C. DARNELL JONES, II, District Judge.
This long pending multidistrict litigation involves antitrust claims brought by several Plaintiffs against companies engaged in the business of providing pharmaceutical benefits management services.
Presently awaiting decision in the lead case are a Motion for class certification pursuant to Fed. R. Civ. P. 23, and a Motion by Caremark to exclude Plaintiffs' expert evidence. Presently awaiting decision in Civil Action 06-4114 ("Express Scripts"), a class action brought by North Jackson against Express Scripts, Inc. ("Express Scripts") and Civil Action 06-4115 ("Medco"), a class action brought by North Jackson against Medco Health Solutions, Inc. ("Medco"), is a Motion to decertify a class that was certified before the case became part of the MDL. Presently awaiting decision in Civil Action 03-4730 ("Brady") is a class certification Motion similar to that pending in Caremark. For the following reasons, Caremark's Motion to exclude expert evidence shall be granted, Plaintiffs' Motions for class certification in the lead case and in Brady shall be denied, and the class previously certified in Express Scripts and Medco shall be decertified.
North Jackson is an IP in Jackson County, Alabama owned by Bryan Hicks. (See No. 06-4305, ECF 1, SAC ¶ 9; see also 06-MD-1782, ECF 248-2, Declaration of Bryan Hicks ("Hicks Decl.") ¶ 1-2
PBMs like Caremark contract with entities that sponsor prescription drug benefit programs such as employers, labor unions, health insurance plans, and health maintenance organizations (collectively "Plan Sponsors") to act as a third-party administrator for the Plan Sponsors' programs. (See ECF 260-2, Aug. 2005 FTC Study Pharmacy Benefit Managers: Ownership of Mail-Order Pharmacies, at 1-3 ("2005 FTC Study").) To fulfil their contracts with Plan Sponsors, PBMs contract with pharmacies in order to build networks of retail pharmacies that can provide the Plan Sponsors' enrollees with convenient access to prescriptions. (See id. at 1.)
Plaintiffs allege two distinct antitrust conspiracies under Section 1 of the Sherman Act wherein Caremark conspired to reimburse IPs at a lower rate than what it paid to Chain Pharmacies. First, Plaintiffs allege that Defendants are agents for their clients, the health plans, in fixing prices to IPs (the "Plan Sponsor Conspiracy"). (See SAC ¶¶ 66, 71; ECF 248 at 1 ("First, Plaintiffs allege that Caremark and other PBMs reached an unlawful agreement with their client payors.").) Plaintiffs allege that various health plans acted through Defendants in a horizontal conspiracy to lower the prices that they paid to IPs for prescription dispensing services.
Plaintiffs allege that Caremark violated the antitrust laws through the following practices:
(ECF 181 at 4-5; SAC ¶ 5.)
The Federal Trade Commission in the 2005 FTC Study explained the mechanics of how PBM reimbursement to retail pharmacies operates. (2005 FTC Study at 1-5.) PBMs contract with entities that provide prescription drug benefits to their enrollees, such as employers, labor union plans, and other entities, to manage those entities' prescription drug coverage benefits. (Id. at 2.) PBMs then establish networks of retail pharmacies to fill prescriptions for the Plan Sponsors' members. (Id. at 3.) Retail pharmacies receive revenue from the consumer in the form of co-payments collected at the point of sale and from the PBM in the form of reimbursements of the dispensed drug's ingredient cost plus any dispensing fee associated with filling the prescription, less the copayment ("the Reimbursement Rate"). (Id. at 4.)
As found in the Illinois Opinion, PBM administration of prescription drug benefit programs achieves a number of efficiencies. See id., 385 F. Supp. 2d at 749 (stating "the arrangement between Plan Sponsors and Caremark clearly has efficiency-enhancing potential. Caremark specializes in various functions of benefit plan administration and is likely able to achieve economies of scale in the performance of those functions that would otherwise be unavailable to Plan Sponsors. And the creation of retail pharmacy networks, which necessarily involves the setting of reimbursement rates, undoubtedly contributes to the success of that larger endeavor. What is more, there is a real question whether, on the other side of the coin, the arrangement actually has any countervailing anticompetitive consequences.") Unlike Plan Sponsors, PBMs are able to specialize in handling a variety of administrative functions involved in running prescription drug benefit programs, such as processing claims, maintaining patient records, creating and managing formularies, and negotiating discounts or rebates with drug manufacturers. Id. (see also Madsen Decl. ¶8; 2005 FTC Study at 1-2.) PBMs thus allow for collective reimbursement rate negotiations, avoiding the unworkable situation where each Plan Sponsor would need to negotiate separately with each pharmacy, an alternative that even the named Plaintiffs recognized would be inefficient and unmanageable. (Cordes Dep. at 328:5-8 (testifying that Big C does not have sufficient personnel to negotiate individually with each Plan Sponsor); Hicks Dep. at 446:14-448:3 (testifying similarly).)
Pharmacies contract with numerous PBMs and Plan Sponsors, and it is not unusual for a pharmacy to contract with more than 100 different PBMs and Plan Sponsors. (Madsen Decl. ¶ 14.) Pharmacies often also participate in multiple networks offered by a single PBM. (See Hicks Dep. at 87:21-89:18 (testifying that North Jackson is a member of 10 or 20 PBM networks, with some managed by the same PBM).) The networks may offer differing reimbursement rates, but the rate for each network is generally expressed according to the industry practice described in the 2005 FTC Study. (Hicks Dep. at 92:8-12; Cordes Dep. at 157:14-23, 175:4-14 (noting same PBM can have multiple networks, with differing reimbursement rates); see also Def. App'x A and B (summarizing different rates from North Jackson's and Big C's network agreements).) Caremark generally offers three kinds of pharmacy networks: (1) "access networks," which enable members to fill prescriptions at a pharmacy but do not specify particular rates; (2) "pricing networks," which provide for specified reimbursement rates and dispensing fees; and (3) "custom pharmacy pricing networks," which are tailored to a Plan Sponsor's particularized requirements. (Madsen Decl. ¶¶ 4-6.) Reimbursement rates and dispensing fees differ from one custom network to another and also within a custom network. (See id.) Caremark solicits a pharmacy's participation in any network that Caremark believes the pharmacy would be interested in joining, including any and all custom networks that have been established to serve clients in the pharmacy's local geographic area. (Id. ¶ 10.) The pharmacy can join all, none, or some of the networks that it is invited to join. (Id.)
According to Big C's Jeff Stewart, approximately 75% of Big C's prescription drug business comes from third-party payors. (ECF 181-3, Aug. 28, 2015 Decl. of Jeff Stewart at ¶ 3.) Medicaid accounts for approximately 20%, Express Scripts accounts for 10%, and all other third-party payors account for 54% of his business. (Id.) In 2004, Big C's average revenue per prescription dispensed was $47.00. (Id. ¶ 5.) Stewart asserts that the number of prescriptions dispensed by Big C has increased, due largely to picking up additional customers after another independent pharmacy in his area closed. (Id. ¶ 7.) He asserts that this other business failed due to low reimbursement rates and that Caremark continues to pay low reimbursement rates and divert plan members to mail order service. (Id. ¶¶ 7-8.) According to North Jackson's owner Bryan Hicks, Express Scripts accounts for approximately 25% of all prescriptions dispensed. (ECF 250-1, Aug. 28, 2015 Decl. of Bryan Hicks at ¶ 3.) His approximate overhead per prescription filled is $5.00. (Id. ¶ 4.) Hicks has attempted to negotiate reimbursement rates with Caremark and other PBMs, but his counteroffers were refused or ignored. (Id. ¶ 6.) He asserts that Caremark's reimbursement rate is far below that of the Alabama and Tennessee Medicaid agencies. (Id. ¶ 8.) For example, the Alabama Medicaid rate includes a dispensing fee 3-3½ times that paid by Caremark. (Id.) He has experienced a steady and dramatic decline in his gross margin on third-party payors, going from 8.94% in 1999 to 5.48% in 2015. His current gross profit per prescription is $2.07, down from $6.18 in 1999. His average revenue per prescription dispensed in 2004 was approximately $34.00 (Id. ¶ 9.) While he too has picked up additional customers due to other independent pharmacies closing, he asserts that his reduced profit is due to low reimbursement rates, and Caremark diverting customers to mail order. (Id. ¶¶ 11-12.)
To support their Motion, Plaintiffs submit multiple expert reports authored by Charles D. Cowan, Ph.D. and Paul J. Seguin, Ph.D.
By way of historical background, Cowan notes that, until the mid-1980s, most pharmacy benefits were offered through indemnity plans whereby the consumer paid the cash price for drugs then sought partial reimbursement — usually 80% — from their health plan. (Id. at 7.) He asserts that the long term impact of PBMs has been to drive a very large number of IPs out of business. In the 1970s and 1980s, the number of pharmacies in the United States was fairly stable or rising, but after 1992, the number started to decline rapidly, at a time when small businesses in other sectors of the economy were "booming." (Id.) Conversely, the number of chain outlets, mass merchandize outlets and supermarket outlets for pharmaceuticals, and their sales volumes, grew each year. (Id. at 8.) IPs lost market share to these other outlets in every year after 1993. (Id. at 9.) Over this time, mail order pharmacy sales grew as well, but only the sales by IPs declined. (Id. at 10; Chart 3.) Cowan notes that,
(Id.) He opines that,
(Id. at 13.)
After reciting the Plaintiffs claims — noted above and contained in SAC ¶ 5 — Cowan opines that the claims "have the effect of price-fixing, eliminating competition, indirect collusion among the PBMs, and an attempt to drive independent pharmacies out of the market." (Id. at 24.) He states that, other than the copay, "all other fees are determined contractually with each pharmaceutical company, in one or more contracts that each pharmacy has with each PBM. These values differ from contract to contract." (Id. at 25.) Cowan opines that "[w]hen there are more sellers and there is unequal strength between the sellers, relative market power is a major determinant of pricing. If there is a well-defined relationship between market power and the fees charged, then it should be possible to determine what fees would be in a market where prices are not artificially depressed." (Id.)
On the cost side of the equation, he notes that Plaintiffs do not claim that
(Id. at 27.) Cowan offers several criteria to determine whether IPs are being treated differentially after other factors, such as size, are accounted for. (Id.)
First, he proposes to determine if the average of "dispensing fee contract values falls below the bounds established in determining the size to fee relationship," opining that this would indicate that prices have been fixed and fees paid to IPs are artificially low. (Id. at 34.) He proposes that, if there is no natural variability in the contract values between IPs and PBMs — in other words all contracts with the PBMs vary much less than they do for the PBMs' contracts with pharmacies with six or more stores — then there is also an indication of price collusion. (Id. at 35.) He has specifically refused to specify the form of the regression tests that may be used to study these criteria since he did not have access to data that would tell him what forms to run.
Second, he proposes to specify a regression to examine reimbursement rates and dispensing fees and then allow the coefficients in the model to vary according to drug type. He opines that the goal of the analysis "is to discover, after accounting for all other available information, whether there is a pattern of practice that compensates independent pharmacist below the rate one would expect." (Id. at 39.) In so doing he would hold constant the pharmacies' network distinctions, the varying rates of reimbursement for different drug types, market concentration, and other factors. (Id.)
Finally, Dr. Cowan opines that his proposed testing methodology "leads to a method for computing damages." (Id. at 50.) He proposed eight different ways in which damages can be measured. First, he would calculate the
(Id.) Dr. Cowan asserts that, if tests show that IPs "are making less than would be expected given the number of stores, then the coefficient d [i.e., some particular drug such as Nexium] will be negative and a measure of the dollar loss for independent pharmacists on each sale of Nexium. If we multiply "d" times the total sales, we have a measure of damages just from sales for this type of drug. We can repeat this calculation for each type of drug and sum to obtain a total estimate of damages for artificially low dispensing fees." (Id.) To calculate the damages resulting from different reimbursement rates, he proposes using the same calculation, except that the coefficient would measure the difference in the percent of the average wholesale price ("AWP") offered to IPs versus chains. He would then apply this gap in AWP to sales to obtain total damages.
Second, to test whether ingredient fees paid to IPs are significantly less than fees paid to larger pharmacies, he would calculate the average difference between IP ingredient fees and the average predicted based on the model described based on size of pharmacy. He would then multiply this difference times all prescriptions processed by a PBM in the last four years, specifying that this calculation could be done at a more refined level for generics, brand drugs, each PBM individually, and by other factors important to understanding the difference between paid and predicted. (Id. at 51.) Third, to test whether "net returns," defined as dispensing fees plus ingredient fees minus charges, are significantly less than fees paid to larger pharmacies, Dr. Cowan would calculate the average difference between IP total compensation and the average predicted by his model, multiplied by all prescriptions processed. (Id.) Fourth, he would examine the variability between prices paid to IPs by PBMs for services since, he asserts, "if independents were able to compete fairly, then their variability of compensation would be higher, and in turn then would be compensated at a higher level too." (Id.) He would use a formula to find the increase in fees resulting from increased variation and apply the increase to all prescriptions written in the last four years. (Id.) Fifth, he would examine the average IP price and average non-independent price paid by uninsured consumers and compare that ratio to a similarly calculated ratio for insured customers. He opines that on a "level playing field" the ratios would be equal, but if IPs are unable to compete fairly, the "insureds" ratio would be less than the "uninsureds" ratio. The product of the ratios multiplied by the IPs' revenue would equal the damages resulting from the lack of a level playing field. (Id. at 52.) Sixth, Dr. Cowan proposes that, for IPs rejected for participation by PBMs, he would compute the average revenue ratio as the revenue that would have been received for rejected contracts and multiply this ratio by the total revenue for the IPs to obtain "corrected" revenue. (Id.) Seventh, since PBMs choose AWPs from a set of AWP rates that favor PBMs rather than their third party payor clients, and which also hurt IPs out of proportion to large pharmacies, Dr. Cowan proposed to use "average AWPs (since real AWP is unknown as they are multiple estimates of same value). Compute fees based on average AWP instead of chosen AWP and compute difference on each transaction — sum of differences is damages." (Id. at 53-54.) Finally, he proposes using the same method described in his fourth alternative, but substituting pricing information from RxHUB. (Id. at 54.)
Paul Seguin, Ph.D., submitted a report for the Plaintiffs entitled "Antitrust Violations by Pharmacy Benefit Managers" on September 23, 2014 ("2014 Seguin Report"), in which he outlines "not only how Cowan's tests were feasibly empirically implemented, but how these tests provide evidence of anti-competitive behavior, provide an estimate of damages and demonstrate injury to identifiable class members." (Id. at 1.) Specifically, Seguin focuses on the first three tests proposed by Dr. Cowan: (1) dispensing fees, (2) ingredient fees, and (3) net returns paid to Plaintiffs, each of which he opines are significantly less than those of larger pharmacies. (Id. at 3.)
Using Caremark data that was unavailable to Dr. Cowan, Dr. Seguin opines he is able to identify that IPs received reimbursements that "were both economically importantly and statistically reliably different than those received by pharmacies that are members of a chain for both years for which data was provided — 2005 and 2008. Specifically, IPs received, on average $1.73 less per non-generic (or "brand-name") prescription filled." (Id. at 1.) Seguin calculates that the aggregate damages suffered by the putative Class for those two years is $23,011,622. (Id.) He asserts he can apply the same methodology to all years deemed relevant and for other PBMs, as well as allocate damages among each class member. (Id.) Using data from 2002-2012 to update Cowan's analysis of IP market share, Seguin opines, consistent with Cowan's findings, that the trend of a reduced market share for IP's continued during this period. He interprets this evidence as being consistent with Plaintiffs' hypothesis that PBMs' anti-competitive practices have a detrimental effect on the competition for the dispensing and sale of prescription drugs that has manifested itself as an elimination of the growth of IPs while their Chain counterparts continue to grow. (Id.)
Dr. Seguin was provided with transactional data from Caremark containing all non-generic prescriptions filled for customers who had a prescription plan managed by a PBM for the two calendar years of 2005 and 2008. In total, 78,543,701 records were provided. (2014 Seguin Report at 3.) He used data on the total reimbursements received by pharmacies for filling a particular prescription and the quantity of drugs filled for that prescription. (Id.) He supplemented these data by creating a set of additional variables that, he opines, capture the locational-specific economic and competitive environment. He used 2000 Census data to calculate income — in the form of the median incomes for each ZIP, and populations for each ZIP. (Id. at 4.) He further calculated the number of pharmacies in a ZIP — specifically the number of storefronts, not just the number of different chains — and the number of pharmacies within a chain. (Id.) He then created an indicator variable representing whether or not the number of pharmacies within a chain was five or fewer, assigning a value of "1" if a particular pharmacy's name has five or fewer unique addresses associated with it and a value of "0" if a pharmacy's name has greater than five unique addresses. (Id.)
Next, Dr. Seguin estimated a regression model
Dr. Seguin also opines that he can similarly calculate damages across all non-generic prescriptions filled in 2005 that were filled at least 120 times by:
Dr. Seguin states that this sum, across all non-generic drugs filled at least 120 times for Caremark PBM members in 2005 equals $12,308,601. Damages for the non-generic drug names filled for Caremark PBM members in 2008 equal $10,703,061. Damages across the two years aggregate to $23,011,622. (Id. at 6-7.)
Dr. Seguin then set out to test the "robustness" of his findings. He opines that his finding is robust for a number of alternative statistical specifications including, but not limited to, (a) adding variables for income, population, and number of pharmacies in a ZIP; (b) including the natural logs (ln) of these three variables that control for cross-sectional differences in the competitive environment facing each individual pharmacy; (c) estimation via weighted-least-squares using (the inverse of) numerous combinations of the dependent variables listed above; (d) estimating models where the slopes varied with NPCFF which represents a direct test of Cowan's test asking whether "ingredient fees paid to plaintiffs are significantly less than fees paid to larger pharmacies." (Id. at 7-8.)
Dr. Seguin has also reviewed data on market share to support Dr. Cowan's conclusions that market share for IPs fell and the share for chain pharmacies grew. He finds that the chain group had an increase in the relative number of outlets of 19.2% over the period from 2002 to 2012, or a compounded annualized growth rate ("CAGR") of 1.8%. For the same period, the data indicate that the market share of IPs fell with a CAGR of -0.15%. (Id. at 10.) Seguin opines that the decline in the number of IP outlets over this period "strongly suggests that the common harm or `suffering' by Independent Pharmacies due to discriminatory reimbursement by PBMs endures." (Id. at 11.)
Finally, Dr. Seguin opines that, although he currently lacks the data to do so, he can feasibly empirically test for the presence of other anti-trust behavior identified by Dr. Cowan, namely, Cowan's hypotheses pertaining to (1) mail order pharmacies; (2) uninsured clients; and (3) the market for pharmacist services. (Id. at 11.) For the allegation that the PBMs' control of their own mail order pharmacy operations diverts high profit refill and maintenance prescriptions from retail pharmacies, Seguin opines that he can "test for the presence of this behavior." Specifically, he contends that "this hypothesis would be confirmed if I could demonstrate that IP's disproportionately filled prescriptions with a Quantity of one month or less, or, equivalently, whether `multi-month' prescriptions were disproportionally filled by mail order pharmacies." (Id. at 12.) For Cowan's hypothesis testing for discrimination in reimbursement for prescriptions filled for uninsured clients, Dr. Seguin opines that the "data fields would be identical to those provided in the most recent data run provided to us by defendants (especially Quantity and Reimbursement). Upon receipt of the data, statistical analyses similar to those outlined above would be performed. Damages would again be easily calculated as the product of reimbursement discrimination and the number of prescriptions filled." (Id. at 13.) For Cowan's hypothesis about the employment market for pharmacists, Dr. Seguin opines that Cowan has already demonstrated that, "although the supply of `active' pharmacists increases over Cowan's sample period beginning circa 1970, the percent of said pharmacists that are self-employed has fallen by roughly 2/3rds from 9.4% in 1983 to but 3.4% in 1998. Once commissioned, I will investigate whether this trend has continued." (Id.)
Caremark has submitted the May 15, 2015 Expert Report of Jerry A. Hausman ("2015 Hausman Report"). (ECF 261-10.)
(Id. at 3-4 (emphasis in original).)
By way of background, Dr. Hausman opines that the work of PBMs "enables them to develop ways to increase efficiency for plan sponsors and patients, and to improve patient care by aiding in the management of prescription drug use by patients through interaction with patients and other health care providers regarding patient care. Thus, PBMs play an important role in promoting efficiency and effectiveness in the health care delivery system." (Id. at 5.) He adds that (1) PBMs compete against each other to obtain plan sponsor clients; (2) PBMs must negotiate with plan sponsors, drug manufacturers, independent pharmacies, pharmacy chains, and businesses that negotiate on behalf of a collective of independent pharmacies; (3) during the period that Plaintiffs allege a conspiracy between PBMs was ongoing, the FTC's analysis of the Caremark/AdvancePCS merger concluded that "competition among PBMs will remain vigorous in the wake of the Caremark/AdvancePCS acquisition, and that this competition is likely to cause PBMs to pass on at least some of their cost savings to their customers in order to gain or retain their business" (see ECF 260-5, Statement of the Federal Trade Commission in the Matter of Caremark Rx, Inc./AdvancePCS, File No. 031 0239); (4) PBMs must meet the geographic needs of its clients; i.e., to gain the business of a plan sponsor with enrollees in a particular market, the PBM must include pharmacies in that market in its network; (5) pharmacies, including chains, must make a determination of the relative costs and benefits of participating in a particular network, including benefits from increased in-store business and non-pharmaceutical sales, and based on these considerations, some pharmacies, including some chains, do not participate in certain networks; (6) the contract between the PBM and the pharmacy governs the reimbursement rates for each particular network in which the pharmacy is a participant; (7) pricing in each network is different for brand name and generic drugs with the pharmacy typically reimbursed by the plan sponsor for the AWP less a negotiated discount percentage, plus a flat fee for dispensing; (8) pharmacies can be reimbursed for the dispensing of generic drugs either based on a similar formula, or based on what is called the maximum allowable cost ("MAC") rate list, which is a rate set by the PBM and which changes over time and plan. (Id. at 5-7.) Dr. Hausman notes that Dr. Seguin performed no analysis of reimbursement for generic drugs in his report.
Dr. Hausman states the following opinions regarding the class certification issues:
Dr. Hausman also provides a critique of Dr. Seguin's specification for a regression model. While Seguin's model would find damages relating to Lipitor in 2005 of $300,103, Hausman asserts that Seguin's test differs from Dr. Cowan's proposal in several important respects. The Cowan factors not accounted for by Dr. Seguin include any relationship between reimbursement and (1) the size of the chain, (2) network membership, (3) market concentration, (4) geographic and demographic factors, (5) membership in a bargaining collective, and (6) tradeoffs between brand drugs and generics in pricing. As a result, Hausman asserts, the test Dr. Seguin actually performs is invalid under even Dr. Cowan's standard.
According to Dr. Hausman, both Dr. Cowan and Dr. Seguin fail to provide a market definition, the first step in assessing the competitive effects of an alleged antitrust violation analyzed under the rule of reason. (Id. at 14.) There is no discussion of either the product market or the geographic market allegedly monopolized. He notes that Seguin testified at his deposition that he does not have "an expert or professional opinion on what is meant by geographic market," and that he has never attempted to define a boundary for a geographic market, or to identify the competitors within a geographic market. (Id. at 14 (quoting ECF 261-5, March 19, 2015 Dep. of Paul J. Seguin, Ph.D. ("Seguin Dep.") at 126-27).) Dr. Cowan, however, stated that geographic and demographic factors "must be included as variables in the set of equations to control for differences that may explain why there is variability in pricing." (Id. at 14 (quoting 2006 Cowan Report at 44).) Nonetheless, Dr. Seguin claimed that his model "has shown that geography is not a factor," and therefore pharmacies compete "regardless of [their] geographic location." (Seguin Dep. at 123-24.) Hausman opines that this "position is wrong as a matter of common sense, is contrary to the FTC's assessment, and is rejected as a matter of econometrics." (2015 Hausman Report at 14).
To rebut Dr. Seguin's assertion that geography is not a factor, Dr. Hausman attempts to test "whether the coefficients for pharmacies in one geographic area are different than the coefficients for pharmacies in another geographic area." (Id. at 15.) Hausman tests "whether the coefficients for pharmacies in Jackson County, Alabama (the location of the named plaintiffs) are different from the coefficients for pharmacies in the rest of the country. I perform the test for the top 5 most-dispensed drugs in Jackson County for both 2005 and 2008." (Id.) Hausman sought to obtain for each "the probability of obtaining the observed difference between the Jackson County coefficients and the coefficients for the rest of the country under the hypothesis that Dr. Seguin's assumption is correct." (Id. at 16.) He finds that the test rejects Dr. Seguin's assumption for 9 of the 10 drug-year combinations and thus concludes that "Dr. Seguin's assumption that `location is irrelevant' is incorrect." (Id., Table 1 (finding p values of 0.000 for 7 drug-year combinations, 0.004 for one drug-year combination, 0.019 for one drug-year combination, and 0.350 for one drug-year combination (Toprol XL).) Hausman reports that, while Dr. Seguin's assumption would calculate $4,273.01 in aggregate damages to Jackson County IPs for those ten drug-year combinations based on his national coefficient, a similar calculation using a Jackson County coefficient results in negative damages of -$1,489.87. (Id. at 16-17.) This would indicate that those IPs are estimated to have received higher reimbursements than chain pharmacies.
Next, Dr. Hausman criticizes Dr. Seguin's failure to demonstrate harm to competition in any relevant market. He opines that, although Dr. Seguin "claims that he was able to `empirically test for the presence of anti-competitive behavior,' Dr. Seguin's results do not demonstrate that competition has been harmed." (Id. at 17-18 (footnote omitted).) While Seguin stated that his hypothesis is that "a condition of a legal world would be that . . . chains and independent pharmacies would be reimbursed at the same levels," (see Seguin Dep. at 200), Hausman opines that "observing a reimbursement differential does not prove illegal behavior, because (as Dr. Cowan has acknowledged) large pharmacies receive better reimbursement than smaller pharmacies because they are in a better bargaining position. Indeed, because Dr. Seguin's test fails to account for the reimbursement-size relationship acknowledged by Dr. Cowan, Dr. Seguin's test would find anticompetitive behavior even when no such behavior has occurred." (2015 Hausman Report at 18 (parenthesis in original).) Hausman asserts that one must
(Id. at 19 (citing ECF 260-5 (FTC Statement on In re Caremark) in which the FTC found that the exercise of monopsony power was unlikely); ECF260-1 (FTC Statement on Express Scripts Acquisition of Medco Health Solutions) in which the FTC again found the exercise of monopsony power unlikely in the PBM industry).)
Next, Dr. Hausman offers a critique of Dr. Seguin's attempt to construct a but-for world to provide common evidence of damages. He notes that Seguin made no assumptions about what a but-for world would look like, and he professed to have no understanding of the conspiracies alleged in the SAC.
(2015 Hausman Report at 21-22 (footnotes citing Seguin Dep. omitted).)
According to Dr. Hausman, Dr. Seguin has only demonstrated that IPs are on average reimbursed less than chain pharmacies, but he has not shown that all or nearly all IPs are reimbursed less than chain pharmacies. Hausman opines, therefore, that Seguin has failed to provide evidence common to the class of antitrust impact. (Id. at 22-23.) Seguin assumed that only two factors affect the reimbursement a pharmacy receives: the quantity of the drug dispensed and whether the pharmacy is an independent or chain. According to Hausman, Seguin's estimated coefficient on the independent-pharmacy variable reflected the average reimbursement differential between independents and chains, holding quantity constant, and demonstrated that nearly all IPs were reimbursed less than chain pharmacies. Hausman attempts to test this assumption using an F test comparing a restricted model to an unrestricted model. Hausman's "restricted" model is Dr. Seguin's model, which assumes that there are no pharmacy-specific factors (other than being an independent or a chain) that affect reimbursement. The "unrestricted" model is a model that allows pharmacy-specific effects by adding what Hausman calls a "fixed effect" for each pharmacy. (Id. at 23.)
Hausman performed the F test for the top 5 drugs in 2005 and the top 5 drugs in 2008. The p value he calculated for these drug-year combinations "is approximately zero. Thus, the F test strongly rejects Dr. Seguin's assumption in favor of a model that allows for pharmacy-specific effects." (Id. at 24; Table 2.) Hausman opines that this result "means that there is significant variation in reimbursement across pharmacies that is not captured by Dr. Seguin's model and that Dr. Seguin's model cannot be relied upon to demonstrate common impact." (Id.) Focusing on one drug-year combination, Lipitor in 2005, Hausman asserts that the data show that that "almost all (99.78%) independents receive reimbursement that is greater than the reimbursement received by the lowest-reimbursed chain. Over half (52.60%) of independents receive reimbursement that is greater than the median chain. Some independents (1.68%) even receive reimbursement that is higher than the highest-reimbursed chain." (Id. at 25; Table 3.) According to Dr. Hausman, "[t]hese results demonstrate that when Dr. Seguin's incorrect assumption is removed, his own model shows that a substantial percentage of putative class members receive reimbursement that is greater than that received by many chains, and hence there is no antitrust impact that is common to members of the putative class."
Next, Dr. Hausman offers a critique of Dr. Seguin's analysis of the number of pharmacy outlets. While Seguin opines that the differential between the growth of IPs and non-IPs "strongly suggests that the common harm or `suffering' by Independent Pharmacies due to discriminatory reimbursement by PBMs endures," (see — Seguin Report at 10), Hausman opines that Seguin "fails to demonstrate that the observed change in the number of outlets is the result of the alleged `discriminatory reimbursement,'" or that "the observed change in the number of outlets demonstrates a harm to competition." (2015 Hausman Report at 26-27.) He notes that Seguin conceded at his deposition (1) that "there are numerous potential explanations" why the relative number of outlets of a chain group rose more rapidly than that of IPs, and (2) that he did not test any of those explanations. (Seguin Dep. at 211.) He further conceded that reimbursement differential "is one but not necessarily the only force" causing the reduction in IP market share. (Id. at 205-06.) Dr. Hausman faults Dr. Seguin for "simply assum[ing] causation and fail[ing] to demonstrate that the observed change in the number of [IP] outlets is the result of `discriminatory reimbursement.'" (2015 Hausman Report at 27.) He also faults Seguin's analysis for failing to demonstrate that the observed change demonstrates a harm to competition since IPs are part of a broader market of chains and mass-merchandizers for which the data show an increase in the number of outlets and the number of prescriptions dispensed. (Id. at 28.)
Finally, Dr. Hausman opines that the tests Dr. Seguin proposes to test for the presence of antitrust behavior are invalid. While Seguin proposes to test whether multi-month prescriptions are disproportionately filled by mail-order pharmacies, Hausman opines this is invalid because it would provide no evidence on whether there was an output reduction in a properly defined relevant market, and hence Dr. Seguin's test fails to test for anticompetitive behavior. (Id. at 29.) According to Hausman, Seguin's proposal to test for discrimination in reimbursement for prescriptions filled for uninsured clients is also invalid because,
(Id. at 30.) Hausman also opines that Seguin's proposal to examine whether the percentage of self-employed pharmacists has continued to decrease would also provide no evidence of anticompetitive behavior, because it would provide no evidence of reduced output in a properly defined relevant market. (Id.)
The United States Court of Appeals for the Third Circuit has joined "certain of our sister courts to hold that a plaintiff cannot rely on challenged expert testimony, when critical to class certification, to demonstrate conformity with [Federal Rule of Civil Procedure] 23 unless the plaintiff also demonstrates, and the trial court finds, that the expert testimony satisfies the standard set out in Daubert [v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993)]." In re Blood Reagents Antitrust Litig., 783 F.3d 183, 187 (3d Cir. 2015). The Court held that expert testimony "that is insufficiently reliable to satisfy the Daubert standard cannot `prove' that the Rule 23(a) prerequisites have been met `in fact,' nor can it establish `through evidentiary proof' that Rule 23(b) is satisfied." Id.
The Daubert analysis governing the admissibility of expert testimony has been codified in Federal Rule of Evidence 702, which provides as follows:
Fed. R. Evid. 702. The proponent of the expert testimony has the burden of establishing its admissibility by a preponderance of the evidence. Padillas v. Stork-Gamco, Inc., 186 F.3d 412, 418 (3d Cir. 1999) (citing Daubert, 509 U.S. at 592 n.10 (1993)); see also Mahmood v. Narciso, 549 F. App'x 99, 102 (3d Cir. 2013) (citing In re TMI Litig., 193 F.3d 613, 663 (3d Cir. 1999)).
There are three requirements for the admissibility of expert testimony pursuant to Rule 702, "`qualification, reliability and fit.'" Calhoun v. Yamaha Motor Corp., U.S.A., 350 F.3d 316, 321 (3d Cir. 2003) (quoting Schneider v. Fried, 320 F.3d 396, 405 (3d Cir. 2003)). "Qualification requires `that the witness possess specialized expertise.'" Id. (quoting Schneider, 320 F.3d at 405). The Third Circuit has `"interpreted this requirement liberally,' holding that `a broad range of knowledge, skills, and training qualify an expert as such.'" Id. (quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 741 (3d Cir.1994) ("Paoli II" ).
The "reliability" prong requires that "the expert's opinion must be based on the `methods and procedures of science' rather than on `subjective belief or unsupported speculation'; the expert must have `good grounds' for his or her belief." Id. (quoting Paoli II at 742 (quoting Daubert, 509 U.S. at 590)). An assessment of `"the reliability of scientific evidence under Rule 702 requires a determination as to its scientific validity.'" Id. (quoting Paoli II at 742). The reliability prong "applies to all aspects of an expert's testimony: the methodology, the facts underlying the expert's opinion, and the link between the facts and the conclusion." ZF Meritor, LLC v. Eaton Corp., 696 F.3d 254, 291 (3d Cir. 2012) (quotations omitted). Where the expert's "factual basis, data, principles, methods, or their application are called sufficiently into question, . . . the trial judge must determine whether the testimony has `a reliable basis in the knowledge and experience of the relevant discipline.'" Id. at 294 (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 149 (1999)).
"Fit" means that "`the expert's testimony must be relevant for the purposes of the case and must assist the trier of fact.'" Calhoun, 350 F.3d at 321 (quoting Schneider, 320 F.3d at 405). It pertains "`primarily to relevance.'" Meadows v. Anchor Longwall & Rebuild, Inc., 306 F. App'x 781, 790 (3d Cir. 2009) (quoting Lauria v. Nat'l R.R. Passenger Corp., 145 F.3d 593, 599 (3d Cir. 1998)). "The expert's testimony must `fit' under the facts of the case so that `it will aid the [fact finder] in resolving a factual dispute.'" Id. (quoting Lauria, 145 F.3d at 599). This element "`requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility.'" Id. (quoting Lauria, 145 F.3d at 600). "In other words, expert testimony based on assumptions lacking factual foundation in the record is properly excluded." Id. (citing Stecyk v. Bell Helicopter Textron, Inc., 295 F.3d 408, 414 (3d Cir. 2002)).
Caremark raises several reliability and fit arguments: (1) Plaintiffs' expert evidence fails to distinguish between legal conduct and illegal conduct, rendering it unfit; (2) the experts offer no reliable opinion about the existence of an antitrust violation; (3) the expert evidence does not track Plaintiffs' different theories of liability, rendering it unreliable under the United States Supreme Court's decision in Comcast Corp. v. Behrend, 133 S.Ct. 1426 (2013); (4) the use of national averages in the expert model cannot demonstrate antitrust impact for individual class members, rendering it unfit; and (5) the regression model is unreliable because it finds damages for class members who have suffered no damage.
Plaintiffs' expert evidence fails to pass Daubert analysis. Dr. Seguin's regression model, while asserting that there is a difference in reimbursement rates paid to IPs and chain pharmacies at the national level, does not attribute that differential solely to illegal conduct alleged in the SAC. As Dr. Hausman notes, Dr. Seguin's underlying hypothesis that "a condition of a legal world would be that . . . chains and independent pharmacies would be reimbursed at the same levels," (see Seguin Dep. at 200), fails to account for legal behavior that could result in different reimbursement rates, such as local market concentration, size efficiencies, and differences in bargaining power. Dr. Cowan has acknowledged that large pharmacies can receive better reimbursement rates than smaller pharmacies for the completely legal reason that they are in a better bargaining position. Dr. Seguin's test fails to account for the reimbursement/size relationship acknowledged by Dr. Cowan, and Seguin's test could find antitrust impact where no anticompetitive behavior has occurred. (See 2015 Hausman Report at 18-19 (faulting Seguin's failure to distinguishing monopsony power from bargaining power); ECF 260-5 (FTC Statement on In re Caremark) (finding that the exercise of monopsony power in the PBM industry was unlikely); ECF260-1 (FTC Statement on Express Scripts Acquisition of Medco Health Solutions) (same).)
Plaintiffs' experts offer no reliable opinion about the existence of an antitrust violation for the plan sponsor conspiracy claim. There are four essential elements of an antitrust claim under § 1 of the Sherman Act analyzed under the rule of reason: (1) concerted action by the defendants; (2) that produced anti-competitive effects within the relevant product and geographic markets; (3) that the concerted action was illegal; and (4) that the plaintiff was injured as a proximate result of the concerted action. Queen City Pizza, Inc. v. Domino's Pizza, Inc., 124 F.3d 430, 442 (3d Cir. 1997). "The relevant geographic market `is that area in which a potential buyer may rationally look for the goods or services he seeks.'" Fed. Trade Comm'n v. Penn State Hershey Med. Ctr., 838 F.3d 327, 338 (3d Cir. 2016) (quoting Gordon v. Lewistown Hosp., 423 F.3d 184, 212 (3d Cir. 2005); U.S. Horticultural Supply v. Scotts Co., 367 F. App'x 305, 311 (3d Cir. 2010). A market's geographic scope is "[d]etermined within the specific context of each case," and "must `correspond to the commercial realities of the industry' being considered and `be economically significant.'" Id. (quoting Brown Shoe Co. v. United States, 370 U.S. 294, 336-37 (1962) (footnote and internal quotation marks omitted in original). Plaintiffs bear the burden of establishing the relevant geographic market. Id. (citing St. Alphonsus Med. Ctr. — Nampa Inc. v. St. Luke's Health Sys., Ltd., 778 F.3d 775, 784 (9th Cir. 2015). Because establishing the geographic market looks to buyer behavior, "the evidence of the geographic market presented by the party claiming a Section 1 violation must therefore speak to buyer behavior." U.S. Horticultural Supply, 367 F. App'x at 311.
Neither Dr. Cowan nor Dr. Seguin have provided a market definition or attempted to relate their conclusions to anti-competitive effects in either a product market or a geographic market. Indeed, Seguin testified that he does not have "an expert or professional opinion on what is meant by geographic market," he has never attempted to define a boundary for a geographic market, and has never attempted to identify the competitors within a geographic market.
Plaintiffs' expert evidence also does not track their different theories of liability, rendering it unreliable under the United States Supreme Court's decision in Comcast. In that case, the Court considered the class certification of a class of more than two million current and former Comcast subscribers who sought damages for purported antitrust violations. 133 S. Ct. at 1429-30. Both the district court and the Third Circuit had determined that the putative class satisfied Rule 23(b)(3)'s predominance requirement, with the Court of Appeals holding that "[a]t the class certification stage," the proposed class did not have to "tie each theory of antitrust impact to an exact calculation of damages." Id. at 1431 (quoting Behrend v. Comcast Corp., 655 F.3d 182, 206 (3d Cir. 2011) (quotations omitted)). The Supreme Court reversed, holding that Rule 23(b)(3) had not been satisfied because the plaintiffs' model of damages fell "far short of establishing that damages are capable of measurement on a classwide basis." Id. at 1433. The majority emphasized that while damages calculations "need not be exact" at the class-certification stage, "any model supporting a `plaintiff's damages case must be consistent with its liability case, particularly with respect to the alleged anticompetitive effect of the violation.'" Id. (citations omitted).
The plaintiffs in Comcast had alleged four theories of antitrust impact, but the district court accepted only one such theory as "capable of classwide proof and rejected the rest." Id. at 1431. The damages model proposed by plaintiffs, however, failed to "isolate damages resulting from any one theory of antitrust impact." Id. The Supreme Court held that this inability to match a damages model with any one theory of liability was fatal to class certification, noting that under the Third Circuit's logic, "any method of measurement" would conceivably be "acceptable so long as it [could] be applied classwide, no matter how arbitrary the measurements." Id. at 1433. The majority held that the Comcast class was improperly certified "[i]n light of the [damages] model's inability to bridge the differences between supra-competitive prices in general and supra-competitive prices attributable to the [one antitrust impact theory found to be viable]." Id. at 1435.
Plaintiffs' expert submissions here are similarly defective. First, they assign any differential found to be the result of Defendants' alleged conduct. Second, while the SAC asserts two distinct antitrust conspiracies — a conspiracy among PBMs and a conspiracy between PBMs, plan sponsors, and chain pharmacies — Dr. Cowan and Dr. Seguin fail to propose a model capable of distinguishing the damages that are attributable to each of the two theories of liability. The teaching of Comcast is that antitrust plaintiffs must match a damages model to their theory of liability. The failure to do so here may well lead to the same paradox condemned in Comcast, i.e., a damages model unattributed to any specific theory of antitrust impact being accepted at class certification with the possibility of one or more of those theories ultimately rejected under Rule 56 or at trial.
In the context of Daubert, the failure to match the damages model to the theory of antitrust impact renders the expert opinion unfit since it cannot assist the Court at this point in the litigation in applying the Rule 23 requirements or assist the trier of fact later on to determine damages attributable to the purported antitrust violation. Indeed, Dr. Seguin testified that he did not seek to understand the nature of the conspiracies alleged by the Plaintiffs — other than "understanding this is an antitrust case" — and did not attempt to connect his findings to any particular claim or attempt to attribute what portion of damages he measured to the two theories. (ECF 261-7 at 72:23-73:11; ECF 261-9 at 132:6-15.)
Another insurmountable Daubert fit problem arises from the use of national averages in the expert model since averages cannot demonstrate antitrust impact for individual class members. Because antitrust impact is an element of Plaintiffs' causes of action,
Numerous courts have rejected the use of average price differentials to show evidence of antitrust impact that is common to the class. See e.g., Sheet Metal Workers Local 441 Health & Welfare Plan v. GlaxoSmithKline, PLC, Civ. A. No. 04-5898, 2010 WL 385552, at *30 (E.D. Pa. Sept. 30, 2010) (noting that methodology using average prices was insufficient as a common method capable of showing class-wide injury because "averaging by definition glides over what may be important differences") (quotations omitted) (citing ABA Section of Antitrust Law, Econometrics: Legal, Practical, and Technical Issues 220 (2005) ("Using averages can lead to serious analytical problems. For example, averages can hide substantial variation across individual cases, which may be key to determining whether there is common impact."); Gates v. Rohm & Haas Co., 655 F.3d 255, 266 (3d Cir. 2011) (averages evidence "is not `common' because it is not shared by all (possibly even most) individuals in the class. Averages or community-wide estimations would not be probative of any individual's claim because any one class member may have an exposure level well above or below the average" and noting that "[a]ttempts to meet the burden of proof using modeling and assumptions that do not reflect the individual characteristics of class members have been met with skepticism.") (citations omitted); In re Optical Disk Drive Antitrust Litig., 303 F.R.D. 311, 321 (N.D. Cal. 2014) (statistical model in which the alleged conspiratorial overcharge was assumed to be the same for all purchasers and throughout the entire class period "cannot serve to establish that all (or nearly all) members of the class suffered damage as a result of defendants' alleged anti-competitive conduct [because the] regression analysis . . . assumes the very proposition that the [Plaintiffs] are now offering it, in part, to show."); Reed v. Advocate Health Care, 268 F.R.D. 573, 591 (N.D. Ill. 2009) ("Measuring average base wage suppression does not indicate whether each putative class member suffered harm from the alleged conspiracy. In other words, it is not a methodology common to the class that can determine impact with respect to each class member."); In re Flash Memory Antitrust Litig., Civ. A. No. 07-0086, 2010 WL 2332081, at *10 (N.D. Cal. June 9, 2010) ("By looking only at an average price trend, [plaintiff's expert's] model obscures individual variations over time among the prices that different customers pay for the same or different products that appear in the data.")
As the Hon. Lawrence F. Stengel noted in Sheet Metal Workers, [j]ust because an average price was increased or decreased by the alleged [anticompetitive activity] does not mean that all members of the proposed class paid supra-competitive prices or that any damage for an individual end-payor could be calculated in a formulaic way by common proof." Id., 2010 WL 385552, at *30. But see, In re Processed Egg Prod. Antitrust Litig., 312 F.R.D. 171, 199 (E.D. Pa. 2015) (stating that "even though the use of a single average overcharge to demonstrate the impact of a conspiracy across the class can be problematic, Plaintiffs have laid a sufficient foundation for the inferential finding that the impact reflected in the single average overcharge was shared by virtually every class member."). Here, Plaintiffs' experts provide no basis on which one can conclude that the average reimbursement differential they find between an IP and a chain pharmacy is shared by virtually every class member. For this reason, the experts' use of the average as a method for establishing antitrust impact fails the fit test because there is no valid scientific connection between the averages evidence and the pertinent inquiry of whether the differential was the result of anticompetitive behavior.
Finally, the regression model is unreliable because it finds damages for class members who have suffered no damage. Dr. Seguin testified in his deposition that his model's use of average reimbursements would result in finding damages for an IP that received the same reimbursement as a chain, as well as for an IP that received a larger reimbursement than a chain. (See ECF 261-8 at 54-57.
Accordingly, the Plaintiffs' expert submissions fail to pass Daubert scrutiny. Caremark's Motion to exclude Plaintiffs' expert testimony is granted. Next, the Court considers the class certification issues.
To prosecute their two conspiracy claims under Section 1 of the Sherman Act, Plaintiffs seek to certify the following Class pursuant to Rule 23(b)(2) and (b)(3) of the Federal Rules of Civil Procedure:
(ECF 248 at 1.) They assert that the requirements for a (b)(2) injunctive relief class are met because Caremark has "acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole," Fed. R. Civ. P. 23(b)(2), and Caremark acted on grounds generally applicable to the class making certification of Plaintiffs' claim for injunctive relief is appropriate. They assert that the requirements for a (b)(3) damages class are met because common questions of law and fact predominate over any potential individualized questions and because a class action is superior to any other means of adjudication. (ECF 248 at 2.) They assert that they can establish common impact and damages utilizing class-wide proof through the expert submissions of Dr. Cowan and Dr. Seguin. Even though Plaintiffs' expert submissions fail the Daubert analysis, the court will nevertheless provide its analysis of all Rule 23 issues.
The United States Court of Appeals for the Third Circuit requires rigorous assessment of the available evidence to assure the prerequisites of Rule 23 are met and to "resolve factual disputes by a preponderance of the evidence and make findings that each Rule 23 requirement is met or is not met, having considered all relevant evidence and arguments presented by the parties." In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 320. A plaintiff "must be prepared to prove that there are in fact sufficiently numerous parties, common questions of law or fact, etc." Wal-mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011) (emphasis omitted). "Failure to meet any of Rule 23(a) or 23(b)'s requirements precludes certification." Danvers Motor Co., Inc. v. Ford Motor Co., 543 F.3d 141, 147 (3d Cir. 2008). It is the plaintiff's burden to prove by a preponderance of the evidence each of the prerequisites under Rule 23(a), and that the class fits within the desired categories of class actions set forth in Rule 23(b). In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 307, 316 n. 14 (citation omitted); Carrera v. Bayer Corp., 727 F.3d 300, 306 (3d Cir. 2013) (stating a plaintiff must show class action prerequisites by a preponderance of the evidence); see Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 354 (3d Cir. 2013) ("It is plaintiff's burden to show that a class action is a proper vehicle for this lawsuit"). Rigorous analysis will frequently "entail some overlap with the merits of the plaintiff's underlying claim. That cannot be helped. `[T]he class determination generally involves considerations that are enmeshed in the factual and legal issues comprising the plaintiff's cause of action.'" Dukes, 564 U.S. at 351 (alteration in original) (quoting General Tel. Co. of SW v. Falcon, 457 U.S. 147, 160 (1982)).
Rule 23(a) requires that Plaintiffs meet four elements for class certification: (1) numerosity; (2) commonality; (3) typicality; and (4) adequacy of representation. If the requirements of Rule 23(a) are met, Plaintiffs seeking to certify a damages class must satisfy additional requirements of predominance and superiority required by Rule 23(b)(3).
Under Rule 23(a), a plaintiff bears the burden of establishing numerosity by a preponderance of the evidence. Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 594-95 (3d Cr. 2012). Plaintiff must prove that the putative class is "so numerous that joinder of all members is impracticable." Fed. R. Civ. P. 23(a)(1). "No minimum number of plaintiffs is required to maintain a suit as a class action, but generally if the named plaintiff demonstrates that the potential number of plaintiffs exceeds 40, the first prong of Rule 23(a) has been met." Stewart v. Abraham, 275 F.3d 220, 226-27 (3d Cir. 2001) (citing 5 James Wm. Moore et al., Moore's Federal Practice § 23.22[3][a] (Matthew Bender 3d ed. 1999)). We cannot "assume," "speculate," or defer to "common sense" with respect to how many class members exist. Marcus, 687 F.3d at 595-97. The plaintiff must produce evidence, direct or circumstantial, specific to the products, problems, parties, and geographic areas actually covered by the proposed class definitions to allow a court to make a factual finding on this requirement. Id. at 596.
Plaintiffs assert there are approximately 25,000 IPs throughout the United States that dispense approximately 1.3 billion prescriptions annually amounting to $60 billion dollars in prescription sales. (SAC ¶ 2.) Caremark's data establishes that in 2005 there were 11,741 IPs with data in Caremark's dataset and in 2008 there were 11,710. (Seguin Rebuttal at 2 n.1.) Caremark makes no specific argument on the numerosity requirement, and Plaintiffs have satisfied the requirement.
"A putative class satisfies Rule 23(a)'s commonality requirement if `the named plaintiffs share at least one question of fact or law with the grievances of the prospective class.'" Reyes v. Netdeposit, LLC, 802 F.3d 469, 486 (3d Cir. 2015) (citing Rodriguez v. Nat'l City Bank, 726 F.3d 372, 382 (3d Cir. 2013)). "Commonality does not require perfect identity of questions of law or fact among all class members. Rather, even a single common question will do." Id., 802 F.3d at 486 (citing Dukes, 564 U.S. 338). The commonality inquiry turns on whether determining the truth or falsity of a common contention will resolve an issue that is central to the validity of each one of the claims in one stroke. Id. at 487. "What matters to class certification . . . is not the raising of common questions — even in droves — but, rather the capacity of a classwide proceeding to generate common answers apt to drive the resolution of the litigation." Dukes, 564 U.S. 338, 131 S.Ct. 2541, 2551 (emphasis and ellipsis in the original). The bar for establishing commonality is "not high" and is "easily met." In re Cmty. Bank of N. Va. Mortg. Lending Practices Litig., 795 F.3d 380, 397 (3d Cir. 2015); Reyes, 802 F.3d at 486 (citing Baby Neal v. Casey, 43 F.3d 48, 56 (3d Cir. 1994)).
Plaintiffs argue that, in addition to the "overarching common question of whether Caremark entered into the conspiracies alleged in the complaint," the following questions of law and fact are common to the Class:
(Pl. Mem. at 7.)
Caremark responds that, although Plaintiffs list these six issues they assert are common, they have not met their burden to show that they are common to all class members. It argues that because Issues (a), (d), and (e) all relate to whether any conspiracy fixed prices below what would prevail in a competitive market and whether class members suffered antitrust injury and damages, and because the factors that affect reimbursement rates in both the real world and the "but-for" world described by Plaintiffs' experts differ from one putative class member to the next, there are no common questions across the proposed class. With respect to Issues (b) and (c), addressing whether Caremark diverted health plan members to CVS and whether Caremark required some pharmacies to dispense and fill drugs as a formulary, Caremark challenges whether they can be "common" questions because they are unrelated to the Sherman Act violations Plaintiffs allege, and Plaintiffs do not describe how any of the actions caused them antitrust injury. Caremark offers no argument to rebut whether Plaintiffs' "overarching" common issue of the existence of the alleged conspiracies is common.
Because the commonality element does not require perfect identity of questions of law or fact among all class members, Caremark's arguments are rejected. Common questions clearly exist, specifically whether Caremark engaged in the antitrust conspiracies detailed in the SAC. Answers to the questions asserted by the Plaintiffs as common will drive the resolution of the litigation.
The typicality requirement aids a court in determining whether maintenance of a class action is economical and whether the named plaintiff's claim and the class claims are so interrelated that the interests of the class members will be fairly and adequately protected in their absence. Marcus, 687 F.3d at 597-98 (citing General Telephone Co. of the Sw., 457 U.S. at 158 n. 13). Typicality "screen [s] out class actions in which the legal or factual position of the representatives is markedly different from that of other members of the class even though common issues of law or fact are present." Id. at 598. To determine whether a plaintiff's position is markedly different from the class as a whole, courts compare three distinct, though related, concerns: (1) the claims of the class representative must be generally the same as those of the class in terms of both (a) the legal theory advanced and (b) the factual circumstances underlying that theory; (2) the class representative must not be subject to a defense that is both inapplicable to many members of the class and likely to become a major focus of the litigation; and (3) the interests and incentives of the representative must be sufficiently aligned with those of the class. Marcus, 687 F.3d at 599.
The named Plaintiffs in Caremark assert that their claims arise from the same course of conduct that gives rise to the claims of absent class members, and their claims are based on the same legal theory. They contend they will advance the interests of absent class member because they must show that Caremark engaged in an unlawful combination or conspiracy to fix the prices that all participating independent pharmacies are paid for dispensing services and/or prescription drugs. They also contend that they possess no antagonistic interests or conflicts with the absent class members.
Caremark responds the typicality requirement is absent because the named Plaintiffs' economic experience is directly contrary to the injury that they allege on behalf of their proposed class. Specifically, it asserts that the relevant markets in this case are highly diverse local retail pharmacy markets, meaning that the named Plaintiffs' proof of antitrust impact and the resulting anticompetitive effects will differ significantly from the proof for the hundreds of other local pharmacy markets across the country. If the named Plaintiffs were able to prove that lower reimbursement had caused anticompetitive effects of higher prices and reduced output in Jackson County, Alabama, Caremark argues that it would prove nothing about whether prices were higher and output was lower in other markets (such as urban areas) with characteristics different from those of Jackson County. It notes that Dr. Hausman has provided evidence that IPs in Jackson County received $1,490 more than they would have received had they been compensated at the rate received by competing local chain outlets. (See Hausman Report at 16-17 and Table 1.) It notes too that Dr. Cowan has conceded that dispensing fees received by the named Plaintiffs were in line with chain outlets in their markets, and they therefore suffered no antitrust injury. (See ECF 261-12 (May 4, 2006 Dep. of Charles D. Cowan, Ph.D.) at 481:24-485:15, 487:2-13.
The named Plaintiffs refute Caremark's assertion that they are atypical of the Class because they received reimbursements above those received by local chain outlets, asserting that this "conclusion is premised on the erroneous notion that an Independent Pharmacy must have been reimbursed at a rate below that of local chains on each and every drug in order to have been injured or sustained damages, while in actuality the injury and damages to Independent Pharmacies are measured against the reimbursements they would have received absent the conspiracy." (ECF 269 at 3.)
The named Plaintiffs in Caremark have failed to meet their burden to show that their claims are typical of the class they seek to represent. They produced no evidence that they received reimbursements that were lower than the chain competitors in their markets while Caremark produced evidence that they received reimbursements equal to or greater than their chain competitors. Given this class certification record, the named Plaintiffs are clearly subject to a defense that is both inapplicable to many members of the class and likely to become a major focus of the litigation, namely whether they have suffered an antitrust injury resulting from the conspiracies they allege. An antitrust injury is an "injury of `the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful. The injury should reflect the anti-competitive effect either of the violation or of anti-competitive acts made possible by the violation." Eichorn v. AT & T Corp., 248 F.3d 131,140 (3d Cir. 2001) (quoting Brunswick Corp., 429 U.S. at 489). Because the record shows the named Plaintiffs failed to put on evidence that they received lower reimbursements, they cannot show they suffered the same effect of the anti-competitive conspiracies they allege. Thus, they fail the typicality requirement.
The fourth requirement in Rule 23(a) is that the representative plaintiffs must "fairly and adequately protect the interests of the class." Fed. R. Civ. P. 23(a)(4). Adequacy concerns both "the experience and performance of class counsel" and "the interests and incentives of the representative plaintiffs." Dewey v. Volkswagen Aktiengesellschaft, 681 F.3d 170, 181 (3d Cir. 2012) (citing In re Cmty. Bank of N. Va., 418 F.3d 277, 303 (3d Cir. 2005). "The principal purpose of the adequacy requirement is to determine whether the named plaintiffs have the ability and the incentive to vigorously represent the claims of the class." Community Bank III, 795 F.3d at 393 (quoting In re Cmty. Bank of N. Va., 622 F.3d 275, 291 (3d Cir. 2010) (Community Bank II)). In fact, "`the linchpin of the adequacy requirement is the alignment of interests and incentives between the representative plaintiffs and the rest of the class.'" Community Bank III, 795 F.3d at 393 (quoting Dewey, 681 F.3d at 183). This inquiry is closely tethered to the typicality inquiry, see Danvers Motor Co., Inc., 543 F.3d at 149, and ensures that the named plaintiff's claims "are not antagonistic to the class." Id. at 150 (citing Beck v. Maximus, Inc., 457 F.3d 291, 296 (3d Cir. 2006)).
While the named Plaintiffs in Caremark assert that they are not aware of any conflict that would prevent them from serving in the capacity as named representatives and contend that their interests in successful prosecution of the claims asserted in the SAC are aligned with the interest of absent class members since the nature of antitrust claims and the injury that flows from an unlawful conspiracy and/or combination to restrain trade are identical, they do not satisfy the adequacy requirement for the same reason that they fail to satisfy the typicality requirement. As they have not demonstrated that they have individually suffered the type of antitrust injury that they allege on behalf of the putative class, their interests diverge from other members of the proposed class who allegedly have received lower reimbursements as a result of anticompetitive activity.
For a class to be certified under Rule 23(b)(3), it must also be "ascertainable." See Carrera, 727 F.3d at 306 ("Class ascertainability is `an essential prerequisite of a class action, at least with respect to actions under Rule 23(b)(3).'") (quoting Marcus, 687 F.3d at 592-93); see also Byrd v. Aaron's Inc., 784 F.3d 154, 162 (3d Cir. 2015), as amended (Apr. 28, 2015) (stating "the ascertainability requirement as to a Rule 23(b)(3) class is grounded in the nature of the class-action device itself").
The ascertainability element "functions as a necessary prerequisite (or implicit requirement) because it allows a trial court effectively to evaluate the explicit requirements of Rule 23." Byrd at 162. It is an independent inquiry, in addition to the Rule 23 requirements, that "ensures that a proposed class will actually function as a class." Id. To satisfy the ascertainability prerequisite, Plaintiffs must show by a preponderance of the evidence that the class is "currently and readily ascertainable based on objective criteria," Marcus, 687 F.3d at 593, and the court "must undertake a rigorous analysis of the evidence to determine if the standard is met." Carrera, 727 F.3d at 306. "[A]scertainability and a clear class definition allow potential class members to identify themselves for purposes of opting out of a class. Second, it ensures that a defendant's rights are protected by the class action mechanism. Third, it ensures that the parties can identify class members in a manner consistent with the efficiencies of a class action." Id. Accordingly, courts must "ensure that class members can be identified `without extensive and individualized fact-finding or "mini-trials."'" Id. (quoting Marcus). "[T]o satisfy ascertainability as it relates to proof of class membership, the plaintiff must demonstrate that his purported method for ascertaining class members is reliable, administratively feasible, and permits a defendant to challenge the evidence used to prove class membership." Id. The Third Circuit has also recently reiterated that "a party cannot merely provide assurances to the district court that it will later meet Rule 23's requirements. . . . Nor may a party `merely propose a method of ascertaining a class without any evidentiary support that the method will be successful.'" Byrd, 784 F.3d at 164 (quoting Carrera 727 F.3d at 306, 307, 311) (internal citation omitted).
While a Rule 23(b)(2) class need not meet the ascertainability requirements that apply to Rule 23(b)(3) classes, see Shelton v. Bledsoe, 775 F.3d 554, 563 (3d Cir. 2015) ("[A]scertainability is not a requirement for certification of a (b)(2) class seeking only injunctive and declaratory relief, such as the putative class here."), such a class must be sufficiently cohesive. Barnes v. Am. Tobacco Co., 161 F.3d 127, 143 (3d. Cir. 1998) ("While 23(b)(2) class actions have no predominance or superiority requirements, it is well established that the class claims must be cohesive."). An injunctive relief class must also be properly defined. "A properly defined `class' is one that: (1) meets the requirements of Rule 23(a); (2) is sufficiently cohesive under Rule 23(b)(2) and [the Third Circuit's] guidance in Barnes, 161 F.3d at 143; and (3) is capable of the type of description by a `readily discernible, clear, and precise statement of the parameters defining the class,' as required by Rule 23(c)(1)(B) and [the Third Circuit's] discussion in Wachtel [v. Guardian Life Ins. Co. of Am], 453 F.3d [179] at 187 [(3d Cir. 2006)]." Shelton, 775 F.3d at 563.
The cohesiveness requirement protects two interests. The first interest is in protecting unnamed class members, who "are bound by the action without the opportunity to withdraw and may be prejudiced by a negative judgment in the class action." Barnes, 161 F.3d at 143. The cohesiveness requirement protects this interest by ensuring that "significant individual issues do not pervade the entire action because it would be unjust to bind absent class members to a negative decision where the class representatives' claims present different individual issues than the claims of the absent members present." Id. (citations and quotation marks omitted). The second interest is in ensuring that the litigation remains manageable. If a class is not sufficiently cohesive, "the suit could become unmanageable and little value would be gained in proceeding as a class action if significant individual issues were to arise consistently." Id. (quotation marks, citations, and alterations omitted).
To satisfy the cohesiveness test, Plaintiffs must show that the "class's claims are common ones and that adjudication of the case will not devolve into consideration of myriad individual issues." Newberg on Class Actions § 4:34. "In other words, Rule 23(b)(2) applies only when a single injunction or declaratory judgment would provide relief to each member of the class. It does not authorize class certification when each individual class member would be entitled to a different injunction or declaratory judgment against the defendant." Dukes, 564 U.S. at 360 (emphasis in original). The Third Circuit has held that any `"disparate factual circumstances of class members' may prevent a class from being cohesive." Gates, 655 F.3d at 264 (citing Carter v. Butz, 479 F.2d 1084, 1089 (3d Cir. 1973). Courts have the discretion to deny certification in the presence of disparate factual circumstances. Geraghty v. U.S. Parole Comm'n, 719 F.2d 1199, 1205 (3d Cir. 1983). "The key to the (b)(2) class is `the indivisible nature of the injunctive or declaratory remedy warranted — the notion that the conduct is such that it can be enjoined or declared unlawful only as to all of the class members or as to none of them.'" Dukes, 564 U.S. at 360 (quoting Nagareda, The Preexisting Principle and the Structure of the Class Action, 103 Colum. L. Rev. 149, 176 n. 110 (2003)).
Plaintiffs in Caremark contend that the proposed Rule 23(b)(3) class can be readily ascertained because Caremark's data contains the name of the pharmacy that filled each prescription. Caremark makes no specific argument on the ascertainability requirement, and the Plaintiffs' evidence satisfies the ascertainability requirement. Neither party has specifically addressed the cohesiveness requirement to certify a Rule 23(b)(2) injunctive relief class. Nonetheless, the lack of cohesion is clear upon examination of the same reasons, writ larger, that the named Plaintiffs fail the typicality test, as well as those discussed earlier in the Daubert context. The record demonstrates that the market for retail pharmacy services is highly diverse and local. Proof by use of averages that some members suffered antitrust injury from the anticompetitive effects of the alleged conspiracy does not constitute common evidence that they all did, that they all suffered the same antitrust injury, or that the same injunctive relief is appropriate. Accordingly, the putative class members' disparate factual circumstances prevent the proposed national injunctive class from being sufficiently cohesive to ensure that injunctive relief is manageable.
A class action can be certified under Rule 23(b)(3) if (1) the questions of law or fact common to class members predominate over any questions affecting only individual members, and (2) a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. Fed. R. Civ. P. 23(b)(3). The Rule provides that the following matters are pertinent to these findings: (1) the class members' interests in individually controlling the prosecution or defense of separate actions; (2) the extent and nature of any litigation concerning the controversy already begun by or against class members; (3) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (4) the likely difficulties in managing a class action. Id.
"Considering whether `questions of law or fact common to class members predominate' begins, of course, with the elements of the underlying cause of action." Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809 (2011); see also In re Modafinil Antitrust Litig., 837 F.3d 238, 260 (3d Cir. 2016) (stating that the predominance inquiry "is especially dependent upon the merits of a plaintiff's claim, since the nature of the evidence that will suffice to resolve a question determines whether the question is common or individual.") (quoting In re Constar Int'l Inc. Sec. Litig., 585 F.3d 774, 780 (3d Cir. 2009) (internal quotation marks omitted in In re Modafinil).) We "must examine each element of a legal claim `through the prism' of Rule 23(b)(3)." Marcus, 687 F.3d at 600 (citing In re DVI, Inc. Sec. Litig., 639 F.3d 623, 630 (3d Cir. 2011)). To obtain class certification, "[a] plaintiff must demonstrate that the element of the [legal claim] is capable of proof at trial through evidence that is common to the class rather than individual to its members." Marcus, 687 F.3d at 600 (citation omitted). If proof of an element of the legal claim requires individual treatment, then class certification is unsuitable. See Tyson Foods, Inc. v. Bouaphakeo, 136 S.Ct. 1036, 1045 (2016) (stating that "[a]n individual question is one where members of a proposed class will need to present evidence that varies from member to member, while a common question is one where the same evidence will suffice for each member to make a prima facie showing [or] the issue is susceptible to generalized, class-wide proof.") (internal quotation marks omitted) (citing Newberg on Class Actions § 4:50, pp. 169-197); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d at 311-312 (stating that "the task for plaintiffs at class certification is to demonstrate that the element [] is capable of proof at trial through evidence that is common to the class rather than individual to its members. Deciding this issue calls for the district court's rigorous assessment of the available evidence and the method or methods by which plaintiffs propose to use the evidence to prove [the elements] at trial."); see also Fed. R. Civ. P. 23 advisory committee's note to 2003 amendment ("A critical need is to determine how the case will be tried.").
Horizontal price-fixing is a per se violation of § 1 of the Sherman Antitrust Act. See, e.g., Catalano, Inc. v. Target Sales, Inc., 446 U.S. 643, 647 (1980); United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 218 (1940). To prevail on their per se inter-PBM price-fixing claim under § 1 of the Sherman Act, Plaintiffs must prove three elements: (1) a conspiracy to fix prices in violation of the antitrust laws; (2) the antitrust impact of the unlawful activity; and (3) damages sustained as a result of the unlawful activity. See, e.g., In re Linerboard Antitrust Litig., 203 F.R.D. 197, 214 (3d. Cir. 2001); (citing In re Plastic Cutlery Antitrust Litig., Civ. A. No. 96-728, 1998 WL 135703, *5 (E.D. Pa. Mar. 20, 1998) ("In a price-fixing antitrust class action, plaintiffs must establish that both the defendants' violations of law and the impact of those violations on the class members involve predominantly common issues. Plaintiffs must therefore make a threshold showing that the element of impact will predominantly involve generalized issues of proof, rather than questions which are particular to each member of the plaintiff class." (internal citations omitted)). As previously discussed, Plaintiffs also must show "that the damages resulting from that injury were measurable on a class-wide basis through use of a common methodology." Comcast Corp., 133 S. Ct. at 1430 (quotation marks omitted). A model purporting to serve as evidence of damages in a class action must measure only those damages attributable to the theory upon which liability is premised. Id. Where the damages evidence does not translate the relevant "`legal theory of the harmful event into an analysis of the economic impact of that event,'" the Comcast Court determined that common questions could not predominate over individual ones. Id. at 1435 (quoting Federal Judicial Center, Reference Manual on Scientific Evidence 432 (3d ed. 2011)).
Plaintiffs' argument that common questions can be proven through evidence that is common to the class rather than individual to its members (see ECF 248 at 15) must be rejected for similar reasons undergirding the Daubert analysis.
Because "[t]he essence of any violation of § 1 is the illegal agreement itself," it is not necessary for a plaintiff to prove that "overt acts [were] performed in furtherance of it." Summit Health, Ltd. v. Pinhas, 500 U.S. 322, 330 (1991) (citing United States v. Kissel, 218 U.S. 601 (1910)); see also, Socony-Vacuum Oil Co., 310 U.S. at 224, n. 59 ("[C]onspiracies under the Sherman Act are not dependent on any overt act other than the act of conspiring."). Nor is it necessary to prove that the conspiracy was actually successful in raising prices. Id. ("It is the contract, combination . . . or conspiracy, in restraint of trade or commerce which § 1 of the Act strikes down, whether the concerted activity be wholly nascent or abortive on the one hand, or successful on the other.") (quotation marks omitted).
Accordingly, to meet the predominance requirement, Plaintiffs seeking class certification of a per se antitrust claim must show by common evidence that a conspiracy existed, which may be done through either direct evidence, such as an express agreement, or circumstantial evidence, such as parallel conduct and course of dealing. See, e.g., In re EPDM, 681 F. Supp. 2d at 166 ("To prove the existence of an express, manifested agreement, the antitrust plaintiff should present `direct or circumstantial evidence that reasonably tends to prove that the manufacturer and others had a conscious commitment to a common scheme designed to achieve an unlawful objective.'") (quoting Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 764 (1984)); Apex Oil Co. v. DiMauro, 822 F.2d 246, 252 (2d Cir.1987) ("[A]t a minimum," a jury must be able to conclude that "the conspirators had a unity of purpose or a common design and understanding, or a meeting of minds in an unlawful arrangement" (quotation omitted)); United States v. Paramount Pictures, 334 U.S. 131, 142 (1948); Am. Tobacco Co. v. United States, 328 U.S. 781, 800 (1946). In the class action context, pre-Hydrogen Peroxide cases generally accepted that "allegations of the existence of a price-fixing conspiracy are susceptible to common proof and, if proven true, would satisfy the first element of . . . [an] antitrust cause of action." Cordes &Co. Fin. Servs., Inc. v. A.G. Edwards & Sons, Inc., 502 F.3d 91, 105 (2d Cir. 2007). In the post-Hydrogen Peroxide world, the "if proven true" part of the formulation takes on new imperative since courts may no longer merely assume that allegations can be proven later at trial.
The allegations here against the PBMs include fixing and artificially depressing the prices paid to IPs, conspiring to use their combined monopolistic market power to force unconscionable reimbursement rates, acting as a conduit for the Client Payors to engage in horizontal restraint of trade by removing IP competitors, using market power to stabilize and repress the fees IPs would be able to charge in a free and open market, and diverting health plan members to their mail order businesses. (SAC ¶ 5.) Dr. Cowan opines that Caremark's actions "have the effect of price-fixing, eliminating competition, indirect collusion among the PBMs, and an attempt to drive independent pharmacies out of the market." (Cowan Report at 24.) He then goes on to offer criteria by which it may be determined whether IPs "are being treated differentially after other factors, such as size, are accounted for." (Id. at 27.) He opines that if the average of "dispensing fee contract values falls below the bounds established in determining the size to fee relationship," this would indicate that prices have been fixed and fees paid to IPs are artificially low. (Id. at 34.) He proposes that, if there is no natural variability in the contract values between IPs and PBMs — in other words all contracts with the PBMs vary much less than they do for the PBMs' contracts with pharmacies with six or more stores — then there is also an indication of price collusion. (Id. at 35.) Dr. Cowan has not, however, actually studied these criteria in order to provide common evidence of price fixing. (Id. at 37.) Dr. Seguin also did not attempt to provide common evidence of price fixing. He assumed price fixing had occurred and endeavored only to calculate the resulting damages. Thus, there has been no actual evidence, common or otherwise, offered on the basic issue of whether a price fixing conspiracy existed.
The Daubert analysis fully sets forth the reasons why Plaintiffs fail to show that their rule of reason violation is capable of being proven by evidence that is common to the class rather than individual to its members. They have offered no evidence to establish the geographic market or Defendants' market power. Not only are their experts' conclusions on antitrust impact unreliable, there is also a failure to demonstrate that the claim is capable of proof through common evidence.
Plaintiffs' proposed damages model fails the Comcast test, and thus cannot serve as common evidence of predominance, because the damages model does not matchup with Plaintiffs' theories of liability. Having already discussed this problem at length in the Daubert analysis, it serves merely to reiterate that Dr. Seguin made no effort to determine which portion of the class's purported damages result from which alleged conspiracy (see Seguin Sept. 11, 2015 Dep. at 132:6-15), and he was not even aware of the nature of the conspiracies alleged by Plaintiffs when he conducted his study. (Seguin Mar. 19 2015 Dep. at 72:19-73:15.)
Importantly, Dr. Seguin failed to isolate the difference in reimbursement rates attributable to an alleged antitrust conspiracy from any difference attributable to legitimate bargaining power or other market factors. Merely assuming that any difference observed has to be the result of Plaintiffs' allegations of antitrust behavior is insufficient, and Dr. Seguin admitted that could not say that the difference he observed was the result of any conspiracy alleged in the SAC. Rather, he did not consider any potential causes other than illegal activity. (See Seguin Mar. 19, 2015 Dep. at 71:18-72:1, 75:21-76:23.) This failure to isolate differentials solely due to anti-competitive activities is fatal since Dr. Cowan recognized that reimbursement rates between IPs and chains could differ in the absence of anticompetitive behavior. (See Cowan Sept. 11, 2015 Dep. at 40:1-4 ("Q: So in the but-for world, you would still expect the PBM to have different reimbursement rates to different pharmacies; correct? A: Within reason, yes."); see also id. at 98:15-25 ("I'm allowing for variation [in reimbursement rates between pharmacies.]").) Both experts also recognized that legitimate bargaining power can explain the differential. (See Cowan Report at 42; Seguin Sept. 11, 2015 Dep. at 72:8-73:11.) While Dr. Cowan warned that, in examining the variability in reimbursements, one must examine if there is natural variability in the contract values between IPs and PBMs, as well as control for market concentration and other factors (see Cowan Report at 34, 39), Dr. Seguin did not do this. Accordingly, Plaintiffs' damages model cannot serve as common evidence.
Dr. Seguin's reliance on averages fails to comply with the Hydrogen Peroxide imperative that Plaintiffs show that every member of the putative class suffered an antitrust injury. His conclusion that there is a nationwide average differential of approximately $1.00 between the reimbursement rate paid to IPs and chains is then used in his regression model that considers only one explanatory variable, namely whether the pharmacy is an IP or a chain. As Dr. Hausman explains, this parsimonious model fails to account for market forces that can explain the differential independent of the conspiracies alleged by the Plaintiffs, including local market density and local costs of doing business. Again, having already discussed this problem at length in the Daubert analysis, the Rule 23(b)(3) predominance problem stemming from the Seguin model's use of averages is the failure to discount other causes of the observed average reimbursement differential in order show common evidence that all putative members were impacted by the anticompetitive conduct. See Areeda, Hovenkamp, Blair, Durrance, Antitrust Law ¶ 398 (4th ed. 2014) (stating that to satisfy predominance, "even in a per se case . . . it is necessary to prove that all class members suffered injury to their business or property using common proof." (emphasis added)). Evidence that IPs received on average $1.00 less than a chain pharmacy is not evidence that every IP received less and was thus impacted. Combined with Dr. Hausman's unrebutted evidence that 52.6% of IPs received a higher reimbursement than the 50th percentile chain pharmacy, and that the named Plaintiffs themselves received above average reimbursements compared to the chain pharmacies they directly competed against, Plaintiffs have failed to show antitrust impact by evidence that is common to the class through Dr. Seguin's parsimonious model.
Because Dr. Seguin assumes a national market for retail pharmacy sales, his model fails to account for the effect of local market forces on reimbursement rates differentials. Dr. Hausman's test of Seguin's model found there was a less than one percent probability to Dr. Seguin's assumption that reimbursement rate differentials was consistent nationwide. While Seguin opines that he has undertaken to judge the robustness of his model by incorporating three local variables, it must be noted that two of those variables — the number of pharmacies within a zip and the median income within a zip — reduced his $1.00/prescription fill differential estimate to $0.54. (See ECF 261-7 at 162:12-23.) The evidence of local variation and lack of robustness supporting a national geographic market is particularly problematic for considering whether common evidence predominates for Plaintiffs' Plan Sponsor conspiracy claim that is subject to rule of reason analysis. It is counterintuitive to think of the geographic market for retail pharmacy sales as national. Dr. Hausman aptly opines that a consumer in Alabama would not consider a pharmacy in Wisconsin to be an option to fill a prescription.
Dr. Cowan's alternative method of using data on the declining market share for IPs and Dr. Seguin's data showing different CAGRs for chains and IPs must also be rejected as common evidence of antitrust impact. Dr. Seguin's data on market share showing that the chain group had an increase in the relative number of outlets of 19.2% over the period from 2002 to 2012, or a CAGR of 1.8%, versus data indicating that the market share of IPs fell with a CAGR of -0.15%, does not attempt to relate the perceived differential to the anticompetitive conduct alleged in the SAC. While Seguin opines that the decline in the number of IP outlets over this period "strongly suggests" common harm to IPs "due to discriminatory reimbursement by PBMs," (Seguin Report at 11), he conceded at his deposition that there were "numerous potential explanations" why the relative number of chain outlets rose compared to IPs, and that he had not tested any of those explanations in order to discount them. (ECF 261-7 at 211.) Because, as already noted, antitrust laws protect competition and not competitors, whether certain competitors' market shares have declined is not alone common evidence of an antitrust violation.
The superiority requirement `"asks the court to balance, in terms of fairness and efficiency, the merits of a class action against those of alternative available methods of adjudication.'" In re Processed Egg Prods. Antitrust Litig., 284 F.R.D. 278, 293-94 (E.D. Pa. 2012) (quoting In re Prudential Ins. Co. Am. Sales Practice Litig. Agent Actions, 148 F.3d 283, 316 (3d Cir. 1998). Courts use the four factors listed in Rule 23(b)(3)-(1) the class members' interests in individually controlling the prosecution or defense of separate actions; (2) the extent and nature of any litigation concerning the controversy already begun by or against class members; (3) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (4) the likely difficulties in managing a class action — to make the determination. While Plaintiffs assert that no putative member of the class has expressed an interest in individually controlling a separate action and that managing a class action will not be difficult, the prior discussion of the typicality and predominance elements makes it clear that class treatment is not a superior method of adjudicating Plaintiffs' antitrust conspiracy claims. The different way that the alleged anticompetitive activities may have impacted individual IPs' reimbursement rates makes collective management of claims particularly difficult.
On March 3, 2006, the transferee court in the Northern District of Alabama certified a class of all independent pharmacies in the United States who contracted with any of the named Defendants "to dispense and sell prescription drugs to members of a defendant network."
Defendants argue that the class created by the Alabama certification be decertified for four reasons: (1) to reconcile the case analytically with the other pending cases since the class definitions overlap; (2) the Alabama certification order fails to comply with Third Circuit law and the specificity required by the amendments to Federal Rule of Civil Procedure 23(c)(1)(B); (3) the order was not grounded upon specific, market-based evidence, which Third Circuit law requires in order to demonstrate that a plaintiff can use common proof to show the impact of the alleged antitrust violation; and (4) the Plaintiffs cannot now meet their burden to show that common questions predominate over questions affecting only individual class members. Plaintiffs respond that (1) the principles of comity require that the Alabama certification order be respected, and (2) Defendants' request is contrary to the recent trend by MDL transferee courts to send class certification issues back to the transferor courts. Because the Alabama certification order has been rendered deficient by intervening changes in the law, the class must be decertified.
Shortly after the cases comprising the MDL were transferred to the United States District Court for the Eastern District of Pennsylvania, the Hon. John P. Fullam concluded that the Alabama certification order "does not comply with the requirements of Fed. R. Civ. P. 23(c)(1)(B), since the order does not define the class and the class claims, issues, or defenses." (Order of Feb. 21, 2007, ECF 42, at 1.) Judge Fullam's order was filed after the pending motion to decertify was filed of record and recognized that Defendants' raised substantial arguments that the certification order was probably defective and inconsistent with prevailing law since he stated that "further information" was needed "in reaching a correct decision as to class certifications." (Id.)
This Order, as well as the pending motion to decertify, are fully consistent with the authority granted by Rule 23(c)(1)(C) stating that an "order that grants or denies class certification may be altered or amended before final judgment." Fed. R. Civ. P. 23(c)(1)(C). They are also consistent with Third Circuit precedent. See Barnes, 161 F.3d at 140 (stating that "[u]nder Rule 23(c)(1), district courts are required to reassess their class rulings as the case develops"); Zenith Labs., Inc. v. Carter-Wallace, Inc., 530 F.2d 508, 512 (3d Cir. 1976) (stating that law of the case rules do not apply to class certification rulings).
Plaintiffs' argument that comity should be granted to the Alabama court's decision is thus ill grounded. Granting comity — and thus setting in stone a transferor court's certification decision — would violate Rule 23's explicit pronouncement that certification orders are subject to later modification. Plaintiffs' citation to the United States Supreme Court's decision in Smith v. Bayer Corp., 564 U.S. 299 (2011) is also inapt. In that decision, the Court held that a federal court had exceeded its authority under the Anti-Injunction Act in enjoining a state court from considering the certification of a class in a products liability action that was largely identical to one in which the federal court had denied certification. The case turned on whether the federal court's prior decision had a preclusive effect on the state court. The Court held that "[n]either a proposed class action nor a rejected class action may bind nonparties. What does have this effect is a class action approved under Rule 23." Id. at 315. The Court also stated that it expects "federal courts to apply principles of comity to each other's class certification decisions." Id. at 317. Although Plaintiffs seek to apply this quoted language here, the context is entirely different. Bayer was a case concerning the application of federal-state comity to a state court copycat lawsuit where certification was initially denied by the federal court. It was not a motion for decertification of a class certified in the same lawsuit before venue was transferred from one federal court to another by order of the Multidistrict Panel. In addition, the doctrine of comity is not a rule of issue preclusion, but rather one of deference. Smentek v. Dart, 683 F.3d 373, 377 (7th Cir. 2012) (stating that, after the decision in Bayer, "[w]e are left with the weak notion of `comity' as requiring a court to pay respectful attention to the decision of another judge in a materially identical case, but no more than that even if it is a judge of the same court or a judge of a different court within the same judiciary"). While the Alabama certification order is entitled to deference, it is clearly not entitled to preclusive effect given the language in Rule 23(c)(1)(C).
The question of which party has the burden on a motion to decertify a previously certified class is not settled. Some courts assign the burden to the proponent of the decertification motion. See, e.g., Doe v. Karadzic, 192 F.R.D. 133, 136-37 (S.D.N.Y. 2000) (holding that "the Court may not disturb its prior findings absent `some significant intervening event,'" and defining as significant the same events that would justify a motion for reconsideration, i.e., "an intervening change of controlling law, the availability of new evidence, or the need to correct a clear error or prevent manifest injustice"); In re Atl. Fin. Fed. Securities Litig., Civ. A. No. 89-645, 1992 WL 50072, *2 (E.D. Pa. Feb. 28, 1992) ("[W]hen seeking decertification of a class, the defendant bears a heavy burden to show that there exist clearly changed circumstances that make continued class action treatment improper."). Other courts look to whether the plaintiff has carried the burden of showing a continued right to proceed as a class. See, e.g., Marlo v. United Parcel Service, Inc., 639 F.3d 942, 947 (9th Cir. 2011) (holding that "as to the class-decertification issue," the party seeking certification continues to bear the burden of showing compliance with Rule 23); In re Credit Suisse First Boston Corp. (Lantronix, Inc.) Analyst Securities Litig., 250 F.R.D. 137, 140 (S.D. N.Y. 2008) ("In order to decide whether or not to decertify this class, the Court must determine whether Plaintiff has carried his burden of demonstrating that each element of Rule 23 is met by a preponderance of the evidence."); Walker v. Bankers Life & Cas. Co., Civ. A. No. 06-6906, 2008 WL 2883614, at *9 (N.D. Ill. July 28, 2008) (holding that "plaintiffs bear the burden of producing a record demonstrating the continued propriety of maintaining the class action"); see also Newberg on Class Actions § 7:39 (5th ed.) (collecting cases). Assigning the burden to either party here, however, leads to the same result.
Defendants have met their burden to show that decertification is appropriate since (1) the Alabama certification order did not originally comply with the requirements of Fed. R. Civ. P. 23(c)(1)(B), and (2) the law governing the consideration of class certification has changed in the intervening years so that the record undergirding the Alabama certification order would no longer be deemed sufficient. On the other hand, Plaintiffs, to the extent that they hold the laboring oar, have not met their burden to show continued compliance with Rule 23. With the advent of Duke, Comcast, Hydrogen Peroxide, and Blood Reagents, all decided after the original certification order was entered, commentators note that the federal class action has become far more challenging to certify. See, e.g., Linda S. Mullenix, Ending Class Actions As We Know Them: Rethinking the American Class Action, 64 Emory L.J. 399 (2014); Edward D. Cavanagh, Antitrust Law and Economic Theory: Finding A Balance, 45 Loy. U. Chi. L.J. 123, 159-60 (2013) (noting that, under these decisions, certification proceedings have "become something of a cottage industry for expert economists. As a result, class certification proceedings have been transformed into complex miniature trials, a practice roundly condemned by the Supreme Court three decades ago"). The requirement of rigorous analysis, the ability to engage in an examination of the merits of the claim where needed, and the imperatives to establish that predominance is susceptible to common proof and that damages can be measured on a class-wide basis tied to each theory of antitrust impact, simply did not exist when the Alabama certification order was entered. Through no fault of the judge that entered it, that order is now deficient in numerous respects.
First, contrary to Dukes, the Alabama certification order specifically states that "the Court may not inquire into the merits of Plaintiffs' claims at this preliminary stage." Compare Alabama certification order at 8; Dukes, 564 U.S. at 351. Second, it does not specifically employ the preponderance of the evidence standard. See Hydrogen Peroxide, 552 F.3d at 320 (citing Gariety v. Grant Thornton, LLP, 368 F.3d 356, 365 (4th Cir. 2004)). Third, it did not resolve concededly disputed evidence, concluding instead that Plaintiffs had met their burden by merely providing some evidence on an issue. See Alabama certification order at 12 (finding antitrust standing sufficiently demonstrated by "substantial, although disputed, evidence".) Finally, it made no attempt to analyze competing expert submissions on the predominance issue before granting certification, deferring that issue to a later stage of the litigation. (Id. at 26 (stating "In the face of Plaintiffs' evidence and their expert's analysis, I find that, at least at this stage, the Plaintiffs have carried their burden of establishing predominance.").)
Having rigorously analyzed the certification record, and having found it wanting in the key respects discussed supra, the Motion for decertification in Express Scripts and Medco is granted.
Similar to Caremark, Plaintiffs in Brady seek to certify a class consisting of pharmacies that contracted or were under contract with Medco to dispense and sell brand name and generic prescription drugs for any prescription drug benefit plan. (See ECF 57.) Plaintiffs allege an antitrust conspiracy essentially identical to the Plan Sponsor conspiracy alleged by Plaintiffs in Caremark. Also similar to Caremark, Plaintiffs rely on the expert reports submitted by Dr. Cowan. (See ECF 45-2 (appending Cowan Rebuttal Report to Pls.' Joint Resp. to Defs.' Suppl. Mem. in Opp. to Pls.' Mot. for Class Cert.).) Because the conspiracy alleged in Brady is judged under the rule of reason, the same Daubert defects and substantive class certification predominance issues identified in the discussion of the Caremark Plaintiffs' rule of reason claim prevent certification of Brady Plaintiffs' claim.
Rigorous analysis of pending class certification Motions leads to the conclusion that Plaintiffs' expert submissions fail to pass Daubert scrutiny and that Plaintiffs are unable to meet their burdens under Rule 23. Accordingly, Caremark's Motion to exclude expert evidence shall be granted, Plaintiffs' Motions for class certification in the lead case and in Brady shall be denied, and the class previously certified in Express Scripts and Medco shall be decertified.
An appropriate Order follows.
This litigation presently consists of six actions listed on the attached Schedule A as follows: two actions each in the Northern District of Alabama and the Eastern District of Pennsylvania and one action each in the Northern District of Califomia and the Northern District of Illinois. Before the Panel is a motion, pursuant to 28 U.S.c. § 1407, by plaintiffs in one Pennsylvania action seeking centralization of all actions in the Eastern District of Pennsylvania. Plaintiffs in the California action support the motion. Medea Health Solutions, Inc. (Medeo), its former parent Merck & Co" Inc., (Merck) and PAID Prescriptions LLC (PAID)
On the basis of the papers filed and hearing session held, the Panel finds that the actions in this litigation involve common questions of fact, and that centralization in the Eastern District of Pennsylvania will serve the convenience of the parties and witnesses and promote the just and efficient conduct of the litigation. All actions arise out of allegations that certain conduct by the phannacy benefit manager (PBM) defendants-including the negotiation of rates for the sale of prescription drugs by retail pharmacies-violated the federal antitrust laws. Centralizing these actions is desirable in order to avoid duplication of discovery, prevent inconsistent or repetitive pretrial rulings (especially on the issue of class certification), and conserve the resources of the parties, their counsel and the judiciary. See In re Managed Care Litigation, 2000 U.S. Dist. LEXIS 15927 (J.P.M.L. Oct. 23, 2000).
Opposing defendants argue that unique questions offact relating to each PBM should produce a different result. We are unpersuaded by this argument. While the contracts between each plan sponsor/PBM will spawn some unique discovery, all plaintiffs allege that these contracts create a price-fixing conspiracy. Moreover, all actions can be expected to focus on similar PBM practices and procedures. Some plaintiffs also allege that the PBMs conspired with each other to further the price-fixing conspiracies. Transfer to a single district under Section 1407 has the salutary effect of placing all actions before one court which can fonnulate a pretrial program that: 1) allows pretrial proceedings with respect to any non-common issues to proceed concurrently with pretrial proceedings on common issues, In re Multi-Piece Rim Products Liability Litigation, 464 F.Supp. 969, 974 (J.P.M.L. 1979); and 2) ensures that pretrial proceedings will be conducted in a manner leading to the just and expeditious resolution of all actions to the overall benefit of the parties. The MDL-1782 transferee court can employ any number of pretrial techniques-such as establishing separate discovery and/or motion tracks-to efficiently manage this litigation. In any event, we leave the extent and manner of coordination or consolidation of these actions to the discretion of the transferee court. In re Equity Funding Corp. ojAmerica Securities Litigation, 375 F.Supp. 1378, 1384-85 (l.P.M.L. 1974).
Given the geographic dispersal of constituent actions, any of the suggested transferee districts would be an appropriate transferee forum. We are persuaded that the Eastern District of Pennsylvania, where two actions are currently pending, has the experience to steer this litigation on a prudent course.
IT IS THEREFORE ORDERED that, pursuant to 28 U.S.c. § 1407, the actions listed on Schedule A and pending outside the Eastern District of Pennsylvania are transferred to that district and, with the consent of that court, assigned to the Honorable John P. Fullam for coordinated or consolidated pretrial proceedings with the actions pending there and listed on Schedule A.
North Jackson Pharmacy, Inc., et al. v. Express Scripts Inc., et al., C.A. No. 5:03-2696 North Jackson Pharmacy, Inc., et al. v. Medea Health Solutions, Inc., et aI., C.A. No. 5:03-2697
Mike's Medical Center Pharmacy, et al. v. Medea Health Solutions, Inc., et aI., C.A. No. 3:05-5108
North Jackson Pharmacy, Inc., et al. v. Caremark RXInc., et aI., c.A. No. 1:04-5674
Brady Enterprises, Inc., et al. v. Medea Health Solutions, Inc., et al., C.A. No. 2:03-4730 Bellvue Drug Co., et al. v. AdvancePCS, C.A. No. 2:03-4731.
After the case was filed in Alabama, it was transferred to the Northern District of Illinois. (See No. 04-cv-05674 (N.D. Ill.) On August 12, 2005, that Court issued an order holding that the Plan Sponsor conspiracy claim was subject to rule of reason analysis. See North Jackson Pharmacy, Inc. v. Caremark Rx, Inc., 385 F.Supp.2d 740 (N.D. Ill. 2005) ("the Illinois Opinion"). There has not yet been a judicial determination whether the inter-PBM conspiracy is to be adjudicated under the per se rule, the rule of reason or a quick-look analysis. For purposes of the pending motions only, it is assumed that the per se rule applies to that claim. Accordingly, the discussion infra related to product and geographic market definitions and market power refers only to the rule of reason claim.
It should be noted that in the introduction to their certification submission, Plaintiffs assert that "[b]oth conspiracies involve horizontal arrangements subject to per se analysis, not evaluation under the rule of reason." (ECF 248 at 1 (emphasis added).) They go on to argue at length that the Illinois Opinion, issued before the case became part of the MDL, erroneously determined that the rule of reason applied to the Caremark-Plan Sponsor conspiracy claim (see ECF 248 at 16-22) and assert that "the Court should reject any attempt by Caremark to justify its agreements to fix prices." (Id. at 22.) The determination that the rule of reason and not the per se rule applied to the Plan Sponsor conspiracy claim is the law of the case, and Plaintiffs fail to couch their argument in terms that acknowledge that significance. This failure is somewhat odd considering that law of the case issues have previously arisen in the MDL with Plaintiffs successfully asserting before the United States Court of Appeals for the Third Circuit that the prior transferor judge erred in failing to grant law of case status to a ruling issued by a transferee judge. See In re Pharmacy Benefit Managers Antitrust Litig., 582 F.3d 432, 439 (3d Cir. 2009) (stating that the `"[l]aw of the case rules have developed "to maintain consistency and avoid reconsideration of matters once decided during the course of a single continuing lawsuit."'" (quoting Casey v. Planned Parenthood of Se. Pa., 14 F.3d 848, 856 (3d Cir. 1994) (quoting 18 Charles A. Wright, Arthur R. Miller, Edward Cooper, Federal Practice and Procedure § 4478 at 788 (2d ed.1981))).
Under law of the case doctrine, the discretion of a court to revisit its own ruling or that of a coordinate court is limited to "extraordinary circumstances" (1) where new evidence is available, (2) where a supervening new law has been announced, (3) where there is a need to clarify or correct an earlier, ambiguous ruling, or (4) where an unambiguous ruling might lead to an unjust result. In re Pharmacy Benefit Managers Antitrust Litig., 582 F.3d at 439 (internal quotations and citations omitted). Because Plaintiffs cite to no new evidence or change in the law, the Illinois Opinion's discussion of why the per se rule should not apply to the Plan Sponsor conspiracy claim is unambiguous, and the result that the rule of reason should apply thereto is not unjust, there is no cause to revisit the ruling. Moreover, to the extent that Plaintiffs argue that the ruling was incorrect, we fully agree with the Illinois Opinion's reasons for determining that the conspiracy alleged between Caremark and the Plan Sponsors should be governed under the rule of reason. See Illinois Opinion, 385 F. Supp. 2d at 749 (because "the bundle of services provided by Caremark reflects a cooperative arrangement between Caremark and the Plan Sponsors that has efficiency-enhancing potential" rule of reason analysis was appropriate).
(2006 Cowan Report at 2.)
(2006 Cowan Report at 38.)
(Id. at 4.)
(Id. at 5.)
To address the criticism that he did not consider pharmacy networks, Seguin has appended a new exhibit to his rebuttal in which he analyzed reimbursements for the 10 most commonly filled prescriptions as defined by Dr. Hausman, selecting the single most prescribed quantity, then, for each unique value of "network ID," calculated the average total reimbursement and the percent of those prescriptions filled under that network ID by IP. (Seguin Rebuttal at 5.) He opines that "[t]he (statistically significant) negative correlations show that as the percent of prescriptions under that network ID filled by independent pharmacies goes up, average compensation declines. In other words, those network IDs associated with greater independent pharmacy participation are also associated with lower rates of reimbursement." (Id.) He finds a 15.3% decline in the average reimbursement for filling a prescription of 30 pills under an IP contract as compared to a chain contract. (Id. at 6.) Dr. Seguin's use of averages is addressed infra.
To address the criticism that he did not consider market concentration, Seguin contends that his NPZ variable, which is the number of pharmacies in a zip code, is used to account for market concentration. While he did not specifically discuss his reasons for using zip codes in his first report, he asserts that using it to define geographic areas is accepted and is used by Dr. Hausman himself. (Id. at 7-8.) He also contends that using zip codes addresses Hausman's criticism concerning the lack of geographic factors. (Id. at 10.) He adds that he has also incorporated median income of each zip code, thereby accounting for demographic factors. (Id.) He concedes that he has not considered "membership in a bargaining collective," but notes that this data was not provided by either side. (Id.) He explains the lack of consideration of the "tradeoff between brand name and generic drugs" by noting that "the decision to fill `as written' is generally up to the prescribing physician, an individual that is not a party to this case and not subject to any influence by the PBM." (Id.)
Atl. Richfield Co., 495 U.S. at 344.
When confronted with Dr. Hausman's evidence that the data show that 52.60% of IPs receive reimbursements that are greater than the median chain (see Hausman Report at 25; Table 3), Dr. Seguin continued to assert that "as a class, they suffered injuries of $300,000 for this drug." (ECF 261-8 at 115.) When asked to assume an example of an IP that had one Lipitor prescription fill in 2005 for which it received a higher reimbursement rate than a chain pharmacy, and asked whether it had "been harmed by the alleged conspiracy," Seguin stated that "[a]s a member of the class, they were harmed. Again, you're asking me — if you're willing to concede there that the class was harmed by $300,000; I'm willing to have a different discussion as to how to apportion that. And, you know, whether or not a particular independent pharmacist who filled one prescription at $80 deserves compensation or not is, I think, up to the counsel and class to decide." (Id. at 117:2-14.)