Wal-Mart Stores, Inc. v. Knickrehm

MEMORANDUM OPINION AND ORDER EISELE, District Judge. By this lawsuit, Wal-Mart challenges the two-tiered reimbursement program established by the defendant under which chain pharmacies, including Wal-Mart and Walgreen, received 6.8% less revenue than independently owned pharmacies for filling prescriptions for the same drugs for Medicaid patients. Briefly stated, Wal-Mart alleges that the defendant failed to consider the appropriate criteria which it says are required by the Medicaid Act and its implementing rules and regulations and, instead, based its decision to go to a two-tier system upon factors that are impermissible under federal law. Wal-Mart also contends that to the degree that the defendant did address factors which are material under federal law, the administrative record provides no support for the defendant’s action. Walgreen is the plaintiff in another case pending on the docket in this Court, the same being Walgreen Company vs. Kurt Knickrehm, in his official capacity as Director, Arkansas Department of Human Services, No. 4-.00CV00382 GTE. In that case, Walgreen also challenges the two-tier system for reimbursement. However, in that case, Walgreen raises additional issues including a challenge to the Dispensing Fee. The Court has permitted Walgreen to participate in this trial as Amicus. Some background might be helpful. Background Medicaid is a cooperative federal-state program through which the federal government provides financial assistance to participating states that choose to reimburse certain costs of medical treatment for needy individuals. See Title XIX of the Social Security Act, 42 U.S.C. § 1396 et seq. Participation in the Medicaid program is voluntary. However, if a state elects to participate it must comply with the requirements of Title XIX and the implementing regulations promulgated by the Secretary of Health and Human Services. See 42 C.F.R. § 447.302. To participate in the Medicaid program a state must submit to the Secretary of HHS a State Plan which complies with the federal law. See 42 U.S.C. § 1396a(a). If the Secretary approves the State Plan, the federal government will pay a specific percentage of the “total amount expended on medical assistance” under that plan. Amendments to the State Plan must be approved by HHS and must meet the requirements of the federal law. See 42 U.S.C. § 1396a(b); 42 C.F.R. §§ 430.10, 430.12. An amendment to the State Plan is deemed to be approved automatically 90 days after its submission unless the administrator of the Health Care Financial Administration (HCFA) notifies the state that it disapproves the amendment or demands additional information. See 42 C.F.R. § 430.16(a). The Medicaid Act requires the states to pay for certain services and it gives them the option to provide additional services. Each state determines what it will pay for *751 services provided to Medicaid recipients so long as the payments are consistent with federal requirements. Payments are made directly by the state to those furnishing the services to Medicaid beneficiaries. The state of Arkansas has elected to participate in the Medicaid program. It has exercised its option to pay for prescription drugs under its State Plan. The Arkansas Medicaid Program is administered by the Department of Human Services. Wal-Mart operates pharmacies in 84 of its stores in Arkansas. Walgreen has 13 pharmacies in the state of Arkansas. Wal-Mart participates in the Arkansas Medicaid Program pursuant to a provider agreement under which the defendant agrees to reimburse Wal-Mart in accordance with federal and state laws and regulations for prescription drugs provided to Medicaid recipients. The federal law establishes general upper and lower limits on the amount for which the federal government will reimburse the states for prescription drugs provided under Medicaid. Pursuant to 42 U.S.C. § 1396a(a)(30)(A), each State Plan must set a reimbursement rate at a level that will: Provide such methods and procedures relating to the utilization of, and the payment for, care and services available under the plan ... as may.be necessary to safeguard against unnecessary utilization of such care and services and to assure that payments are consistent with efficiency, economy, and quality of care and are sufficient to enlist enough providers so that care and services are available under the plan at least to the extent that such care and services are available to the general population in the geographic area. HCFA’s regulations implementing this provision are found in 42 C.F.R. Part 440; Subpart F. HCFA has established separate systems for reimbursing providers for the cost of multi-source (generic) drugs on the one hand and brand name drugs on the other. This case deals only with the reimbursement provisions applicable to brand name drugs. For brand name drugs, HCFA reimbursement regulation 42 C.F.R. § 447.331(b) provides that: [State Medicaid] agency payments for brand name drugs... must not exceed in the aggregate payment levels that the agency has determined by applying the lower of the— (A) Estimated Acquisition Costs plus reasonable Dispensing Fees established by the agency; or (B) Providers’ usual and customary charges to the general public. HCFA’s rules define the term “Estimated Acquisition Costs” as follows: Estimated Acquisition Cost means the agency’s best estimate of the price generally and currently paid by providers for a drug marketed or sold by a particular manufacturer or labeler in the package size of drug most frequently purchased by providers. See 42 C.F.R. § 447.301. Most pharmacies in Arkansas are compensated under the Estimated Acquisition Cost branch of § 447.331(b) because this amount ordinarily is less than the usual and customary charges for a drug. This case challenges the legality of the defendant’s amendment to the Estimated Acquisition Cost reimbursement formula in the State Plan. The state reimburses pharmacies through payments of two separate components. First, the Department determines a “Dispensing Fee” which is a payment for each prescription filled that compensates pharmacies for the cost involved in providing the service of dispensing a prescription for a Medicaid patient. Effective July 1999, the Arkansas State Plan was amended to reduce the Dispensing Fee to a flat fee of $5.51 per prescription. *752 The second component requires the Department to determine the pharmacies’ “Estimated Acquisition Cost” of drugs. See 42 C.F.R. § 447.331(b)(a). As previously mentioned, Section 447.301 defines “Estimated Acquisition Cost” as the Department’s best estimate of “the price generally and currently paid by providers for a drug.” As stated, the level of the Dispensing Fee is not presently an issue in this case. However, the history regarding the establishment of that Dispensing Fee back in July of 1999 is pertinent to the issues of this case. In its trial brief Wal-Mart states: Wal-Mart does not challenge the Department’s position that, at least for some drugs, it may be paying a higher reimbursement rate to pharmacies than some private third party payors in Arkansas. Wal-Mart would not have objected to a reduction in the pharmacy reimbursement rate to a level that was reasonable, based on facts and non-discriminatory. The two-tiered ingredient cost reimbursement mechanism, however, blatantly discriminates against chains and is both impermissible and inequitable. Wal-Mart Tr.Bf. at p. 2. Further, in their oral argument, counsel for Wal-Mart in effect acknowledged that it would not be bringing this lawsuit if all of the other pharmacies filling brand name prescriptions for Medicaid patients under the State’s Medicaid plan had been treated the same as it. Under, the new State Plan, which was to go into effect at the end of April, 2000, Wal-Mart, as a “chain” pharmacy, is reimbursed for the ingredient drug cost at “Average Wholesale Price” (AWP) minus 17.3%, plus a flat Dispensing Fee of $5.51 for each prescription. But the reimbursement rate for “independent” pharmacies remains at AWP minus 10.5% plus $5.51. If all other pharmacies providing brand name drugs in the Arkansas Medicaid Program were reimbursed under the plan according to the same formula, i.e., AWP minus 17.3.% or AWP minus 10.5%, then Wal-Mart states it would not be making this challenge. 1 Because of the position taken by Wal-Mart, the defendant contends that Wal-Mart has no standing to challenge the two-tier reimbursement plan which went into effect for the first time in Arkansas in late April, 2000. It argues that since Wal-Mart acknowledges that the administrative record would support, /or all pharmacies participating in the program, the reimbursement of AWP minus 17.3% plus the $5.51 Dispensing Fee, it should not be able to complain simply because some three-fourths of the pharmacies participating in the program (i.e., the so-called “independent” pharmacies) are reimbursed at the rate of AWP minus 10.5% plus the $5.51 Dispensing Fee per prescription. Stated otherwise, the defendant argues that, although Wal-Mart may have a claim under the Equal Protection Clause of the U.S. Constitution, it has no claim under Section 1983 that the defendant violated federal law when it imposed upon Wal-Mart the lower reimbursement fee. This brings us to a consideration of the law controlling the disposition of this case. Relevant Law In Arkansas Medical Society, Inc. v. Reynolds, 6 F.3d 519, 522 (8th Cir.1993), the Eighth Circuit discussed the standard that should be applied in this case: The Medicaid providers brought suit in the district court under 42 U.S.C. § 1983 alleging that DHS had deprived them of federal rights by violating the Medicaid statute, specifically, 42 U.S.C. § 1396a(a)(30)(A). This particular sec *753 tion of the Medicaid statute and its corresponding regulation at 42 C.F.R. § 447.204 require states to ensure that Medicaid recipients have access to medical care that is at least equal to that of the general population. This feature of the Medicaid law is typically called the equal access provision. Later in the opinion, the Court discussed the applicable standards under the rubric “Violation of the Equal Access Provision”: The Medicaid providers allege that DHS has violated their rights under 42 U.S.C. § 1983 because DHS did not comply with the substantive requirements of the equal access provision. Specifically, the Medicaid providers contend that in cutting reimbursement rates to noninstitu-tional sources by 20%, DHS did not consider the relevant factors of equal access, efficiency, economy, and quality of care and thereby violated the statute. The parties are in agreement that the challenged action involves state agency rate-making as opposed to adjudication. In reviewing DHS’s rate-making decision, we must decide whether the action is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 414, 91 S.Ct. 814, 822, 28 L.Ed.2d 136 (1971). An agency acts arbitrarily and capriciously if it fails to “consider whether the decision was based on a consideration of the relevant factors.” Id. at 416, 91 S.Ct. at 824; American Paper Inst., Inc. v. American Elec. Power Serv. Corp., 461 U.S. 402, 413, 103 S.Ct. 1921, 1928, 76 L.Ed.2d 22 (1983). Review under the arbitrary and capricious standard is narrow, and a court may not substitute its judgment for that of the agency. Motor Vehicle Mfrs. Ass’n of the United States, Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983). While it is a deferential standard of review, the agency must still articulate a “rational connection between the facts found and the choice made,” Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87, 105, 103 S.Ct. 2246, 2256, 76 L.Ed.2d 437 (1983), and this court must engage in a “substantial inquiry.” Citizens to Preserve Overton Park, 401 U.S. at 415, 91 S.Ct. at 823. Our review of the State Plan must include “a determination whether the [state] plan complies with the requirements of federal law.” Illinois Health Care Ass’n v. Bradley, 776 F.Supp. 411, 417 (N.D.Ill.1991) (citation omitted), aff'd, 983 F.2d 1460 (7th Cir.1993). We agree that the trial court’s conclusion that the relevant factors that DHS is obliged to consider in its rate-making decisions are the factors outlined in 42 U.S.C. § 1396a(a)(30)(A). As already discussed, the equal access provision provides an unambiguous and compulsory framework to guide substantive agency decisions regarding reimbursement rates for noninstitutional providers. The statute requires that the reimbursement rates are sufficient “to assure that payments are consistent with efficiency, economy, and quality of care and are sufficient to enlist enough providers so that care and services are available under the plan at least to the extent that such care and services are available to the general population in the geographic area.” 42 U.S.C. § 1396a(a)(30)(A). The purpose of this subsection is to ensure adequate access and quality of care in the context of noninstitutional Medicaid providers, just as the purpose of the Boren Amendment is to ensure adequate access and quality of care in the context of institutional providers. Moreover, the language in the two subsections closely parallel each other. Accordingly, DHS must consider the relevant factors of equal access, efficiency, economy, and quality of care as designated in the statute when setting reimbursement rates. ;-s % # # # % *754 DHS has offered no evidence to show that the relevant factors have been considered. DHS argues that it followed the requirements of the Arkansas Administrative Procedures Act, but such procedural compliance is not the same as substantive adherence to the requirements of the Medicaid statute. DHS contends that it considered reimbursement rates for providers in other states. DHS has not shown, however, how such a comparison has any bearing on equal access, efficiency, economy, and quality of care in Arkansas. Id. at 529-30. Statutory Claim Prior to July of 1999, the State of Arkansas reimbursed participating pharmacies by paying them the Average Wholesale Price (AWP) minus 10.5% for the ingredient cost of the drug and, in addition, paid a Dispensing Fee of $4.51 plus a percentage of the cost of the drug. This Dispensing Fee percentage on an average added between 7.5% and 8% to the Dispensing Fee. In December 1997, the defendant Department hired the firm of Myers & Stauf-fer to conduct a survey of Arkansas pharmacies. That firm conducted a survey of both the dispensing and acquisition costs incurred by Arkansas pharmacists. See DHS Exhibit 2 at 1384-1466. This survey, dated July 15, 1998, studied the dispensing cost of 438 pharmacies in Arkansas (out of a total of some 740) which included 141 chain pharmacies and 297 independent 2 pharmacies. Id. Myers & Stauffer then statistically analyzed the data, and prepared a detailed report that was submitted to DHS on July 15, 1998. Id. DHS contends that it heavily relied upon this report when it established the two-tiered reimbursement rate that is the subject of this lawsuit. With respect to the Dispensing Fee component, the survey proposed three possible options for reducing outlays: Pursuant to the first Option, entitled “One Statewide Fee,” the Department would pay each pharmacy the same amount regardless of the services actually provided. Id. at 1389-90. The advantages of this approach were said to be “simplicity and equality of treatment.” Id. at 1390. The chief disadvantage was said to be that: no consideration is given to pharmacies regarding the nature and expense of the services they offer. For example, a pharmacy offering free, 24-hour delivery service received the same reimbursement as one not providing delivery service. Since items such as delivery service can be a significant portion of overhead, this disparity can be quite large. Id. Pursuant to the second Option proposed by Myers & Stauffer, each individual pharmacy would be reimbursed upon the basis of the cost data submitted by it. As stated by Myers & Stauf-fer the major advantage would be that “reimbursement rates are responsive to individual differences in cost. Pharmacies are neither overpaid or underpaid with respect to cost.” Id. The major disadvantage of this approach was its potential for “rewarding inefficiency” by paying unreasonably high Dispensing Fees to a few individual pharmacies. Id. Finally, under the third suggested Option, called the “Formula Based System,” Dispensing Fees would be based upon an objective formula that would “most nearly predict the individual pharmacy’s actual cost of dispensing prescriptions,” based on such factors as “urban location, percent of prescriptions delivered and percent of prescriptions *755 dispensed by owners.” Id. at 1391. According to Myers & Stauffer, this approach would “take into consideration the expense of an individual pharmacy’s services, while ... it does not reward inefficiéncy ....” Id. It was the recommendation of Myers & Stauffer that the difference in services provided by pharmacies should be taken into account, if at all, in the Dispensing Fee component of the formula. That firm did not suggest that such differences in the services provided by individual pharmacies should affect in any way the ascertainment of the appropriate reimbursement under the Estimated Acquisition Cost component. Furthermore, the firm did not attach any weight to the circumstance that an individual pharmacy was a member of a chain or not. To Myers & Stauffer, the only relevant factor in setting a variable Dispensing Fee was whether a pharmacy actually provided services in connection with the dispensing of drugs that were not provided by other pharmacies. The form of ownership of the pharmacies played no role in Myers & Stauffer’s analysis or recommendation. The defendant Department chose Option 1, i.e., one statewide flat fee of $5.51. 3 It is that decision which undercuts many of the arguments it makes in an attempt to justify the 6.8% reimbursement differential in favor of “independent” pharmacies. The federal law and federal regulations clearly authorize the defendant to establish a reimbursement formula which takes into account the various costs that might be necessary to dispense drugs to Medicaid recipients, e.g., delivery costs, 24-hour service, etc. But the law requires such costs to be determined and compensated for under the “Dispensing Fee” component, not the “Estimated Acquisition Cost” component. So, the defendant’s argument that it should be permitted to increase the reimbursement rate paid to the “independents” for the cost they pay for drugs to offset the additional costs it believes they incur in dispensing the drugs comes to naught, not only because the law does not permit that approach, but also because the record contains no data supporting the contention that all “independents” incur such costs or that all “chains” do not incur such costs. Why cannot the “Estimated Acquisition Cost” component be used to cover such dispensing services? The answer is clear. As pointed out earlier, the Estimated Acquisition Cost “means the agency’s best estimate of the price generally and currently paid by providers for a drug marketed and sold by a particular manufacturer or labeler ...” 42 C.F.R. § 447.301 (emphasis supplied). The costs of home-delivery of drugs and staying open 24 hours per day are real costs. No one argues to the contrary. But they do not affect the “price paid by providers for drugs.” Such costs may not be indirectly compensated for by over estimating what some' providers pay for drug ingredients. The Medicaid Act provides that the pay *756 ment for brand name drugs must not exceed the Estimated Acquisition Cost which is defined as the price paid by providers of the drug sold by a particular manufacturer or labeler. Therefore, the defendant may not give away taxpayer’s money by paying one group, here “independents,” more than the Estimated Acquisition Cost of the drug ingredients. This would be a subsidy and is clearly prohibited. The Statute and the federal regulations do not permit discrimination in payments made to providers that are similarly situated and within the same provider class, regardless of the form of ownership of those providers. Furthermore, the HCFA has clearly stated that the determination of the Dispensing Fee should be “separate and distinct from the EAC determination and unrelated to the cost of the drug product.” See August 17, 1994, Memorandum from the Director of the Medicaid Bureau to HHS Regional Officer, Wal-Mart’s Exhibit 24. Therefore, if there is to be a differential in reimbursement of “chains” and “independents” in connection with the “Estimated Acquisition Cost” component, that differential must be based upon information in the administrative record which rationally and reasonably supports the contention that “chains” pay something like 6.8% less than “independents” for the same drugs. The defendant believes Wal-Mart, and, assumedly, other “chains,” get secret rebates from the manufacturers of the drugs. But the administrative record is devoid of any evidence that this is true. To determine the Estimated Acquisition Cost, Myers & Stauffer asked 41 of the 438 pharmacies that responded to the survey to produce “all of their drug purchase invoices for a specified month in 1997.” DHS Exhibit 2 at 1388. Of the 38 pharmacies that responded to this request, only five were chains. Id. at 1407. The report declared that a sample of 38 was “an adequate basis for making valid findings” because “[o]ur review of the acquisition cost data and data we were collecting for a similar study in another state at the same time showed that the prices charged by wholesalers to pharmacies exhibit little variation.” Id. at 1388-89. Ultimately, Myers & Stauffer concluded that pharmacies were paying 82.7% of the “Average Wholesale Price” (AWP) for brand name drugs, and thus the Estimated Acquisition Cost should be AWR — 17.3%. Id. at 1391, 1408. Importantly, Myers & Stauffer concluded in their report that: “There was no significant difference between the acquisition costs of drugs for chains and independents shown by the invoice data.” Id. at 1407 (emphasis added). However, in January of 2000, Myers & Stauffer recanted this declaration in an email sent to Suzette Bridges of DHS. See DHS Exhibit 43. In this e-mail, Myers & Stauffer explained that because there were only five chains included in the sample for the acquisition cost survey, “it can not be determined based on the 1997/98 Arkansas acquisition cost study whether there is or is not a difference between the discounts obtained by chains and independents.” Id. Hence, Myers & Stauffer could not determine whether there was a statistically significant difference in the acquisition costs paid by independent and chain pharmacies in Arkansas. DHS also claims that it relied upon an acquisition cost study Myers & Stauffer performed in Louisiana for the calendar year ending December, 1998. See DHS Exhibit 19 at 448-458. Similar to the Arkansas survey, the Louisiana survey consisted of a relatively small, yet purportedly rehable, sample of 43 pharmacies, 20 of which were chains. 4 Id. 449, 453. Also similar to the Arkansas study, Myers & Stauffer found that the average Estimated *757 Acquisition Cost for pharmacies in Louisiana was AWP minus 17.0% for brand name drugs. Id. at 449, 451. However, in Louisiana, Myers & Stauffer found a statistically significant difference in the acquisition cost paid by chains and independent pharmacies. Id. at 453. In particular, the Louisiana report indicated that chains paid 83.2% to 84% of the AWP for brand name drugs, while independents paid 81.5% to 82.9% of the AWP. Id. at 453. Hence, there was a 1.1 to 1.7% difference in the price chains and independents paid for brand name drugs. The Court finds the Louisiana Myers & Stauffer inapplicable to the Arkansas Medicaid reimbursement rate for several reasons. First and foremost, the Arkansas appropriation statute declared that any changes in the Medicaid dispensing fee must be based upon an independent study of “pharmacists in Arkansas,” and not any other state. 1997 Ark. Acts 1360, § 119. Likewise, in Reynolds, the Eighth Circuit emphasized that “reimbursement rates for other states” had no “bearing on equal access, efficiency, economy, and quality of care in Arkansas,” and thus, the Department should not have relied upon the reimbursement rates used in other states. Reynolds, 6 F.3d at 530. Thus, this Court does not agree with the Department’s contention that the finding in the Louisiana report can provide a rational basis for treating chain and independent pharmacies differently in Arkansas. Moreover, even if the Department were allowed to rely upon the Louisiana report, there is no data therein that would support a differential of 6.8% in the reimbursement rates for independents and chains. In fact, this Court has searched the record in vain for any support for the Department’s decision to impose the 6.8% differential. Hence, it is the Court’s opinion that the Louisiana study, at the most, might have suggested to the Department that a further study in Arkansas should have been done to determine whether any possible price differentials existed between chains and independents. It does-not appear, however, that such a study was ever undertaken. Next, this Court must consider whether the definition of a “chain” as eleven or more retail pharmacies (located in or out of Arkansas) under one ownership bears any rational relatio'n to any of the substantive issues with which the defendant Department has expressed concerns. Consider the following' Concerns: 1) Whether independent or chain stores provide delivery services; 2) Whether independents ánd chains stay open 24 hours per day; 3) Whether independents or chains service nursing homes; 4) Whether independents or chains offer sales on credit; 5) Whether independents or chains pay more, less or the same, for drugs. The record reflects that the defendant Department relied on the “chain” definition suggested by the Arkansas Pharmacy Association which believes that providers who own more than eleven stores are able to obtain wholesale rebates. However, nothing in the record supports this statement. Rather, the enclosure submitted simply notes Bureau of Census statistical studies that use the “eleven or more” level as a basis for dividing the collection of data about all retail stores, not just pharmacies. Nothing supports the Department’s contention that a group of eleven or more pharmacies is able to obtain wholesale discounts unavailable to non-chain pharmacies. This Court cannot say on this record whether the definition used might turn out to have some merit upon further investigation and study. It can, however, state that the present administrative record does not support the rationality of using that definition in the context of the issues presented in this lawsuit. There are other circumstances that must be considered. The manufacturers and suppliers of brand name drugs, holding patents on such drugs, essentially enjoy a monopoly position. \ This lessens their incentive to offer discounts. Recognizing this circumstance, Congress amended the *758 Medicaid statute in 1990 to add a provision that requires that drug manufacturers, as a precondition for having their products covered by the program, agree to rebate to the states the difference between the manufacturer’s normal price and its “best price” to any customer. See 42 U.S.C. § 1396r-8. This rebate is collected by HCFA and distributed to the states (including Arkansas). Id. Thus, the defendant Department has the benefit of the most favorable acquisition cost as determined nationwide whether or not an Arkansas provider is able to buy at this “best price.” 5 Of course, this does not prove that chains such as Wal-Mart are not getting secret rebates as suspicioned by the defendant, but it does tend to lessen the force of such intuitive judgments. In any event, such suspicions and conjectures cannot substitute for an administrative record which factually supports the agency’s theories. It was acknowledged during the trial that some large companies like Wal-Mart can arguably obtain a competitive advantage by agreeing to assume some of a drug manufacturer’s costs, such as warehousing and distribution. For example, if Wal-Mart will store, warehouse, and deliver drugs to its stores that otherwise the manufacturer would have to store or deliver, then the manufacturer might agree to charge Wal-Mart the ordinary cost of the drugs less the cost for providing such facilities and services. However, a true price differential would only exist to the extent that Wal-Mart could perform these services and provide such facilities at a lower cost than could the manufacturer. Otherwise, it would be a “wash” or, if it cost Wal-Mart more than the manufacturer, then Wal-Mart would be paying more for the drugs than would the manufacturer’s customers that do not provide such services and facilities. Nevertheless, to the extent Wal-Mart can make a “profit” by taking over functions ordinarily performed by the manufacturers, then, to that extent, it could be argued that its costs of acquiring the drugs is less than other providers. There, however, is no data in the administrative record which deals with this possibility, the propriety of considering the same, or the quantification thereof. The use of the “chain” vs. “independent” dichotomy has other obvious problems. Myers & Stauffer stated in their report that any discounts or rebates that might be permitted by manufacturers would be equally available to chains, large independent pharmacies, and pharmacies that participate in buying groups to purchase drugs collectively. There is no data in the administrative record identifying the pharmacies falling into these groups. Still, we know from the report that they exist. Their existence undercuts further the rationality of dividing all pharmacies into the two categories. The defendant attempts to justify its two-tiered reimbursement program not only upon undisclosed rebates which it believes chains can and do exact, but also upon its belief that the chains give discounts to Blue Cross and other private payors. The factual predicate for this latter belief arises apparently out of a telephone conversation that Mr. Ray Hanley, the Director of the Arkansas Medicaid Program, had with someone at Blue Cross/ Blue Shield, suggesting that Blue Cross had been able to negotiate discounts amounting to AWP minus 15%. Mr. Han-ley described this as a “most favored nation” approach in which market discounts would be exacted from pharmacy providers depending upon what those providers accepted from private third party payors. Mr. Hanley observed that under this ap *759 proach the chains would be most likely the hardest hit, but he conceded that if one applied the approach across the board, then some independents might also be hit to the extent they granted such discounts to private insurers such as Blue Cross. Mr. Hanley stated that, “we all agreed on the chain discount as a compromise that cut some slack for the small providers.” See Plaintiffs Exhibit 13 at DHS 2320. The federal statutory and regulatory formula would not permit this approach even if there were a factual basis for it in the record. The Estimated Acquisition Cost is the price paid by the provider to the manufacturer. It is not the price that the provider might be willing to accept upon the subsequent transfer of that drug to a private insurance plan’s beneficiary. Furthermore, the Department was aware of information in July, 1999, that some pharmacies were already refusing to participate in private insurance prescription programs because of revenue reductions to a minus 15% reimbursement level which, as we all know, is more favorable than the Department’s proposal of AWP minus •17.3% for chains. Moreover, an e-mail sent to Suzette Bridges, the Administrator of the Arkansas Medicaid Prescription Program, states: In fact, one person I talked with yesterday, who is an independent, said that even the larger chains were not accepting some insurance because the discount off of AWP was so low. The AWP minus 15% looks like the lowest at this point. See Wal-Mart Exhibit 28. Such information puts in question the degree to which the Department was attempting to comply with the “equal access” requirement of the Medicaid statute when it adopted the AWP minus 17.3% for “chains.” The first public explanation of the Department’s two-tier proposal is found in the document entitled “Medicaid Pharmacy Cost: Slowing the Rate of Growth is Mandatory” which was released at the same time that the official notice was published. The document mentioned that Arkansas Medicaid drug expenditures were running 18% ahead of year earlier levels and that “the funded budget will not support this growth.” The Department stated that it would use its status as a “dominant purchaser” (i.e., the largest purchaser of prescription drugs in the state) “to claim market based prices” from chains. The Department is suggesting that instead of focusing on the cost pharmacies actually paid to manufacturers, as required by the law, the Department would use its dominant purchasing status to impose reductions in the revenues that chain pharmacies would receive for providing the drugs to the Medicaid beneficiaries. These reductions would not, however, be imposed upon “independent” pharmacies whether or not those independents made similar concessions to such private insurers. The Department’s justification for this approach was admittedly based upon anecdotal information leading it to believe that the “largest insurance companies are paying most chains an ingredient cost of Average Wholesale Price (AWP) less 15% to 16%.” The Department took the position that its proposed two-tier reimbursement system would “put reimbursement very much in line with what the largest insurance payors allow.” The justification document does not discuss the degrfee of discount the largest insurance payors were able to exact from “independent” pharmacies that participated in their programs. But it is clear the Department was aware that some independents accepted some reductions in revenues from private insurers such as Blue Cross. The document justified keeping the higher reimbursement rate for independent pharmacies (AWP minus 10.5%) as follows: More importantly ... is the recognition of the invaluable role small pharmacies play in the health care infrastructure of a rural state like Arkansas. It is the *760 independents, for the most part, who provide services like delivery (critical to homebound patients), charge accounts, nursing home services, and twenty-four hour access. Very few of these chains offer these services at the same time that the chains are putting tremendous market place pressure on independents to survive. If the independents continue to go out of business it is unlikely that the chains will add the services so important to low income, elderly and disabled patients. Plaintiff Exhibit at 10,286. The justification document did not mention that within the administrative record there are numerous e-mails and other comments showing that many chains provide many of these services and many independents do not. Wal-Mart Exhibit 29 at 3567-68; 3571. The simple answer to the whole “most favored nation,” Blue Cross insurer argument is that it is contrary to the federal law and regulations which control the determination of the Estimated Acquisition Cost of brand name drugs. In addition, there is no reliable data in the record upon which one could, with reasonable accuracy, estimate the amount of discounts which private insurance companies have been able to negotiate with either “chain” or “independent” pharmacies. Finally, there has been no showing of parallelism between private negotiated insurance programs and the Medicaid program. 6 The defendant Department has expressed concern for maintaining the “infrastructure” by which prescription drugs are distributed to Medicaid beneficiaries. This is of course, a serious and legitimate concern. What if pharmacies in the Delta or other sparsely settled rural areas are forced out of business for any reason? How then would the prescription needs of the poor be met? There are, however, limitations on what the defendant Department may do in dealing with such problems. If a small pharmacy is forced into bankruptcy because it cannot compete successfully with larger chains or independents, the assumption is that the survivor in that economic struggle will be there to provide the Medicaid prescriptions to the program’s beneficiaries. The Court, like the defendant, is sympathetic with the plight of small independent pharmacies when forced to compete with the “mega-stores.” But the Medicaid law does not permit the defendant to interfere with the free operation of the marketplace. There are laws, of course, which do deal with unfair competitive practices, should same exist, but the Medicaid Act is not one of them. Furthermore, there is no data in the record which reveals the extent to which this is a real problem for the defendant. In oral argument, the defendant appeared to acknowledge that the number of drug stores in the Medicaid program has remained fairly stable over the years. Nor can we estimate the impact upon the economic viability of “independents” of a 6.8% subsidy of their Medicaid prescription business. We do not know the percentage of total revenue derived from prescriptions or the percentage of total prescription rev *761 enue that is represented by brand name prescriptions for Medicaid beneficiaries. We do know that, on this record, no justification can be found for the two-tiered system pursuant to which “independents” are reimbursed 6.8% more than “chains.” Although the Myers & Stauffer report would support the defendant’s adopting a plan requiring reimbursement at AWP minus 17.3% across the board, the Court notes the defendant Department nevertheless, in 1999, chose not to change the EAC payment to AWP minus 17.3% because it determined that payment of AWP minus 10.5% was essential to comply with the statutory requirements of efficiency and economy while still ensuring access. To quote from a letter dated November 3, 1999, from Mr. Ray Hanley to Mr. Andrew A. Fredrickson of the HCFA: The acquisition cost survey performed by Myers & Stauffer determined that the average acquisition cost of brand name drugs is AWP minus 17.3%. To reiterate, this was the average. There were many brand name drug products sampled which had average acquisition costs in the range of AWP minus 11% to 16%. Setting the reimbursement level at AWP minus 10.5% ensures adequate reimbursement for a broad range of products. Wal-Mart’s Exhibit 2. Nevertheless, without any new study, the defendant Department initiated the new two-tier plan basing it principally on the old Myers & Stauffer study. The proposal was published within five weeks of the date the Department wrote to HCFA representing that AWP minus 10.5% was reasonable, supported by historical evidence, and necessary to meet the requirements of the law. When the Arkansas Legislative Council asked the purpose of the new two-tiered rule the defendant’s sole justification was: The purposed rule is necessary to slow the growth of retail drug costs by allowing Arkansas Medicaid to obtain discounts similar to those afforded private sector purchasers. Wal-Mart’s Exhibit 11. Although the defendant argued to the contrary, the Court finds that the two-tier plan was driven by budget and political considerations. It appears that the Department concluded that it could not obtain the necessary political support to impose the AWP minus 17.3% reimbursement standard across the board as suggested by its consultants. So, to obtain the budget savings the Department decided to impose the AWP minus 17.3% only on “chains,” which it defined with an eye toward the amount of savings it wished to generate. Moreover, it sought the support of the “independent” pharmacies by altering the balance of competition between them and the chains. The court understands that political realities cannot be ignored. However, the federal law and the Untied States Constitution do not permit such discrimination among those pharmacies which supply the identical drugs to Medicaid beneficiaries. The defendant estimated that the savings to be realized by imposing the lower reimbursement figures on the “chains” only will amount to in excess of $4 million per year. 7 Because less than 25% of the pharmacies fall within the defendant’s definition of a “chain,” one might expect that if the 17.3% figure were applied to all the pharmacies in the state, the taxpayers would be saved in excess of $16 million per year. If one uses the plaintiffs calculation, this figure could exceed $30 million. From the defendant’s brief and oral argument, the Court gets the clear impres *762 sion that the 17.3% for the “chains” was put out as a bargaining chip. The defendant expected the chains to come in and bargain with it to determine a more reasonable and viable figure. 8 And the record indicates that the defendant itself was, and is, unconvinced that such a rate will not interfere with the “equal access” requirements of the statute. The defendant argues that three other states have two-tiered systems: Texas, Washington and Michigan. However, in Washington the Dispensing Fee is tiered based upon the dispensing pharmacy’s actual volume, with large volume pharmacies being reimbursed at a lower dispensing fee than smaller pharmacies. Moreover, the distinction is based on the pharmacy’s actual volume, not the pharmacy’s ownership. The pharmacies are classified as high-volume, mid-volume and low-volume based upon the number of prescriptions dispensed. In contrast to the Arkansas Reimbursement Plan, the Estimated Acquisition Cost (EAC) in the Washington plan is not tiered. Therefore, the Washington Plan clearly is not like the two-tiered system in Arkansas. Neither is the Texas system. Texas has not adopted an EAC which discriminates between chains and independents. Instead, in Texas, the EAC is based on whether the dispensing pharmacy usually obtains a particular drug from a wholesaler or from a manufacturer (or distributor). No distinction is made between chain ownership and independent ownership. The Court cannot tell too much about the Michigan program from the information supplied, although it appears to have a two-tiered system, one for chains and one for independents. But the Court cannot determine from the record information about the differentials in reimbursement fees and whether such differentials are based on record data. In sum, the information about the Texas, Washington and Michigan systems adds nothing to the discussions of the pros and cons of the Arkansas two-tiered system. For the above stated reasons, the Court concludes that the two-tiered system is indefensible under the federal Medicaid statute and the supporting regulations. Equal Protection Wal-Mart argues that the Department’s two-tiered reimbursement plan violates equal protection rights guaranteed it by the Fourteenth Amendment to the Constitution. The Equal Protection Clause “commands that no State shall ‘deny to any person within its jurisdiction the equal protection of the laws,’ which is essentially a direction that-all persons similarly situated should be treated alike.” City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 439, 105 S.Ct. 3249, 87 L.Ed.2d 313 (1985). However, “dissimilar treatment of dissimilarly situated persons does not violate equal protection.” Klinger v. Department of Corrections, 31 F.3d 727, 731 (8th Cir.1994). If two similarly situated persons are treated differently, then, in reviewing economic and social welfare regulations, a court must determine if the rule authorizing the disparity bears a rational relation to a legitimate state interest. Arkansas Pharmacists Ass’n v. Harris, 627 F.2d 867, 870 (8th Cir.1980). First, the Department argues that chain pharmacies and independents are not similarly situated with respect to the issues raised in the lawsuit, an averment which, if true, would preclude further consideration of Plaintiffs equal protection claim. See Klinger, 31 F.3d at 731. Specifically, the Department contends that differences ex *763 ist between the types of services provided by chains and independents, that the United States Department of Commerce treats chains and independents differently in the area of reporting requirements, and that, in a previous lawsuit, Wal-Mart representatives testified that it was not in direct competition with small, independent pharmacies operating in Faulkner County, Arkansas. See Wal-Mart Stores, Inc. v. American Drugs, 319 Ark. 214, 891 S.W.2d 30 (1995). 9 First, it is important to note that the discrimination in this case is not predicated upon these differentiating factors. Here, the discrimination is based on the defendant’s reimbursing “independents” more than “chains” for the cost of drugs when there is no evidence that “chains,” as defined by the defendant, pay less for the same drugs. So, in connection with the plaintiffs equal protection clause argument, the Court finds that in purchasing the drugs for resale all of the pharmacies, “chain” and “independent,” are similarly situated and that there is no factual data in the administrative record that would permit such pharmacies to be treated differently with respect to the price paid for drug ingredients. Furthermore, the dissimilarities suggested by the defendant are not reliably supported by the record. While the Department contends that independent stores provide more services than chains, such as prescription benefits to the State’s nursing home population, 24-hour a day availability, and home delivery, there is no evidence in the record to support that contention. 10 In addition, it is undisputed that some chain stores do provide service to nursing homes, are open all day and/or deliver to home bound customers Likewise, it is clear that not every independent offers these extra services. So, as the Department’s two-tier plan now stands, it is, even under its own formulation, both over inclusive, rewarding some “independent” stores for providing services which they do not provide, and under inclusive, for failing to reward those “chain” stores which do furnish these additional benefits to customers. 11 Although the Court has determined that the chains and independents are similarly situated, it must still uphold the regulation against an equal protection challenge if the rate in question is rationally related to a *764 legitimate state interest. Arkansas Pharmacists Ass’n, 627 F.2d at 870. Under this test, substantial deference is given to the government in both “determining what constitutes a legitimate governmental objective and in selecting the means to accomplish the chosen objective.” Stiles v. Blunt, 912 F.2d 260, 263 (8th Cir.1990). So long as the challenged state action bears a rational relationship to a state objective not prohibited by the Constitution, the regulation survives an equal.protection challenge. More v. Farrier, 984 F.2d 269, 271 (8th Cir.1993). The Department contends that its two-tiered rate regulation is rational in that it protects smaller pharmacies which offer additional services that greatly aid the low-income, elderly and disabled citizens of the state. In effect, it argues that unless it is able to give independent pharmacies an opportunity to recover some of the costs of these additional services, they will eventually fail in the face of the chain pharmacies’ financial strength, causing the loss of substantial assistance to the poor, elderly, and infirm. Again, however, the Department’s contentions are not based on the facts present in the record. As mentioned, only some chains provide the services for which the Department gives all independents the credit. At trial, the Department admitted that it had no evidence that independent stores were failing, or that if a store did fail, whether another store, chain or independent, did or did not take its place and provide the same services. In short, the Department’s two tiered reimbursement program is based on factually unsupported, and sometimes in-disputedly incon-ect, assumptions, and, therefore, is not rationally related to any important state objective. Ergo, it violates the Equal Protection Clause of the Fourteenth Amendment. Conclusion The Court finds, concludes, and declares that the two-tiered reimbursement plan at issue here is arbitrary, capricious, an abuse of administrative discretion, discriminatory and in violation of the Medicaid Act and the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution. The plaintiff is entitled to judgment declaring that the application to it of the discriminatory reimbursement rate of AWP minus 17.3% violates the Medicaid Act and is unconstitutional under the Equal Protection Clause. The plaintiff is also entitled to a permanent injunction prohibiting the defendant Department from imposing the AWP minus 17.3% discriminatory ingredient cost reimbursement formula upon it, and it is also entitled to an order vacating the Amendment to the State Plan that established that reimbursement rate. Judgment will be entered accordingly. Footnotes [1] Walgreen, which is the plaintiff in the other suit, takes a different position. It states if the reimbursement rate is not increased above AWP minus 17.3% plus the $5.51, which it, as a "chain” receives, then its 13 stores in Arkansas may have to withdraw from the program. [2] In the Arkansas report, Myers & Stauffer clarified that "[tjhroughout this report, the term independent pharmacies will include chains with less than five individual pharmacies.” Id. at 1393. [3] The Myers & Stauffer report determined that the mean total cost to dispense a prescription in Arkansas is $4.71. The report suggested that Medicaid consider a profit allowance of $.80 per prescription. Id. at 1389. This forms the basis for the Dispensing Fee of $5.51. The report also provides additional information about the average weighted per prescription cost for "chain” stores (defined as five or more) and "independent” pharmacies. According to the report such cost for "independent” pharmacies amounts to $5.10 per prescription, whereas the cost for "chains” was determined to be $4.26. Id. at 1405. This suggests that the average "independent's” dispensing cost amounts to $.84 more per prescription than such cost to the average "chain.” However, it is clear from the report that this deferential is not based upon "independent” stores providing such extra services as free 24-hour delivery. So, what accounts for the $.84 deferential? The most obvious answer is that it reflects the deferential in economies of scale between larger volume pharmacies and the smaller operations. Whether the defendant has the power to adjust such deferentials is a question not before the Court. However, if it does, thát adjustment clearly must be made with respect to the Dispensing Fee component of the formula and not the acquisition cost component. [4] Although no definition of "chain pharmacies” can be found in the portion of the Louisiana report provided to the Court, the parties agreed in their briefs and during the hearing that "chains,” as used in the Louisiana report were pharmacies with five or more stores nationwide. Hence, the Myers & Stauffer reports in Arkansas and Louisiana defined chains and independents in the same manner. [5] The State of Arkansas received back in excess of $34 million dollars in such rebates for the year 1999. The amount returned constituted almost 17.6% of the total dollars spent on drugs by the Arkansas Medicaid program over fiscal years 1996 to 1999. This rebate is not credited to the Department's drug expenditures but, rather, to the State's general Medicaid budget. [6] In their Amicus brief and during the hearing, Walgreen argued that the Department's reliance on reimbursement rates paid by private insurance carriers, such as Blue Cross/ Blue Shield, was misplaced because comparing private insurance carriers with Medicaid was like comparing "apples with oranges.” The Court agrees. For example, Walgreen emphasized that patients who are insured by private insurance carriers generally pay a higher co-payment for their prescriptions, and, importantly, pharmacies will not fill their prescriptions unless the co-payment is first paid in full. In contrast, Medicaid patients are required to pay only a nominal co-payment, and the pharmacies must fill the prescriptions even if the Medicaid patients are unable to pay their co-payment. Additionally, a private insurance carrier can reduce its acquisition costs by obtaining a rebate or discounts from manufacturers in exchange for an agreement that the manufacturer’s drug, and not a similar drug made by a competitor, will be included on the private insurance carrier’s formulary. In contrast, Medicaid does not exclude certain brand name drugs from coverage. [7] The plaintiff states that the defendant Department erroneously estimated the impact of its proposal at $4,420,780 by using an incorrect Dispensing Fee of $7.79 per claim instead of the flat fee of $5.51 per claim that went into effect on July 1, 1999. If this error is corrected, the savings impact would exceed $9 million, i.e., double the defendant's estimate of $4.4 million. [8] Indeed, the defendant even argues that the plaintiff and the other ''chains” have an obligation to come in and show the Department why its amended plan is wrong, unfair, or unsupported. But this burden of proof argument represents a distorted view of the administrative process. The burden is upon the agency to justify its proposed changes. The administrative record which it creates must support, and rationally justify, the reimbursement rates it establishes. [9] In that case, Wal-Mart representatives, defending accusations that Wal-Mart priced some products below cost, testified to the contrary, asserting that the store was able to obtain volume discounts and take advantage of its efficiencies to drive down prices and still make a profit. This testimony, however, is not relevant in the instant case. First, the testimony did not outline any specific prices paid, instead using vague terms such as "volume discounts” and "efficiency.” Secondly and most importantly, the two-tiered rate employed by the Department applies and is discriminatory toward all chain stores, not just Wal-Mart. The defendant does not argue that it based its two-tiered system for all chains upon information it obtained about this Unfair Practices Act case from Faulkner county. Indeed, it appears from the record that the defendant had proposed this two-tiered amendment before it even learned of this case. The Faulkner County case might legitimately add to defendant’s suspicions and cause it to seek additional information, but it in no way adds to the legality of its two-tiered reimbursement plan. [10] At trial, the only evidence the Department submitted to support this position were, e-mail communications from independent pharmacists sent to Mr. Hanley supporting the new regulation and asserting that they provide some services which chains do not. [11] The Department’s other two mentioned differences, i.e., that the Department of Commerce maintains different reporting requirements for chains and independents and that Wal-Mart agents previously testified that they were not in direct competition with the smaller, independent pharmacies in Faulkner County, do not buttress its position. The fact that companies with eleven or more stores are required by the Census Bureau to file different reports for unrelated reporting purposes has no bearing on the reimbursement levels in question. Wal-Mart is not the only store affected by the regulation. Wal-Mart, Walgreen, and all other chain pharmacies are Medicaid providers. They provide the same medicines for indigents that independents provide, and are therefore similarly situated in that most relevant respect.