAGE, AND THE SHIFTING ALLIANCE BETWEEN THE FEDERAL GOVERNMENT AND THE STATES (Urban Institute Press 1994).
 Id . at 87.
 On states’ use of the waiver process to shift costs to the federal government, see Andersen HCFA Waivers , supra note 228, at 229.
 MANN , supra note 14, at 14 (2001) (noting that thirteen out of the seventeen waivers approved up through 2001 included eligibility expansions and new eligibility groups).
 Statements on Introduced Bills and Joint Resolutions, Beneficiary Access to Care and Medicare Equity Act, 148 CONG. REC . S9715-01 (2002) (statement of Sen. Baucus regarding Section 706 of the bill, entitled “Improvement of the Process for the Development and Implementation of Medicaid and SCHIP Waivers”).
 42 U.S.C. § 1396.
 See 42 U.S.C. § 1396a, “State plans for medical assistance.”
 See letter from Senator John Dingell et al, Committee on Energy and Commerce Democrats, to Tom Scully, Administrator, Centers for Medicare & Medicaid Services, (April 18, 2002) (commenting on the HIFA initiative, see below, and noting that under a cap on state expenditures for pharmacy benefits, a state “would have to cut benefits from some people in order to have money to expand coverage to others” and stating concern regarding “suggestions [by CMS] that states may limit spending by either cutting off eligibility for people who are currently getting benefits”). See also Rosenberg, supra note 3, at 555 (noting that “The [Oregon’s] original Medicaid population, and not the entire population of the state, therefore bore the burden of health care reform by giving up important medical services to enable the previously uninsured to receive basic health care.”)
 For a discussion of the improper use of waivers as health care policy reformers based on Congressional intent, see Andersen HCFA Waivers , supra note 228 at 223-227.
 On the topic of Chevron’s basis in the separation of powers, see Kenneth A. Baumberger, Provisional Precedent: Protecting Agency Flexibility in Administrative Policymaking , 77 N.Y.U. L. REV . 1272, 1273 (2002).
 MANN , supra note 14, at 19 .
 Id. at 11.
 CMS HIFA INTRODUCTION, supra note 254.
 See HIFA GUIDELINES , supra note 210.
 MANN , supra note 14, at 19. Mandatory groups include individuals that states are required to cover under the federal Medicaid law, including the elderly and disabled persons receiving SSI; children age six and over with household incomes up to the federal poverty level; and younger children and pregnant women with incomes below 133% of poverty. Optional groups include individuals who are not mandatory, but who may be covered under Medicaid without a waiver. Pregnant women, parents, children, and disabled and elderly persons with incomes over Medicaid mandatory eligibility levels fall into the optional group.
 Id . at 24.
 42 U.S.C. § 1396o(b)(3), as cited in PhRMA I.
 See HIFA GUIDELINES , supra note 210.
 See MANN , supra note 14.
 See Dayna Bowen Matthew, The “New Federalism” Approach to Medicaid: Empirical Evidence that Ceding Inherently Federal Authority to the States Harms Public Health . 90 KY. L.J. 973, 974 (2002) (citing President Clinton’s 1993 call for HCFA to streamline the waiver process; further noting that “[t]wo types of Medicaid waivers allowed states to bypass the federal Medicaid restrictions during the early 1990s in order to enroll their citizens in managed care programs. By 1998, thirty-five states operated mandatory managed care enrollment using section 1915(b) ‘Freedom of Choice’ waivers while another seventeen states operated Section 1115 Research and Demonstration projects.”)
 See MANN , supra note 14. For a summary of the NGA’s position on Medicaid reform, see NATIONAL GOVERNORS ASSOCIATION, HR-43 MEDICAID REFORM PRINCIPLES POLICY , available at www.nga.org/nga/legislativeUpdate/1,1169,C_POLICY_POSITION^D_5113,00.html (last visited Apr. 28, 2003).
 See MANN , supra note 14.
 Medicaid Program; Demonstration Proposals Pursuant to Section 1115(a) of the Social Security Act; Policies and Procedures. 59 FED. REG . 49249 (Sep. 27, 1994) (see Section VIII).
 See MANN , supra note 14, at 24.
 See id.
 See NATIONAL HEALTH LAW PROGRAM, WHAT IS HIFA AND WHY SHOULD WE BE CONCERNED (2002), available at www.nhelp.org/waiver.shtml.
 See Rosenberg, supra note 3, at 555 (1995) (citing Thomas L. Friedman, President Allows States Flexibility on Medicaid Funds , N.Y. TIMES, Feb. 2, 1993, at A1, A13. See also Larry Stevens, States Test Medicaid Reforms , BUS. & HEALTH, Aug. 1994, at 51).
 For an economic analysis of the Reagan administration’s proposal, see THOMAS W. GRANNEMANN AND MARK V. PAULY, CONTROLLING MEDICAID COSTS: FEDERALISM, COMPETITION, AND CHOICE (American Enterprise Institute for Public Policy Research, 1983).
 Press Release, United States Department of Health & Human Services, Bush Administration Will Propose Innovative Improvements in States’ Health Coverage for Low-Income Americans (Jan. 31, 2003), available at www.hhs.gov/news/press/2003pres/20030131d.html.
 American Political Network, Medicaid: Washington Post Examines Bush Reform Proposal , 10 AMERICAN HEALTH LINE 9 (Feb. 10, 2003); see also Melanie Nathanson and Iris J. Lav, The Bush Administration’s Medicaid Proposal Would Shift Risks and Costs to States , CENTER FOR BUDGET AND POLICY PRIORITIES (2003), available at www.cbpp.org/2-12-03health.htm.
 See Melanie Nathanson and Iris J. Lav, The Bush Administration’s Medicaid Proposal Would Shift Risks and Costs to States , CENTER FOR BUDGET AND POLICY PRIORITIES (2003), available at www.cbpp.org/2-12-03health.htm.
 For a discussion of the financial structure of the Bush proposal and its negative effects on state health policy decisions, see CINDY MANN ET AL, ADMINISTRATION’S MEDICAID PROPOSAL WOULD SHIFT FISCAL RISK TO THE STATES, CENTER FOR BUDGET AND POLICY PRIORITIES (2003), available at www.cbpp.org/4-1-03health.htm.