Cason–Merenda v. Detroit Med. Ctr.

Full title: Pat CASON–MERENDA and Jeffrey A. Suhre, on behalf of themselves and others…

Court: United States District Court, E.D. Michigan, Southern Division.

Date published: Mar 22, 2012

862 F. Supp. 2d 603 (E.D. Mich. 2012)

Opinion

Case No. 06–15601.

2012-03-22

Pat CASON–MERENDA and Jeffrey A. Suhre, on behalf of themselves and others similarly situated, Plaintiffs, v. DETROIT MEDICAL CENTER, Henry Ford Health System, Mount Clemens General Hospital, Inc., St. John Health, Oakwood Healthcare Inc., Bon Secours Cottage Health Services, William Beaumont Hospital, and Trinity Health Corp., Defendants.

Stephen F. Wasinger, Stephen F. Wasinger PLC, Royal Oak, MI, Daniel Cohen, Cuneo, Gilbert, David P. Dean, James and Hoffman, Washington, DC, Mark A. Griffin, Raymond J. Farrow, Tana Lin, Keller Rohrback, Seattle, WA, for Plaintiffs. Alethea A. Wilson, Charles N. Raimi, David A. Ettinger, Honigman, Miller, Schwartz and Cohn LLP, Terrence J. Miglio, Gouri G. Sashital, Keller Thoma, PC, David A. Hardesty, Thomas M. J. Hathaway, Clark Hill, Fred K. Herrmann, Kerr, Russell, Detroit, MI, David Marx, Jr., Amy J. Carletti, David L. Hanselman, Jr., Stephen Y. Wu, McDermott, Will, Margo Weinstein, Snr Denton U.S. LLP, Chicago, IL, Michael R. Shumaker, Jones Day, Shari Ross Lahlou, Crowell & Moring, Washington, DC, Howard B. Iwrey, Dykema Gossett, Sheldon H. Klein, Butzel Long, Bloomfield Hills, MI, Mark T. Nelson, Butzel Long, Ann Arbor, MI, William B. Slowey, Butzel Long, Detroit, MI, Cathrine F. Wenger, Trinity Health, Novi, MI, Corey M. Shapiro, David B. Gunsberg, Birmingham, MI, Sandra D. Hauser, Snr Denton U.S. LLP, New York, NY, for Defendants.

GERALD E. ROSEN

Stephen F. Wasinger, Stephen F. Wasinger PLC, Royal Oak, MI, Daniel Cohen, Cuneo, Gilbert, David P. Dean, James and Hoffman, Washington, DC, Mark A. Griffin, Raymond J. Farrow, Tana Lin, Keller Rohrback, Seattle, WA, for Plaintiffs. Alethea A. Wilson, Charles N. Raimi, David A. Ettinger, Honigman, Miller, Schwartz and Cohn LLP, Terrence J. Miglio, Gouri G. Sashital, Keller Thoma, PC, David A. Hardesty, Thomas M. J. Hathaway, Clark Hill, Fred K. Herrmann, Kerr, Russell, Detroit, MI, David Marx, Jr., Amy J. Carletti, David L. Hanselman, Jr., Stephen Y. Wu, McDermott, Will, Margo Weinstein, Snr Denton U.S. LLP, Chicago, IL, Michael R. Shumaker, Jones Day, Shari Ross Lahlou, Crowell & Moring, Washington, DC, Howard B. Iwrey, Dykema Gossett, Sheldon H. Klein, Butzel Long, Bloomfield Hills, MI, Mark T. Nelson, Butzel Long, Ann Arbor, MI, William B. Slowey, Butzel Long, Detroit, MI, Cathrine F. Wenger, Trinity Health, Novi, MI, Corey M. Shapiro, David B. Gunsberg, Birmingham, MI, Sandra D. Hauser, Snr Denton U.S. LLP, New York, NY, for Defendants.

OPINION AND ORDER REGARDING DEFENDANTS' MOTIONS FOR SUMMARY JUDGMENT

GERALD E. ROSEN, Chief Judge.

I. INTRODUCTION

In this case, the Plaintiff registered nurses (“RNs”), Pat Cason–Merenda and Jeffrey A. Suhre, seek to recover on behalf of themselves and a class of RNs against eight Detroit-area hospitals, alleging that the Defendant health care providers have violated § 1 of the federal Sherman Act, 15 U.S.C. § 1, by (i) conspiring among themselves and with other local hospitals to hold down the wages of RNs employed by these institutions, and (ii) exchanging compensation-related information among themselves in a manner that has reduced competition among Detroit-area hospitals in the wages paid to RNs. This Court's subject matter jurisdiction rests upon Plaintiffs' assertion of claims arising under federal law. See28 U.S.C. § 1331.

Through the motions presently before the Court, the five remaining Defendants seek awards of summary judgment in their favor on each of Plaintiffs' claims against them.¹ In separate motions brought by Defendant Detroit Medical Center and the four other Defendant hospitals—specifically, Defendants Henry Ford Health System, Mount Clemens General Hospital, Inc., William Beaumont Hospital, and Trinity Health Corp.—Defendants argue that Plaintiffs have failed to produce either direct or circumstantial evidence of any agreement among Detroit-area hospitals to fix RN compensation, and that the evidence, to the contrary, reflects independent decisionmaking by each of the Defendant health care institutions. The five moving Defendants further contend that the record fails to establish any anticompetitive effects resulting from the exchange of compensation-related information among Detroit-area hospitals. ²

1 Following lengthy negotiations overseen by a court-appointed special master, Professor Eric Green, three of the eight Defendant hospitals—St. John Health, Oakwood Healthcare Inc., and Bon Secours Cottage Health Services—agreed to settle the claims against them, and these settlements were embodied in final orders and judgments entered by the Court on September 8, 2010. More recently, following the August 11, 2011 hearing on the five remaining Defendants' summary judgment motions, the Court appointed the Honorable Layn R. Phillips as special master to conduct further settlement negotiations. To date, this latest round of extensive and complex negotiations has yielded tentative settlements between Plaintiffs and two more Defendant hospitals, Mount Clemens General Hospital, Inc. and William Beaumont Hospital. Because these latter settlements have not yet been finalized or formally submitted to the Court for approval, the Court will continue to include Mount Clemens and Beaumont among the group of five remaining Defendant hospitals referenced in the balance of this opinion.

2 In addition to joining in the motion brought by the other Defendant hospitals, Defendant Mount Clemens General Hospital separately moved for summary judgment on the ground that the wages paid to its largely unionized RN workforce have been determined solely through a lawful collective bargaining process, and have not been affected by any alleged conspiracy to fix RN compensation or any sharing of wage-related information with other local hospitals. In light of the tentative settlement reached between Plaintiffs and Mount Clemens, the Court denies this separate motion without prejudice, and will revisit this motion only in the event that this settlement is not finalized.

These two summary judgment motions have been fully and thoroughly briefed by the parties.³ In addition, the Court held an August 11, 2011 hearing on these motions, at which counsel offered extensive and skillful argument in support of their respective positions. Having reviewed the parties' briefs and the accompanying, voluminous record, and having carefully considered the arguments of counsel at the August 11 hearing, the Court now is prepared to rule on Defendants' motions for summary judgment. This opinion and order sets forth the Court's rulings on these motions.

3 Because of the considerable overlap in the arguments advanced in the summary judgment motions filed by the Detroit Medical Center and the four other Defendant hospitals, Plaintiffs have filed a single, consolidated response in opposition to these two motions.

II. FACTUAL AND PROCEDURAL BACKGROUND

According to Plaintiffs' third corrected class action complaint, Plaintiff Pat Cason–Merenda is a registered nurse (“RN”) who has been employed by the Defendant Detroit Medical Center (“DMC”) since November of 2002. During this same time frame, Plaintiff Jeffrey A. Suhre has worked as an RN at Providence Hospital, a health care facility owned and operated by Defendant St. John Health. In their complaint, Plaintiffs allege that the eight Defendant hospitals, along with other hospitals in the Detroit metropolitan area, have conspired among themselves to depress the level of compensation paid to their RN workforces, and that they have implemented a scheme of exchanging compensation-related information that has reduced competition among Detroit-area hospitals in the compensation of their RN employees. In bringing these claims under federal antitrust law, Plaintiffs seek to represent a class of individuals who were employed as RNs by any of the Defendant hospitals at any time from December 12, 2002 through the present.⁴

4 Plaintiffs' pending motion for class certification defines the class somewhat more narrowly, but this broader characterization suffices for present purposes. In addition, the Court notes that the December 2002 commencement of the class period presumably reflects the four-year statute of limitations for federal antitrust claims, see15 U.S.C. § 15b, with this suit having been filed in December of 2006. Finally, because two of the Defendant hospitals, Beaumont and Trinity, were not added as parties until June of 2007, the limitation period for the claims against these Defendants presumably extends back only to June of 2003, rather than December of 2002.

A. The Exchange of Compensation–Related Information Among the Defendant Hospitals

Although Plaintiffs have accused the Defendant hospitals of committing two distinct violations of federal antitrust law, these two theories of recovery rest upon a common factual predicate—namely, that Defendants have routinely engaged in substantial exchanges of information about how they compensate their RN workforces. Accordingly, the Court finds it appropriate to recount in considerable detail the evidence in the record reflecting these exchanges of wage-related information. In particular, Plaintiffs have identified three principal mechanisms through which this information has been shared among Detroit-area hospitals: (i) direct contacts between employees of the various hospitals who were involved in the process of determining RN compensation at their respective institutions; (ii) health care industry organizations and meetings that addressed nursing issues, including compensation; and (iii) third-party surveys of RN compensation sponsored by the Defendant hospitals. The Court reviews each of these mechanisms in turn.

1. Direct Exchanges of Wage–Related Information Among Employees of Detroit–Area Hospitals

The record reveals that it was not uncommon—particularly in the early days of the relevant time period from December of 2002 forward—for an employee of one of the Defendant hospitals to contact his or her counterpart at another Detroit-area hospital and obtain information relating to the compensation of RNs. Indeed, as Plaintiffs observe, each of the Defendant hospitals has acknowledged in the course of this litigation that its employees communicated with employees of other hospitals during the pertinent time period regarding RN compensation. ( See Plaintiffs' Consolidated Response Br. at 14 (citing Defendants' answers and interrogatory responses).) As detailed below, these direct contacts took a variety of forms.

First, from 1989 until October of 2003, Thomas Dabrowski of Defendant William Beaumont Hospital conducted quarterly surveys of the compensation paid to nurses and other employees at a number of Detroit-area hospitals. ( See Plaintiffs' Response, Ex. 12, compilation of survey results for October 2002, January 2003, April 2003, and July 2003.) Each of the Defendant hospitals other than Bon Secours participated in at least one of these surveys.⁵ The participant hospitals then were provided with aggregated survey results, ⁶ which disclosed, among other information, the average minimum and maximum pay ranges and actual pay rates for a number of positions. As Plaintiffs observe, the information provided by the participant hospitals typically reflected current (as opposed to historical) pay rates and ranges, ( see, e.g., Plaintiffs' Response, Exs. 13, 30), and on at least one occasion, one of the survey participants, Defendant Trinity, disclosed its plan for a future merit increase, ( see Plaintiffs' Response, Ex. 14).⁷

5 As Plaintiffs note, the human resources manager for Defendant Mount Clemens, Paula Mutch, testified at her deposition that “[i]t was practice in Mount Clemens to participate in third-party surveys, so the hospital did not participate in direct surveys,” and she further clarified that “direct” meant “hospital-to-hospital” surveys. (Plaintiffs' Response, Ex. MM, Mutch Dep. at 40.) This testimony, however, is contradicted by the evidence showing that Mount Clemens did, in fact, participate in at least two of the surveys conducted by Mr. Dabrowski of Beaumont Hospital. ( See Plaintiffs' Response, Exs. 12C, 12D.) Although Defendants complain that Plaintiffs have “misquote[d]” this testimony and taken it “out of context,” (Defendants' Reply Br. at 7 n. 7), the Court fails to see how Ms. Mutch's testimony can be squared with the record on this point.

6 The parties draw a distinction in their briefs between “aggregated” compensation data—in which the information from a number of sources is combined and reported collectively—and “disaggregated” data—in which the information received from each source is kept and reported separately, albeit sometimes with a generic name ( e.g., “Hospital A”) substituted for the true source of the data. As discussed below, this distinction between “aggregated” and “disaggregated” data has significance under a statement issued by the U.S. Department of Justice and Federal Trade Commission in August of 1996, which sets forth the antitrust enforcement policies of these federal agencies with regard to certain forms of joint activity in the health care industry. See U.S. Dep't of Justice & Federal Trade Comm'n, Statements of Antitrust Enforcement Policy in Health Care (1996) (“DOJ/FTC Guidelines”) (attached as Ex. 32 to Defendant DMC's Motion for Summary Judgment).

7 As noted, Beaumont's direct hospital-to-hospital surveys ceased in October of 2003, when a Beaumont management official issued a memo stating that the hospital would thereafter rely on twice-yearly surveys conducted by a third party, Sullivan, Cotter & Associates. ( See Plaintiffs' Response, Ex. 11.) This memo further urged hospital officials to “[a]ttempt to use published salary survey information if at all possible,” and stated that “[a]d hoc survey requests should be limited to positions that are critical or where there is a recruiting and/or retention issue.” ( Id.)

In addition to these regular Beaumont surveys, the Defendant hospitals requested and provided RN compensation-related informationon an ad hoc basis. Plaintiffs have produced, for example, a number of documents exchanged among the Defendant hospitals in 2001—before the commencement of the class period in December of 2002—disclosing various aspects of their respective RN compensation packages. ( See Plaintiffs' Response, Ex. 5 (Henry Ford providing information to Bon Secours); Ex. 6 (summary of information obtained by Oakwood from a number of the Defendant hospitals); Ex. 7 (Oakwood providing nurse wage information to Trinity); Ex. 8 (Henry Ford disclosing information it obtained about retention bonus practices of several Defendant hospitals); Ex. 9 (Oakwood advising Beaumont of its practices regarding shift differential pay).) These ad hoc information exchanges continued into 2002, ( see, e.g., Plaintiffs' Response, Ex. 17 (February 2002 e-mail in which Trinity human resources employee states that she spoke to a Mount Clemens employee about nurse technician rates and “in exchange” obtained the latter hospital's physician assistant rates); Ex. 21 (Beaumont responding in July 2002 to Oakwood request for information); Ex. 22 (March 2002 fax from Trinity to Oakwood providing a copy of Trinity's “Salaried Employee Merit” policy); Ex. 26 (Henry Ford document dated May 2002 disclosing information learned in survey of signing bonuses paid by other Defendant hospitals); Ex. 27 (November 2002 chart summarizing information obtained by Henry Ford regarding other Defendant hospitals' projected merit increases for 2003); Ex. 31 (Oakwood providing current wage information in mid–2002 in response to survey by Trinity)), and also extended into the post-December 2002 class period, ( see, e.g., Plaintiffs' Response, Ex. 15 (May 2003 e-mail exchange between DMC and St. John employees disclosing nurse pay rates); Ex. 19 (September 2004 e-mail exchange between DMC and Beaumont employees disclosing the two hospitals' most recent bedside nurse pay increases); Ex. 23 (April 2003 e-mail from St. John employee reporting information obtained from Mount Clemens); Ex. 25 (December 2006 chart compiled by Mount Clemens consultant reporting signing bonuses and loan forgiveness programs offered by other Detroit-area hospitals); Ex. 28 (December 2006 response by Bon Secours employee to St. John's request for information regarding signing bonuses and loan forgiveness); Ex. 32 (May 2003 e-mail from St. John employee reporting information obtained from Beaumont regarding pay raises awarded earlier that month and new RN pay ranges); Ex. 33 (July 2005 e-mail from DMC employee reporting call received from St. John employee disclosing St. John's and Oakwood's midnight shift premiums for RNs); Ex. 94 (March 2006 e-mail correspondence indicating that St. John's personnel “check[ed] with our competition” to determine how a proposed pay increase compared with the increases given by Oakwood and Henry Ford); Ex. 102 (September 2004 Beaumont report of survey of other Detroit-area hospitals regarding recent pay increases for bedside nurses)).

More generally, the record discloses that Defendants' human resources and compensation staff retained contact lists of their counterparts at other Detroit-area hospitals. ( See Plaintiffs' Response, Ex. 10A (June 2002 contact list maintained by Oakwood's compensation department in order to survey “local competitors” as part of a “job evaluation process”); Ex. 10B (contact list maintained by St. John); Ex. 10C (June 2003 contact information maintained by DMC for use in conducting “local surveys”); Ex. 10D (June 2006 list of contacts used by DMC to obtain compensation-related information from Henry Ford); Ex. JJ, Logan Dep. at 72–73 (testifying to a list of contacts maintained at Henry Ford prior to August of 2003).) The documentary record, as well as the deposition testimony of certain of these employees, reflects a widespread belief in the quid pro quo nature of these contacts—that is, an understanding that if one of the Defendant hospitals wished to obtain compensation-related information from other Detroit-area hospitals, the best way to achieve this result was to share its own information with these other institutions. ( See, e.g., Plaintiffs' Response, Ex. 10E (January 2006 memo from DMC's director of compensation to her staff stating that “I am a big believer in networking and always find my compensation colleagues to be very generous in sharing information, and of course appreciate any information in return”); Ex. 16 (undated fax from Trinity human resources employee to her counterpart at Oakwood stating that “I'm returning your wage survey and hope you can help me out with one I'm conducting”); Ex. 20 (July 2003 e-mail from Oakwood's chief nursing officer, Barbara Medvec, authorizing the disclosure of certain wage information to a St. John's hospital “if you can also get what they are currently at and/or considering”); Ex. D, Barcome Dep. at 83 (explaining that Trinity would participate in surveys even if it was not given the results, “[b]ecause then if we wanted to have that information, then more than likely they would share it with us”); Ex. M, Dabrowski Dep. at 60–61 (testifying that he felt comfortable calling his counterparts at other hospitals when he had questions, and that they would likewise call him with their questions).) As aptly summarized by Jan Wiseman of Trinity's corporate compensation department in a January 2007 e-mail, “I know that in the past, there have been many informal calls made between organizations and that information has been pretty freely shared.” (Plaintiffs' Response, Ex. 35.) ⁸

8 Defendants again complain that Plaintiffs have presented this quote in a misleading light, given that Wiseman was responding to an e-mail from an employee at a Trinity hospital in Grand Rapids (and not a Detroit-area hospital) referencing a past practice of calling competing hospitals for salary information and asking whether this was “still permissible.” (Plaintiffs' Response, Ex. 35.) Yet, while Wiseman's comment may have been prompted by an inquiry from a Grand Rapids employee, her response—prefaced as it was by the words “I know”—presumably was based upon her personal knowledge of what she referred to as a past practice of “many informal calls” and “pretty freely shared” information. ( Id.) Before moving to Trinity's corporate compensation department, Wiseman was employed from 1997 to 2003 as a compensation analyst and compensation specialist at Trinity's St. Joseph Mercy Hospital in Ann Arbor. ( See Plaintiffs' Response, Ex. DDD, Wiseman Dep. at 12–13.) In that position, she likely would have been familiar with St. Joseph Mercy–Ann Arbor's participation in the April and July 2003 wage surveys conducted by Beaumont, ( see Plaintiffs' Response, Exs. 12C, 12D), and she testified that she provided compensation-related information to her counterparts at other hospitals on at least a “sporadic basis” while employed at St. Joseph Mercy, (Wiseman Dep. at 121). Thus, there is an evidentiary basis for Wiseman's claim of personal knowledge as to a past practice of informal contacts and freely shared information. In addition, the record certainly includes examples of Trinity's contacts with other Detroit-area hospitals to exchange wage-related information, ( see, e.g., Plaintiffs' Response, Exs. 7, 17, 31), and Wiseman's statement could well have been informed by these instances of information sharing. Regardless, then, of Wiseman's somewhat ambiguous deposition testimony as to her awareness of the exchange of salary information between Trinity and other hospitals, ( see Defendants' Reply, Ex. SSS, Wiseman Dep. at 101–02), her above-quoted statement need not be read as limited to Grand Rapids hospitals or based on something other than her own knowledge and experience as a compensation professional at Trinity.

2. The Exchange of Wage–Related Information at Industry Meetings or Through Health Care Industry Organizations

Plaintiffs next point to certain health care industry meetings and organizations as mechanisms through which the Defendant hospitals exchanged wage-related information. First, Plaintiffs cite working group meetings attended by the chief executive officers of some of the Defendant hospitals in 2004 through 2006. Yet, as Defendants point out, there is no evidence in the record that issues of nurse compensation were discussed at these meetings, or that compensation-related information was exchanged.

Nonetheless, other industry organizations, as well as contacts established through these organizations, evidently provided a means for disseminating wage-related information among employees of the Defendant hospitals. First, in late 2005, a “Health Care Roundtable” was formed as a subgroup of the Michigan Ontario Compensation Association (“MOCA”), an all-industry organization. This group met quarterly and although its membership is not clear, all eight Defendant hospitals evidently received notice of its meetings. ( See Plaintiffs' Response, Ex. 38.) ⁹

9 In addition, all eight Defendant hospitals participated in a 2006 salary survey conducted by MOCA. ( See Plaintiffs' Response, Ex. 72.)

When members of the Health Care Roundtable were invited to propose topics for discussion at a scheduled meeting in January of 2006, Carolyn LeGault of Trinity's corporate compensation group suggested that the group could address “market sensitive jobs, recruitment challenges and how current pay plans are affected by the external market and competition for a limited pool of candidates.” (Plaintiffs' Response, Ex. 37.) In response, Eileen Vernor of DMC stated that she was “so glad those are issues you want to address, Carolyn, as those are issues for the DMC as well, but I did not think other health care systems would want to discuss their trade secrets!” ( Id.) The minutes of the January 2006 meeting indicate that the attendees discussed “[m]arket [s]ensitive” positions, including RNs, and addressed the problem of “[b]idding wars” that “inflate market rates” and cause employees to “go down [the] street for $.10/hr[.]” (Plaintiffs' Response, Ex. 39.)

Following this meeting, Amy Lumetta of Henry Ford sent an e-mail to MOCA's then-president, Carol Breen, expressing an interest in “continuing the discussion from the last round table, but trying to focus a bit more on the pay practices, etc. vs. supply and demand concerns,” and suggesting that the results of a MOCA survey of Detroit-area hospitals' pay practices for market sensitive jobs could be used “as a reference point in the discussion.” (Plaintiffs' Response, Ex. 40.) The minutes of the Health Care Roundtable's April 2006 meeting reflect that this survey was, in fact, a principal topic of discussion, with an outside presenter opining that health care institutions “always use[ ] compensation-relat[ed] solutions” to their RN shortages but urging them to “shift their thinking to break this cycle.” (Plaintiffs' Response, Ex. 41.) ¹⁰

10 Beyond the matters actually discussed at the roundtable meetings, these gatherings provided an opportunity for employees who handled compensation-related issues at their respective Detroit-area hospitals to meet their counterparts at other local institutions. Following one of these meetings in late 2005, Eileen Vernor of DMC circulated to her co-workers a list of “health care compensation contacts” who had attended the meeting, stating that “[t]hese may or may not be familiar names but could be useful as you market pric[e] jobs or benchmark compensation practices.” (Plaintiffs' Response, Ex. 10E.) Vernor further opined that “[o]pportunities to meet [these individuals] face-to-face are always valuable in reinforcing these contacts and establishing relationships,” with these “compensation colleagues” having proved “to be very generous in sharing information.” ( Id.) Thus, she urged her co-workers to attend the forthcoming January 2006 roundtable meeting hosted by DMC.

Next, employees of the Defendant hospitals exchanged compensation-related information and discussed their pay practices at meetings of the Health Care Recruiters Association of Metropolitan Detroit (“HCRAMD”). Frances Dombrowski of Beaumont—who served as vice president of the HCRAMD in 2006 and president in 2007—testified that this organization helped its members to “understand the [nursing] market [and] what was happening with other organizations,” with one relevant aspect of this market being compensation. (Plaintiffs' Response, Ex. N, Dombrowski Dep. at 33–34.) Ms. Dombrowski was aware of discussions at the HCRAMD meetings regarding the pay practices of the member health care institutions, and she agreed that one of the benefits of membership in the organization was the opportunity to obtain this information from other members. ( See id. at 54–56.) Likewise, Alicia Horvath of Bon Secours—who served as vice president of the HCRAMD in 2002 and president in 2003—agreed that it was “typical” for members to share RN compensation information at meetings, and for members to discuss salary ranges, planned raises, and the timing of annual increases. (Plaintiffs' Response, Ex. AA, Horvath Dep. at 101–02, 110; see also Ex. XX, Stacey Dep. at 30–31 (Gloria Stacey of Mount Clemens testifying that information relating to nurse compensation was exchanged at HCRAMD meetings “[m]aybe a couple times a year”); Ex. 24 (sheet circulated by Ms. Stacey at an HCRAMD meeting requesting information from other hospitals regarding the hiring bonuses paid to ICU nurses); Ex. 43 (minutes of January 2002 HCRAMD meeting reflecting that members shared details of their retention/sign-on bonuses).)

Beyond their interactions at meetings, employees of the Defendant hospitals used their membership in the HCRAMD as a mechanism for obtaining compensation-related information from other hospitals. Ms. Dombrowski of Beaumont testified that she received e-mails from other HCRAMD members sharing pay information. ( See Dombrowski Dep. at 86–87.) In addition, Plaintiffs have produced a number of documents evidencing exchanges of compensation-related information among HCRAMD members. ( See, e.g., Plaintiffs' Response, Ex. 45A (August 2003 e-mail to members reporting disaggregated results—with each participating hospital identified by name—of survey of RN wages, along with dates of anticipated pay increases); Ex. 45B (August 2004 e-mail from Henry Ford seeking pay information for operating room nurses, with response from Mount Clemens providing the name and contact information for its nurse recruiter); Ex. 45C (October 2004 response by Henry Ford to member's request for pay and other information for licensed practical nurses (“LPNs”)); Ex. 45D (June 2005 Oakwood document compiling information learned from HCRAMD members regarding scholarship and loan forgiveness programs for nurses); Ex. 45E (July 2005 e-mail from St. John employee seeking (and receiving) information from HCRAMD members regarding midnight shift premiums); Ex. 45F (September 2005 Henry Ford e-mail reporting sign-on bonus and loan forgiveness information obtained from other HCRAMD members); Ex. 45G (December 2005 e-mail from St. John employee to other HCRAMD members summarizing information received from membership regarding new nurse graduate pay practices).)

Finally, Plaintiffs point to the Southeast Michigan Healthcare Workforce Partners as a means though which human resources executives at the Defendant hospitals met and discussed compensation-related issues. According to St. John's senior vice president and chief worklife officer, Mary Naber, the members of this group pooled their money to hire a third party, Watson Wyatt, to “conduct a survey about the workforce labor shortage,” and the resulting survey report included nurse compensation data. (Plaintiffs' Response, Ex. NN, Naber Dep. at 166–67, 175.) The group then met to discuss the survey data and to “plan[ ] tactics ... from the data.” ( Id.)

3. The Exchange of Wage–Related Information Through Third–Party Surveys

The Defendant hospitals observe, and Plaintiffs acknowledge, that salary surveys conducted by third parties are commonly used in a wide range of industries, including health care, as a legitimate tool in establishing competitive wage rates. As the parties recognize, an August 1996 policy statement issued by the U.S. Department of Justice and the Federal Trade Commission expressly addresses such surveys, explaining that these federal agencies “will not challenge, absent extraordinary circumstances, provider participation in written surveys of ... wages, salaries, or benefits of health care personnel,” so long as (i) the survey in question is “managed by a third-party,” (ii) the “information provided by survey participants is based on data more than 3 months old,” (iii) “there are at least five providers” participating in the survey, with no participant's data “represent[ing] more than 25 percent” of a given reported statistic, and (iv) the “information disseminated is sufficiently aggregated such that it would not allow recipients to identify the ... compensation paid by any particular provider.” DOJ/FTC Guidelines, Statement 6(A) at 63. In Plaintiffs' view, however, the record in the present case indicates that at least some of the third-party surveys sponsored or relied upon by the Defendant hospitals in determining their RN compensation have deviated in one or more respects from the criteria set forth in the DOJ/FTC Guidelines.

First, Plaintiffs have produced evidence that from January of 2001 ¹¹ to March of 2004, the results of various third-party surveys were reported to the sponsoring Defendant hospitals with unmasked, named hospital data, including wage rates and effective dates. ( See Plaintiffs' Response, Ex. 46 (compiling several third-party surveys sponsored by Defendants Henry Ford and St. John, the results of which were reported to the sponsoring hospitals in disaggregated form and with the name of each participating hospital (and its data) expressly identified).) ¹² In addition, Oakwood's former director of compensation, Rhonda Bunce, testified that she had received the results of nurse compensation surveys with individual hospitals explicitly identified. ( See Plaintiffs' Response, Ex. J, Bunce Dep. at 82.) Similarly, Linda Budd of Budd & Associates—a one-person firm frequently used by the Defendant hospitals to conduct wage surveys—testified that she initially reported the results of surveys with the participating hospitals identified by name, but that she switched to code letters “after a few years.” (Plaintiffs' Response, Ex. I, Budd Dep. at 34.)

11 As Plaintiffs point out, this was the earliest date for which documents were produced during discovery in this case. ( See 8/29/2007 Order at 4 (ordering the production of documents dating back to January 1, 2001).)

12 Notably, in a document prepared by St. John's wage and salary division, a report of the results of a July 2002 third-party survey—with disaggregated, unmasked hospital-by-hospital RN wage data for most of the Defendant hospitals in this case—is immediately preceded by a description of the DOJ/FTC Guidelines, including the recommendation that only aggregated salary data should be disclosed. ( See Plaintiffs' Response, Ex. 46C.)

Plaintiffs also point to instances where a Defendant hospital was given a “key” that permitted it to identify the hospitals participating in a third-party survey. Ms. Bunce of Oakwood, for example, testified that she had received keys that would enable her to identify the hospitals that had participated in third-party surveys sponsored by Oakwood. ( See Bunce Dep. at 84–85.) In addition, the record includes evidence that Ms. Budd provided St. John with a key that would enable it to identify each hospital's data as reported in a March 2004 survey, ( see Plaintiffs' Response, Ex. 48), and that Ms. Budd once advised a Bon Secours employee that Oakwood was “Hospital L,” but that “[y]ou never heard that from me!,” (Plaintiffs' Response, Ex. 49).

Next, Plaintiffs point to several instances in which survey results were reported in disaggregated form, under circumstances that made it possible for the Defendant hospitals to “break the code” and determine which data corresponded to which hospital. At times, for example, Ms. Budd would provide updated survey results to a sponsoring hospital with a new participant identified by the next consecutive letter of the alphabet, making it easy to identify this new participant and its corresponding data. ( See Plaintiffs' Response, Ex. 50 (compiling examples of this practice in updated survey data provided by Ms. Budd to the sponsoring hospital, Mount Clemens, in October of 2006).) In other instances, a Defendant hospital could be identified through characteristics unique to it but known to other hospitals— e.g., the time of year that a given hospital awarded pay increases, ( see Plaintiffs' Response, Exs. 51, 52, 56), or a reference to job titles that were unique to a particular institution, ( see Plaintiffs' Response, Exs. 53, 54). As stated by Ann Vano, St. John's former director of compensation, there were “many sources” from which she could determine which hospital corresponded to which letter code. (Plaintiffs' Response, Ex. BBB, Vano Dep. at 77.)

Moreover, while the DOJ/FTC Guidelines advocate the disclosure of only aggregated data, the results of third-party surveys often were reported to the Defendant hospitals in disaggregated form. Most notably, until some time in 2004, Ms. Budd provided disaggregated data to all of the hospitals involved in her surveys, both sponsors and participants. ( See Plaintiffs' Response, Ex. 57; see also Budd Dep. at 154–55.) When sponsors began to complain that they were receiving the same data as participants, Ms. Budd ceased to supply the survey participants with disaggregated data, but continued to provide this data to the survey sponsors. ( See Budd Dep. at 154–55; see also Plaintiffs' Response, Ex. 122 (disaggregated data from August 2004 survey provided to sponsor Mount Clemens).) Even then, Ms. Budd provided the disaggregated results of a November 2005 Oakwood-sponsored survey to Henry Ford, stating that she was “[a]lways happy to oblige [her] old ‘alma mater’ ” where she had worked before forming her own business. (Plaintiffs' Response, Ex. 196; see also Ex. 200 (providing disaggregated results of November 2002 St. John-sponsored survey to non-participant Bon Secours as a “freebie,” with Ms. Budd noting “don't say I never gave you anything”).) Similarly, in 2004, Defendant Henry Ford sponsored a “Nursing Pay Policy Survey” in which the participants—including each of the eight Defendant hospitals—were asked to share a wide range of competitively sensitive information, including whether the participant targeted its RN base pay to a market percentile and, if so, what this percentile was. ( See Plaintiffs' Response, Ex. 74.) The results of this survey were reported in disaggregated form, albeit with the participating hospitals identified by letter codes rather than by name. ( Id.)

In addition, the results reported in Ms. Budd's surveys sometimes ran counter to the recommendation in the DOJ/FTC Guidelines that participant data should be more than three months old. ( See Plaintiffs' Response, Ex. 58 (compiling examples of survey data reported by Ms. Budd that were less than three months old).) At least on some occasions, it apparently was a matter of some importance that the survey data be as up-to-date as possible. As one example, Rachelle Hulett, DMC's director of compensation programs, sent a March 25, 2002 e-mail to Ann Vano of St. John in connection with a critical jobs survey sponsored by St. John and conducted by Ms. Budd, stating that “[a]s promised, we will report our 2002 ranges (effective April).” (Plaintiffs' Response, Ex. 64.) ¹³ In another instance, St. John requested in early March of 2004 that Ms. Budd conduct a “brief survey to see if the RN data reported in their December [2003] survey is already obsolete,” (Plaintiffs' Response, Ex. 66), even though St. John had received this survey data just two months earlier, in January of 2004, ( see Plaintiffs' Response, Ex. 46E).¹⁴ Indeed, Plaintiffs have identified several instances of the Defendant hospitals divulging information in third-party surveys as to their projected future pay increases, with the bulk of this data reported in a disaggregated format. ( See Plaintiffs' Response, Ex. 71 (compiling surveys conducted by Ms. Budd each year between 2001 and 2005 of the participating hospitals' planned pay increases for the coming year); Ex. 72 (report of similar survey conducted by MOCA Health Care Roundtable in 2006 to determine the participating hospitals' planned 2007 pay increases, albeit with results reported in aggregated form).)

13 As Plaintiffs note, the results of this survey reflect that one participant did, in fact, report data effective April of 2002, making it evident to St. John that this data had been supplied by DMC.

14 Again, Plaintiffs note that Ms. Budd evidently complied with this request and provided updated survey data to St. John in March of 2004. ( See Plaintiffs' Response, Ex. 46G.)

Finally, Plaintiffs point to evidence that certain surveys were “third-party” in name only, with participant data routed through the sponsoring hospital on its way to the third party. In June of 2003, for instance, DMC provided its average rates of pay for a number of positions directly to Oakwood for inclusion in a survey sponsored by the latter hospital, and Oakwood then forwarded this information to Ms. Budd so that it could be incorporated into the results of her “third-party” survey. ( See Plaintiffs' Response, Exs. 60, 61.) Again in January of 2004, Oakwood personnel gathered survey data directly from several other health care institutions and forwarded this information to Ms. Budd, who combined this data with information received directly from other hospitals and distributed the results of this “third-party” survey to the participants. ( See Plaintiffs' Response, Ex. QQ, Perez Dep. at 40–41; Ex. 62.) More generally, the record discloses instances in which the sponsoring institutionswere significantly involved in the design of surveys and the presentation of their results, which seemingly runs counter to the notion of a “third-party” survey. ( See Plaintiffs' Response, Ex. MM, Mutch Dep. at 92, 106–07 (Mount Clemens employee testifying that she instructed Ms. Budd as to which categories of nurses to include in her survey and what types of information to gather); Ex. I, Budd Dep. at 183 (testifying that Henry Ford devised the questions to ask in a 2005 survey); Ex. 59 (compilation of e-mails sent by Oakwood and St. John employees to Ms. Budd between 2001 and 2003, specifying such details as the survey participants, the positions to be addressed in the survey, the specific information to be obtained regarding these positions, and the manner in which the results should be presented).)

B. The Role of Competitor Wage Information in Defendants' Compensation Decisions

Plaintiffs concede that there is “no smoking gun here,” (Plaintiffs' Response Br. at 7)— i.e., no direct evidence of an agreement among the Defendant hospitals to depress the wages of their RNs. Likewise, nothing in the record explicitly contradicts the evidence presented by Defendants—principally in the form of declarations offered by high-ranking executives at each of the Defendant hospitals—that they looked to the wages paid by other hospitals as merely one of several factors used in establishing the rates of compensation for their RN workforces, and that each Defendant institution independently set its own RN compensation rates. Nonetheless, Plaintiffs point to evidence that they view as illustrating each Defendant hospital's significant—and, in Plaintiffs' view, inappropriate—use of and reliance upon compensation-related information that it obtained directly from competitors, or that otherwise failed to comport with the “safety zone” criteria set forth in the DOJ/FTC Guidelines for the exchange of wage data among health care providers. The evidence pertaining to each Defendant hospital is summarized below, albeit with the three settling Defendants omitted from this review.¹⁵

15 Although this survey of the record focuses on the five remaining Defendants, this should not be construed as suggesting that the conduct of the three settling Defendant hospitals is wholly irrelevant to the proper disposition of Plaintiffs' claims against the remaining Defendants. Rather, the activities of each alleged co-conspirator in the claimed antitrust conspiracy may properly be considered in resolving Plaintiffs' claims, even if a particular co-conspirator is no longer (or has never been) a party to the case. Accordingly, while the compensation practices of the settling Defendants have been omitted from the present review, some of these practices are mentioned later in this opinion as pertinent to the Court's analysis of the parties' arguments in support of or opposition to Defendants' motions.

1. Detroit Medical Center

According to Robert Griswold, Jr., DMC's director of compensation, his institution has a compensation policy of “try[ing] to approximate the median of the market, giving consideration to our budget constraints and ability to pay.” (Plaintiffs' Response, Ex. V, Griswold 9/17/2008 Dep. at 26.) In order to implement this policy, DMC's compensation team would “look at the survey data that we have to establish what the market is,” and arrive at a “market pricing” from which a recommendation would be forwarded to management. (Plaintiffs' Response, Ex. U, Griswold 4/10/2008 Dep. at 21; see also Ex. EE, Krueger Dep. at 26.) These market analyses were based in part upon two third-party surveys commissioned by DMC during and shortly before the class period—one conducted by Towers Perrin in 2001–02, and the second conducted by Sullivan Cotter in 2003. ( See Defendants' Motion, Ex. L.1, Krueger Decl. at ¶ 7.) In addition, DMC participated in wage surveys conducted or sponsored by other hospitals— e.g., the surveys conducted by Beaumont between October 2002 and July 2003, ( see Plaintiffs' Response, Ex. 12), and the surveys conducted by Linda Budd, ( see Plaintiffs' Response, Ex. V, Griswold Dep. at 47)—because this participation would enable DMC to “get a copy of the results for free.” ( Id.)

Yet, if no published survey data was available, DMC's manager of compensation programs, Rachelle Hulett, testified that it was acceptable to directly contact the hospital's competitors. ( See Plaintiffs' Response, Ex. BB, Hulett Dep. at 96–97.) As corroboration for this testimony, Plaintiffs point to specific occasions when such contacts occurred. In August of 2006, for example, DMC's then-director of compensation, Eileen Vernor, directed Thomasine Krueger of the compensation team to obtain wage information from DMC's Detroit-area competitors. ( See Plaintiffs' Response, Ex. EE, Krueger Dep. at 49.) Notably, Krueger refused to do so, stating her view that Vernor's request was “outside the [antitrust] guidelines that we try to follow.” ( Id. at 50.) Krueger testified that Vernor was “angry” with this refusal and instructed her to get another member of the compensation team to obtain the information, ( id.), and Vernor ultimately contacted the other hospitals herself to secure the desired wage data, ( see Plaintiffs' Response, Ex. CCC, Vernor Dep. at 165; see also Ex. 157 (wage data obtained from Beaumont, Henry Ford, and St. John, including scheduled St. John pay increase in September 2006)).¹⁶ Apart from this incident, Plaintiffs have produced evidence of other instances in which DMC employees—often at the instruction of their superiors—directly contacted competing hospitals to obtain wage-related information. ( See Plaintiffs' Response, Ex. 160 (February 2001 e-mail from Ruthann Liagre, DMC's then-vice president of human resources, asking Ms. Krueger to make a “couple of quick phone calls to our closest competitors” to inquire about their certification bonuses); Ex. 162 (August 2003 e-mail from Ms. Hulett stating that the “VPs are very interested in what other health care systems do for PDL [professional development ladder], in terms of how they pay (hours worked, benefit level, ladder level, etc.),” and asking Ms. Cracium to “do a call around to see what's up”); Ex. 163 (Ms. Liagre conducting a “[q]uick and dirty” September 2003 survey of other health care systems' projected or scheduled pay increases, and then reporting this information to DMC's chief operating officer, Gwen MacKenzie); Ex. 164 (June 2006 e-mail from Ms. Krueger reporting staff nurse pay rates obtained from “SUPER confidential” source at Henry Ford, apparently at the request of DMC's director of human resources, Paulette Griffin).)

16 As Plaintiffs observe, another member of the compensation team had reminded Ms. Vernor about the antitrust guidelines just a few months earlier. Specifically, upon receiving Vernor's January 2006 e-mail (quoted earlier) in which she advised the team of some “health care compensation contacts” and characterized these colleagues as “very generous in sharing information,” (Plaintiffs' Response, Ex. 10E), team member Cynthia Cracium responded that “[a] few years ago, many of our former contacts made the decision to work through 3rd party agencies due to the anti-trust laws, and we lost some valuable contacts,” and she further stated that “[i]t's still very awkward asking for information directly from other hospitals due to the legal issues that could arise.” (Plaintiffs' Response, Ex. 158.)

Once the compensation team gathered this data, whether from surveys or through direct contacts, and formulated its “market pricing” recommendations, the recommendations and underlying “base rate information” were forwarded to DMC's “executive leadership for decision-making by them.” (Plaintiffs' Response, Ex. V, Griswold Dep. at 165.) As one example, the data obtained by Ms. Vernor through her August 2006 contacts with other Defendant hospitals was forwarded to Deloris Hunt, DMC's corporate vice president of human resources, and this information evidently was shared with DMC's chief executive officer, Mike Duggan, at a meeting shortly thereafter. ( See Plaintiffs' Response, Ex. 165.) ¹⁷ The wage rates for nurses at competitor hospitals also were provided to Michael Prusaitis, DMC's director of patient care budget monitoring, for use in DMC's budgeting process. ( See Plaintiffs' Response, Ex. CCC, Vernor Dep. at 172–74; Ex. SS, Prusaitis Dep. at 13.)

17 When asked at his deposition whether he thought it was appropriate for DMC's human resources personnel to communicate with other hospitals about the pay packages offered by their respective institutions, Mr. Duggan responded that “I never gave [it] any thought one way or the other.” (Plaintiffs' Response, Ex. O, Duggan Dep. at 57.)

Plaintiffs view the record as revealing, at least in one instance, that this data gathered from other hospitals formed the basis for a lower merit pay increase than had been recommended. Specifically, they point to a December 2005 memo stating that DMC's compensation department had recommended a 3 percent merit increase for 2006, but that only a 2 percent merit increase had been approved. ( See Plaintiffs' Response, Ex. 166.) When asked to identify the basis for this 2 percent merit increase, as opposed to the 3 percent increase advocated by the compensation department, Mr. Duggan testified that “I would certainly not raise [employee] wages unnecessarily given the narrow margin that we're always operating on to keep our hospitals open,” and he explained that “I try to raise in every specialty what is necessary to keep us competitive and keep our positions filled, and I try not to pay more than that because it jeopardizes the viability of our hospital system.” (Plaintiffs' Response, Ex. O, Duggan Dep. at 53.) ¹⁸

18 DMC complains in its reply brief that Plaintiffs have misrepresented Mr. Duggan's testimony on this point. In testimony immediately preceding the above-quoted passage, Mr. Duggan noted that the 2 percent merit increase for 2006 was only one of the pay increases for which nurses could be eligible—for example, nurses could receive additional increases by moving to a higher step in DMC's “Promoting Excellence” program, ( see DMC's Motion, Ex. 4, Krueger Decl. at ¶¶ 28, 33)—and he explained that he “look[ed] at all of the nurses' increases in a year together” in making compensation decisions. (DMC's Motion, Ex. 10, Duggan Dep. at 52.) In DMC's view, then, Mr. Duggan had “rejected the premise” of the subsequent question by Plaintiffs' counsel that elicited his above-quoted testimony. ( See DMC's Reply Br. at 14.)

2. Henry Ford Health System

According to Robert G. Riney, Henry Ford's executive vice president and chief operating officer, Henry Ford has established “market targets” for each job title, and the market target for “market sensitive positions”—a category that includes RNs—is “the 66th percentile of the market.” (Defendants' Motion, Ex. N, Riney Decl. at ¶ 11; see also Plaintiffs' Response, Ex. Y, Harland Dep. at 53, 55–56.) In assessing how the actual wage rates paid to Henry Ford employees compare to these market targets, and in determining more generally the salary structures and pay practices for RNs and other positions, a team of executives on the “Total Rewards Committee” relies in part on a “Competitive Pay Analysis” prepared each year by Henry Ford's compensation department. ( See Riney Decl. at ¶ 11.) In this Competitive Pay Analysis, the compensation department makes recommendations as to the percentage wage increases for market sensitive positions and all other positions. ( See id. at ¶ 15; see also Plaintiffs' Response, Ex. 117 (November 2002 Competitive Pay Analysis making recommendations for 2003).)

In formulating its Competitive Pay Analysis each year, Henry Ford's compensation department relied in part upon information it gathered or learned regarding local competitors' pay rates and projected increases. ( See Harland Dep. at 19; see also Plaintiffs' Response, Ex. PP, Parkinson–Tripp Dep. at 46, 109.) In the early days of the class period, there were occasions when employees in the compensation department would obtain this information by directly contacting their counterparts at other Detroit-area hospitals. ( See Plaintiffs' Response, Ex. 118 (November 2002 e-mail stating that compensation department employee had been directed to “call around to local competitors to get a feel for what they are projecting”); Ex. 119 (report of information learned from other local hospitals regarding projected 2003 merit increases). ¹⁹) In 2003, however, the compensation department staff were reminded of the antitrust guidelines regarding exchanges of information among competitors, ( see Plaintiffs' Response, Ex. KK, Lumetta Dep. at 33–34), ²⁰ and Henry Ford employees thereafter largely ceased to request compensation-related information directly from competitors, ( but see Plaintiffs' Response, Ex. 109 (September 2005 fax from human resources employee with chart of compensation data from other Detroit-area hospitals and names of contacts from whom this information evidently was obtained); Ex. 116 (minutes of July 2006 meeting indicating that Henry Ford's chief nursing officer, Ronnie Hall, planned to “send[ ] someone to Oakwood to find out” this competitor's contingent rate of pay for RNs).) ²¹

19 As Defendants point out in their reply brief, the information obtained on this latter occasion evidently led the compensation department to recommend a greater wage increase than it had recommended in an earlier draft of its Competitive Pay Analysis. ( See Defendants' Reply, Ex. JJJ, Parkinson–Tripp Dep. at 192–94.)

20 Although the cited testimony of Henry Ford employee Amy Lumetta appears to make reference to a memo that has been held to be protected by the attorney/client privilege, ( see 2/24/2009 Order), the Court nonetheless may consider Ms. Lumetta's testimony as to the actions taken by Henry Ford's compensation department following the dissemination of this memo.

21 Beyond these examples of direct contacts, Plaintiffs point more generally to a March 2002 report generated by Henry Ford's “Market–Based Compensation Recalibration Team.” (Plaintiffs' Response, Ex. 108.) As Plaintiffs observe, this report referenced such measures as “[f]requent audit of competitor pay rates and practices,” “[a]ggressive collection of market compensation information from informal sources,” and “[c]onstant monitoring of compensation competitiveness” by “keep[ing] in touch with other Compensation representatives from the various hospitals throughout the year as to future plans and any changes to current pay practices.” ( Id.) In a supplemental affidavit, however, Mr. Riney states that he was a co-leader of this team, and he asserts (i) that the March 2002 report included only “draft recommendations,” (ii) that “Henry Ford did not take any action with regard to any specific recommendations” in the report, (iii) that the report “does not reflect compensation setting practices, processes, or philosophy at Henry Ford,” and (iv) that “no department at Henry Ford, including the Compensation Department, ever implemented any of the [report's] recommendations.” (Defendants' Reply, Ex. HHH, Riney Suppl. Decl. at ¶¶ 5–6.)

Yet, despite this reminder, Henry Ford employees continued to provide wage-related information in response to direct inquiries from other hospitals. ( See, e.g., Plaintiffs' Response, Ex. 110 (Henry Ford providing St. John with information regarding nurse wages and practices in January of 2006); Ex. 111 (March 2006 e-mail in which St. John employee states that she “confirmed [pay increases] with HFH 2 weeks ago”); Ex. 113 (June 2006 e-mail from member of DMC compensation team reporting information received from “SUPER confidential” Henry Ford source regarding staff nurse pay range and recent change in pay scale).) Henry Ford also continued to both sponsor and participate in third-party surveys, with the results of these surveys occasionally reported to the hospital in a disaggregated format. ( See, e.g., Plaintiffs' Response, Ex. 71D (Henry Ford-sponsored survey conducted by Linda Budd in November 2003, reporting salary planning information in disaggregated form); Ex. 74 (results of 2004 “Nursing Pay Policy Survey” sponsored by Henry Ford, reported in disaggregated form); Ex. 196 (disaggregated results of November 2005 Oakwood-sponsored survey given to Henry Ford by Ms. Budd, who stated that she was “[a]lways happy to oblige [her] old ‘alma mater’ ”).)

Finally, in at least some years during the relevant period, Henry Ford sponsored third-party surveys of other Detroit-area hospitals' planned salary increases for the coming year. ( See Plaintiffs' Response, Exs. 71D, 120 (survey conducted in late 2003 of projected increases for 2004); Ex. 121 (survey conducted in September of 2004 of projected increases for 2005).) Although Defendants claim that the “Competitive Pay Analysis” reports furnished by Henry Ford's compensation department to the executive-level Total Rewards Committee presented the survey data “in highly aggregated form,” (Defendants' Motion, Br. in Support at 7), Plaintiffs correctly observe that the December 2003 and October 2004 reports incorporated the disaggregated results of the two above-referenced salary planning surveys, albeit with the participant hospitals identified by letter codes rather than by name. ( See Plaintiffs' Response, Exs. 120, 121.)

3. Mount Clemens General Hospital

Mount Clemens is the only Defendant hospital with a unionized RN workforce. In preparation for its union negotiations, Mount Clemens commissioned third-party surveys of nurse wage-related information at other Detroit-area and Michigan hospitals. ( See, e.g., Plaintiffs' Response, Ex. 57B (February 2004 survey conducted by Linda Budd reporting detailed RN salary structure data in disaggregated form, but with hospitals identified by letter code rather than name); Ex. 122 (updated August 2004 survey conducted by Ms. Budd due to prolonged union negotiations, with data again in disaggregated form and reporting some data that was only a month old).) Mount Clemens states without contradiction, however, that the data from these surveys were shared with the nurses' bargaining team during negotiations, ( see Defendants' Motion, Ex. BB, Klinger Dep. at 189; Ex. CC, Horde Dep. at 71, 92), and, on some occasions, were shared directly with the RNs themselves, ( see Horde Dep. at 92–94). In addition, the compensation actually paid by Mount Clemens to its unionized RN workforce is disclosed in the collective bargaining agreements (“CBAs”) reached with the nurses' union, copies of which are distributed to each union member.²²

22 Consequently, on those occasions when Mount Clemens employees responded to inquiries from other hospitals or third-party surveys regarding the hospital's wage rates for RNs or plans for future increases, such information was readily obtainable from the current CBA, which was widely available to all members of the Mount Clemens RN union.

Beyond this, Plaintiffs have identified an instance in March of 2006 when, in preparation for upcoming union negotiations, two of the hospital's representatives in the negotiations—David Klinger, Mount Clemens' then-vice president of human resources, and Priscilla Horde, a human resources consultant—obtained from an Oakwood employee a copy of an Oakwood-sponsored survey of RN pay. ( See Plaintiffs' Response, Ex. 123 (January 2006 survey of RN wages paid by local hospitals, reported in an aggregated format).) When these 2006 negotiations again extended through the year, Mount Clemens employees obtained the results of a more recent Oakwood-sponsored survey, ( see Plaintiffs' Response, Ex. 124 (August 2006 survey results provided to Mount Clemens)), and also initiated direct contact in November of 2006 with their counterparts at Beaumont, St. John, and Trinity to obtain the specific wage rates for these hospitals, with this disaggregated, unmasked data then included in a packet given to Mount Clemens' negotiating representatives, ( see Plaintiffs' Response, Ex. 125; see also Ex. CC, Klinger Dep. at 58–60, 83–84). Finally, Plaintiffs note that on one occasion, a Mount Clemens human resources consultant, Gloria Stacey, obtained information from her counterparts at an HCRAMD meeting regarding the hiring bonuses paid to ICU nurses. ( See Plaintiffs' Response, Ex. 24 (undated sheet disclosing this information for various local hospitals).) ²³ As Defendants point out, however, there is no evidence that any of the information obtained by Mount Clemens through direct contacts with other hospitals had any impact on Mount Clemens' negotiations with the RN union.

23 While Defendants claim that this information was gathered outside the class period, in 2001, the passage of Ms. Stacey's deposition cited in support of this contention is silent as to the date of this document.

4. William Beaumont Hospital

Defendant Beaumont's vice president of human resources, Ronald Lilek, has stated in an affidavit that Beaumont “established a minimum target for Inpatient RN base wage paygrade maximum at between the 60th and 80th percentile, with a target of the 75th percentile, of the highest actual rate paid in the market.” (Defendants' Motion, Ex. S, Lilek Decl. at ¶ 25.) In determining its base hourly rate for RNs and endeavoring to meet this target, Beaumont relied in part on survey data. ( See id. at ¶¶ 21, 26.) As discussed earlier, until October of 2003, Beaumont obtained this survey data through its own quarterly surveys conducted by one of its employees, Tom Dabrowski, who directly requested compensation information from other Detroit-area hospitals, and all of the Defendant hospitals other than Bon Secours participated in one or more of these surveys. Thereafter, Beaumont switched to surveys conducted by a third party, Sullivan Cotter & Associates.²⁴

24 As Plaintiffs note, at around the time Beaumont began to use Sullivan Cotter to conduct wage surveys, this outside firm sent Beaumont a September 2003 letter stating that “[a]s a result of the Justice Department guidelines regarding survey data exchanges, you would like to ‘outsource’ the collection of competitive pay data from key healthcare organizations.” (Plaintiffs' Response, Ex. 202.)

Even after Beaumont made this transition to third-party surveys, its employees continued to occasionally contact their counterparts at other hospital for compensation-related information. In July of 2006, for example, Mr. Dabrowski sought and received information from DMC regarding its student loan programs for nurses and other positions. ( See Plaintiffs' Response, Ex. 101.) In September of 2004, Mr. Dabrowski compiled a list of information he had obtained from other local hospitals regarding their most recent pay increases for bedside nurses. ( See Plaintiffs' Response, Ex. 102.) More generally, Beaumont's director of compensation, David Misner, testified that he would permit direct contacts on occasions where “we were maybe looking to survey a singular job, and to go out and get a custom survey through a third party would have been expensive [and] very time consuming,” but he stated that, so far as he was aware, such contacts occurred only “on a sporadic basis.” (Plaintiffs' Response, Ex. LL, Misner Dep. at 141, 183.)

In Plaintiffs' view, the record discloses several instances in which Beaumont used information from the local market to award smaller wage increases or to reduce planned or requested increases in other forms of compensation. In September of 2003 and again in May of 2005, Beaumont's chief operating officer, Kenneth Matzick, determined that budgeted 2.5 percent wage increases for the hospital's nurses should not be awarded, citing the recommendations of human resources personnel that “[b]ased upon a review of the recently completed nursing survey,” a smaller 2.0 percent increase was appropriate. (Plaintiffs' Response, Exs. 104A, 104C.) Likewise, a pay increase that was “anticipated” for December of 2006 was cancelled, with Beaumont's director of human resources, Lucy Vail, explaining that “pay is market-based” and that “[w]e led the market when the increases were given in the summer, resulting in pay as much as $2.00 over top of the market,” so that “there is no need to increase pay at this time.” (Plaintiffs' Response, Ex. 99.) In addition, a length-of-service (“LOS”) increase was reduced by one percent in 2005, based on a “survey of six area hospitals indicat[ing] that Beaumont[']s LOS increases ha[ve] been inconsistent with what is occurring in the wage market.” (Plaintiffs' Response, Ex. 105.) Finally, the results of a June 2005 ad hoc survey of local hospitals were cited as a basis for rejecting a requested increase in the then-current $3.00/hour on-call rate for the acute hemodialysis staff. ( See Plaintiffs' Response, Ex. 106; see also Ex. 103 (April 2003 recommendation, based on market data, that the on-call rate for cardiac cath lab staff be increased to $3.00/hour rather than the requested $4.00/hour, although the requested increase to $4.00/hour was later granted in September 2004).)

5. Trinity Health Corp.

There are three Trinity hospitals involved in this case: (i) St. Joseph Mercy Oakland in Pontiac, Michigan (“Trinity–Oakland”), (ii) St. Mary Mercy Hospital in Livonia (“Trinity–Livonia”), and (iii) the former St. Joseph Mercy Macomb in Clinton Township (“Trinity–Macomb”).²⁵ Although these hospitals are affiliated with Trinity, each of them operates independently, and each has a separate nurse compensation structure with distinct rates of pay and no effort to coordinate with the other Trinity facilities.

25 In July of 2007, after the commencement of this suit, Trinity sold its interest in Trinity–Macomb to Defendant Henry Ford, and this hospital is now part of Henry Ford Macomb Hospital.

None of the three Defendant Trinity hospitals commissioned any third-party wage surveys during the class period. However, Trinity's corporate compensation group collected market wage data from a variety of sources, and used this informationto prepare and disseminate market analyses to each of the three Trinity affiliates. Trinity's corporate recommendation to these affiliates was to target their base pay to the 50th percentile of the market, and all three affiliates adopted this target. In addition to the market analyses furnished by Trinity's corporate compensation group, each of the three Trinity affiliates participated in and received the results of third-party surveys, such as Linda Budd's surveys. ( See, e.g., Plaintiffs' Response, Ex. 128 (affiliates received disaggregated results of survey conducted by Ms. Budd in May 2003); Ex. 129 (affiliates received disaggregated results of survey conducted by Ms. Budd in November 2003).)

Plaintiffs also cite evidence of employees at the three Trinity affiliates initiating direct contacts with their counterparts at other Detroit-area hospitals to obtain wage-related information, although some of these contacts predate the June 2003 commencement of the class period applicable to Trinity. As to Trinity–Livonia, for example, Plaintiffs point to the testimony of Judy Seybert, an employee in the hospital's human resources department until February of 2003, who stated that she would request wage information from other Detroit-area hospitals when instructed by her supervisor to do so. ( See Plaintiffs' Response, Ex. VV, Seybert Dep. at 82–83; see also Ex. 139 (December 2001 request for wage information from Oakwood); Ex. 140 (September 2002 request for wage information from a number of hospitals, with response from Oakwood).) Similarly, Christine Barcome, a compensation analyst at Trinity–Oakland, testified that prior to the arrival of a new supervisor, Martha Murphy, in 2002 or 2003, she would contact competitor hospitals at management's request to obtain market data for RN compensation. ( See Plaintiffs' Response, Ex. D, Barcome Dep. at 15, 97–98, 108–09; see also Ex. 135 (September 2001 request to Oakwood); Ex. 136 (January 2003 request to Oakwood).) ²⁶

26 Shortly after Ms. Murphy became her supervisor, however, Ms. Barcome was directed not to contact other hospitals to obtain compensation-related information. ( See Barcome Dep. at 97–98.)

Next, Plaintiffs have produced evidence that Trinity–Macomb employees also made direct contacts with their counterparts at other Detroit-area hospitals to obtain compensation-related information. ( See Plaintiffs' Response, Ex. 131 (May 2004 request for information with response from St. John); see also Ex. 130 (April 2005 e-mail from Trinity–Macomb's then-vice president of human resources stating that he “may possibly be able to validate/nullify ... through some additional sources” the information obtained by his staff regarding cath-lab on-call rates paid at other hospitals).) The hospital's administrative director of human resources, Lynn Urban, has testified that she considered it part of her duties to obtain wage range information from hospitals outside of the Trinity network. ( See Plaintiffs' Response, Ex. AAA, Urban Dep. at 38–39.) Ms. Urban further testified that she occasionally used the RN wage range information she received through these direct contacts in analyzing whether Trinity–Macomb's RN wages should be increased and the level to which they should be increased. ( See id. at 38.) Finally, beyond the contacts she initiated, Ms. Urban responded to direct contacts from other hospitals seeking nurse wage information, ( see Plaintiffs' Response, Ex. 133 (February 2002 e-mail reflecting that Ms. Urban provided information effective January 2002 to a counterpart at St. John)), and employees at the other two Trinity hospitals did likewise, ( see Plaintiffs' Response, Ex. 137 (Ms. Barcome of Trinity–Oakland providing copy of documents from the hospital's policy and procedure manual); Ex. 14 (Ms. Barcome responding to Beaumont October 2003 survey with current information and indication of planned merit increase in November); Ex. 102 (Ms. Barcome responding to September 2004 Beaumont survey); Ex. 141 (Ms. Seybert of Trinity–Livonia responding to August 2002 Beaumont survey)).

III. ANALYSIS

A. The Standards Governing Defendants' Motions

Through the two motions addressed in the present opinion, each of the five remaining Defendants—Detroit Medical Center, Henry Ford Health System, Mount Clemens General Hospital, William Beaumont Hospital, and Trinity Health Corp.²⁷—seeks summary judgment in its favor on each of Plaintiffs' two claims brought under § 1 of the Sherman Act. Under the pertinent Federal Rule, summary judgment is proper “if the pleadings, the discovery and disclosure materials on file, and any affidavits show that there is no genuine issue as to any material fact and that the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(c).²⁸ As the Supreme Court has explained, “the plain language of Rule 56(c) mandates the entry of summary judgment, after adequate time for discovery and upon motion, against a party who fails to make a showing sufficient to establish the existence of an element essential to that party's case, and on which that party will bear the burden of proof at trial.” Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 2552, 91 L.Ed.2d 265 (1986).

27 As noted earlier, two of these Defendants, Mount Clemens General Hospital and William Beaumont Hospital, recently reached tentative settlement agreements with Plaintiffs, but these settlements have not yet been finalized.

28 .Rule 56 has recently been revised in various respects, but the language quoted here (and immediately below) reflects the Rule as it read when Defendants filed their motions. Under the current Rule, the bulk of the quoted language from subsection (c) has been moved to subsection (a), but the recent amendments do not appear to have altered the overarching standards for resolving a summary judgment motion.

In deciding a motion brought under Rule 56, the Court must view the evidence “in a light most favorable to the party opposing the motion, giving that party the benefit of all reasonable inferences.” Smith Wholesale Co. v. R.J. Reynolds Tobacco Co., 477 F.3d 854, 861 (6th Cir.2007). Yet, the non-moving party “may not rely merely on allegations or denials in its own pleading,” but “must—by affidavits or as otherwise provided in [Rule 56]—set out specific facts showing a genuine issue for trial.” Fed.R.Civ.P. 56(e)(2). Moreover, any supporting or opposing affidavits “must be made on personal knowledge, set out facts that would be admissible in evidence, and show that the affiant is competent to testify on the matters stated.” Fed.R.Civ.P. 56(e)(1). Finally, “[a] mere scintilla of evidence is insufficient” to withstand a summary judgment motion; rather, “there must be evidence on which the jury could reasonably find for the non-moving party.” Smith Wholesale, 477 F.3d at 861 (internal quotation marks and citation omitted).

As Plaintiffs point out, the Sixth Circuit recently observed that “[i]n this circuit, motions for summary judgment are disfavored in antitrust litigation.” Smith Wholesale, 477 F.3d at 862. Yet, the court then emphasized:

This general rule ... does not preclude the use of summary judgment in an antitrust case in which there clearly are no genuine issues of fact to try—indeed, the very purpose of a motion for summary judgment, to eliminate a trial where it would be unnecessary and merely result in delay and expense, warrants summary disposition of such cases when appropriate. Accordingly, the absence of any relevant probative evidence in support of a litigant's antitrust claims will expose such claims to summary judgment disposition.
477 F.3d at 862 (internal quotation marks and citations omitted). Moreover, Defendants correctly note the willingness of both the Supreme Court and the Sixth Circuit to award or affirm summary judgment in antitrust cases, so long as “there is no genuine issue for trial.” Matsushita Electric Industrial Co. v. Zenith Radio Corp., 475 U.S. 574, 587, 106 S.Ct. 1348, 1356, 89 L.Ed.2d 538 (1986) (internal quotation marks and citation omitted); see also Smith Wholesale, 477 F.3d at 880–81;Wallace v. Bank of Bartlett, 55 F.3d 1166, 1170 (6th Cir.1995); Nurse Midwifery Associates v. Hibbett, 918 F.2d 605, 617 (6th Cir.1990). With these principles in mind, then, the Court turns to the challenges advanced in Defendants' motions.

B. Plaintiffs' Per Se Claim of a Conspiracy to Depress RN Compensation

1. The Law Governing Per Se Claims Under § 1 of the Sherman Act

Section 1 of the Sherman Act prohibits “[e]very contract, combination ..., or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations.” 15 U.S.C. § 1. The Sixth Circuit has stated:

[T]o establish a claim under section 1, the plaintiff must establish that the defendants contracted, combined or conspired among each other, that the combination or conspiracy produced adverse, anti-competitive effects within relevant product and geographic markets, that the objects of and conduct pursuant to that contract or conspiracy were illegal and that the plaintiff was injured as a proximate result of that conspiracy.
Expert Masonry, Inc. v. Boone County, Ky., 440 F.3d 336, 342 (6th Cir.2006) (internal quotation marks and citations omitted). Yet, if the challenged anticompetitive practice is deemed illegal per se, “further examination of the practice's impact on the market or the procompetitive justifications for the practice is unnecessary for finding a violation of antitrust law.” Expert Masonry, 440 F.3d at 342 (internal quotation marks and citations omitted). Rather, in the case of practices that are “proscribed per se, the plaintiff need only prove that (1) two or more entities engaged in a conspiracy, combination, or contract, (2) to effect a restraint or combination prohibited per se (wherein the anticompetitive effects within a relevant geographic and product market are implied), (3) that was the proximate cause of the plaintiff's antitrust injury.” 440 F.3d at 342 (citations omitted).

In Count I of their complaint, Plaintiffs allege that the Defendant hospitals “and their co-conspirators have engaged in a continuing conspiracy in restraint of trade to depress the compensation of RNs employed at hospitals in the Detroit [metropolitan area].” (Third Corrected Class Action Complaint at ¶ 51.) The parties agree that such a conspiracy among competing hospitals to fix wages, like an analogous horizontal price-fixing conspiracy, would be subject to per se treatment. See Todd v. Exxon Corp., 275 F.3d 191, 198 (2d Cir.2001); Fleischman v. Albany Medical Center, 728 F.Supp.2d 130, 157 (N.D.N.Y.2010); see also DOJ/FTC Guidelines, Statement 6(B) at 64 (“If an exchange among competing providers of ... cost information results in an agreement among competitors as to ... the wages to be paid to health care employees, that agreement will be considered unlawful per se.”). Absent direct evidence of an explicit agreement among the Defendant hospitals to fix RN wages—which Plaintiffs acknowledge they have not produced—the requisite conspiracy may be established through circumstantial evidence of “business behavior which evidences a unity of purpose or a common design and understanding, or a meeting of the minds in an unlawful arrangement.” Wallace, 55 F.3d at 1168 (internal quotation marks and citations omitted).

The Supreme Court has adopted what this Court previously characterized as a “fairly stringent” standard for assessing circumstantial evidence of an alleged conspiracy in violation of § 1. In re Northwest Airlines Corp., 208 F.R.D. 174, 195 (E.D.Mich.2002). In particular, the Supreme Court has emphasized:

[A]ntitrust law limits the range of permissible inferences from ambiguous evidence in a § 1 case. Thus, in Monsanto Co. v. Spray–Rite Service Corp., 465 U.S. 752, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984), we held that conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy. Id., at 764, 104 S.Ct., at 1470. To survive a motion for summary judgment or for a directed verdict, a plaintiff seeking damages for a violation of § 1 must present evidence “that tends to exclude the possibility” that the alleged conspirators acted independently. 465 U.S., at 764, 104 S.Ct., at 1471. [The plaintiffs] in this case, in other words, must show that the inference of conspiracy is reasonable in light of the competing inferences of independent action or collusive action that could not have harmed [the plaintiffs].
Matsushita, 475 U.S. at 588, 106 S.Ct. at 1356–57 (citations omitted). The Court further cautioned that “mistaken inferences” in assessing the evidence of a claimed antitrust conspiracy “are especially costly, because they chill the very conduct the antitrust laws are designed to protect.” Matsushita, 475 U.S. at 594, 106 S.Ct. at 1360.

In the wake of this Supreme Court precedent, the Sixth Circuit has adopted a “two-part inquiry for evaluating the propriety of summary judgment in an antitrust conspiracy case.” Riverview Investments, Inc. v. Ottawa Community Improvement Corp., 899 F.2d 474, 483 (6th Cir.1990) (internal quotation marks and citation omitted); see also In re Northwest Airlines, 208 F.R.D. at 195. First, the Court must determine whether Plaintiffs' evidence of conspiracy is ambiguous—that is, whether it is “as consistent with the defendants' permissible independent interests as with an illegal conspiracy.” Riverview Investments, 899 F.2d at 483 (internal quotation marks and citation omitted). If so, the Court must inquire whether Plaintiffs have produced “evidence that tends to exclude the possibility that the defendants were pursuing these independent interests.” 899 F.2d at 483 (internal quotation marks and citation omitted). “A plaintiff thus fails to demonstrate a conspiracy if, using ambiguous evidence, the inference of a conspiracy is less than or equal to an inference of independent action.” 899 F.2d at 483;see also In re Northwest Airlines, 208 F.R.D. at 195–96.

2. There Is No Inflexible Requirement that Plaintiffs Must Produce Evidence of Parallel Conduct in Order to Establish a Conspiracy Through Circumstantial Evidence.

As a threshold matter—and, indeed, as the principal argument Defendantshave advanced against Plaintiffs' per se claim in the brief in support of their combined summary judgment motion—Defendants contend that Plaintiffs “cannot even get to first base” on their claim of a per se violation of § 1 without proof that the Defendant hospitals engaged in parallel conduct in their setting of RN wages. (Defendants' Motion, Br. in Support at 22.) In Defendants' view, the law is clear that “[w]here antitrust plaintiffs rely on circumstantial evidence to prove a Section 1 conspiracy claim, they must first demonstrate that the defendants' actions were parallel.” ( Id. at 21 (quoting In re Beef Industry Antitrust Litigation, 907 F.2d 510, 514 (5th Cir.1990)).) Defendants suggest that Plaintiffs cannot make this threshold showing, citing a report and accompanying graph reflecting the opinion of their economic expert, Dr. Daniel Rubinfeld, that the RN wages paid by the Defendant hospitals “were not similar and did not change in a parallel way.” (Defendants' Motion, Br. in Support at 22–23.) As discussed below, however, while parallel conduct can serve as one form of circumstantial evidence of a § 1 conspiracy to fix prices or wages—and while the absence of parallel conduct might serve to undercut an inference of conspiracy that might otherwise arise from the evidentiary record—the Court does not read the pertinent case law as mandating this particular form of proof in order to sustain a § 1 conspiracy claim.

There is no question that many (perhaps most) § 1 price-fixing claims rest in part upon allegations of parallel conduct. See, e.g., Wallace, 55 F.3d at 1168 (plaintiffs attempted to establish § 1 claim by contending that the defendant banks “all charge essentially the same fee for NSF checks”); In re Flat Glass Antitrust Litigation, 385 F.3d 350, 354–55 (3d Cir.2004) (plaintiffs alleged that “[s]everal times during the class period, [the defendant] flat glass producers raised their ‘list prices' for flat glass by the same amount and within very close time frames”); In re Baby Food Antitrust Litigation, 166 F.3d 112, 122 (3d Cir.1999) (“In an effort to reinforce their claim of collusive pricing, plaintiffs presented the testimony of an expert for the purpose of showing a pattern of parallel price increases in each of the five baby food product categories throughout the certified time period.”); In re Beef Industry, 907 F.2d at 512 (plaintiff cattlemen alleged that the defendant meat packers “violated § 1 of the Sherman Act by using an information exchange ... and consciously parallel pricing to depress live cattle prices”). This stands to reason because, in the absence of direct evidence of an agreement to fix prices, the most likely way that such an arrangement would manifest itself to outsiders—and potential plaintiffs—would be through parallel action that gives rise to a suspicion of unlawful coordination in restraint of trade. See In re Travel Agent Commission Antitrust Litigation, 583 F.3d 896, 903 (6th Cir.2009) (observing that “[a]llegations of concerted action by competitors are frequently based on a pattern of uniform business conduct”).

Nonetheless, this Court is aware of no case law—nor have Defendants identified any—that mandates that a plaintiff's portfolio of circumstantial evidence in a § 1 case must include proof of parallel conduct. To the contrary, in Re/Max International, Inc. v. Realty One, Inc., 173 F.3d 995, 1009 (6th Cir.1999), the Sixth Circuit cited the defendants' “uniform[ity] in their actions” as merely one of several “[i]mportant factors” a court should evaluate in determining whether a plaintiff's circumstantial evidence of antitrust conspiracy is sufficient to withstand summary judgment. Similarly, Plaintiffs point to the decision in InterVest, Inc. v. Bloomberg, L.P., 340 F.3d 144, 163–65 (3d Cir.2003), in which the court first addressed the plaintiff's attempt to establish a conspiracy “on the [b]asis of [c]ircumstantial [e]vidence,” and then, upon finding that this effort failed to produce evidence that would support a “reasonable inference[ ] of concerted action,” proceeded to ask as a separate matter whether the plaintiff had shown that the defendants “engaged in activity that approximates conscious parallelism.” ²⁹

29 Defendants' effort to explain away the InterVest decision is unpersuasive. In Defendants' view, while the court in that case used separate headings in its opinion entitled “Conspiracy on the Basis of Circumstantial Evidence” and “Conspiracy on the Basis of Consciously Parallel Behavior,” InterVest, 340 F.3d at 163, 165, this does not necessarily mean that the court viewed these as “distinct ways” of proving a § 1 agreement. (Defendants' Reply Br. at 20.) Yet, at the very beginning of the second of these two sections of its opinion, the court expressly stated that “[w]ithout any evidence of communication between the [defendants] or other reasonable inferences of concerted action,” the plaintiff was obligated to proceed through a showing of conscious parallelism. InterVest, 340 F.3d at 165. The import of this language seems fairly clear—namely, that a showing of conscious parallelism is only one of the available avenues of proving a § 1 conspiracy by circumstantial means. Here, of course, there is substantial evidence of communications among the Defendant hospitals, which InterVest identifies as an alternative method for raising an inference of concerted action.

Indeed, in another RN wage-fixing case presently pending in the Northern District of New York, the court expressly rejected the very same argument advanced by Defendants here, holding that the plaintiffs in that case “need not prove parallel pricing in order to prevail on [a] per-se claim based on circumstantial evidence.” Fleischman v. Albany Medical Center, 728 F.Supp.2d 130, 158 (N.D.N.Y.2010). As succinctly stated by the court in that case, “[p]arallel pricing is merely one such form of circumstantial evidence” upon which a plaintiff may permissibly rely to establish concerted action in violation of § 1. Fleischman, 728 F.Supp.2d at 158 (internal quotation marks and citation omitted).

The cases cited by Defendants are not to the contrary. In the decision upon which they principally rely for their posited rigid requirement of parallel conduct, In re Beef Industry, Defendants quote only the language that aids their cause— i.e., that a plaintiff in a § 1 conspiracy case “must first demonstrate that the defendants' actions were parallel”—while notably omitting the language that immediately precedes it—namely, that this standard applies to plaintiffs who “rel[y] on circumstantial evidence of conscious parallelism to prove a § 1 claim.” In re Beef Industry, 907 F.2d at 514 (emphasis added). As Plaintiffs aptly note, this “unsurprising statement” by the court is “frankly little more than a tautology—if your theory of the case is based upon parallel conduct then you must show parallel conduct!” (Plaintiffs' Response Br. at 92.) Likewise, Defendants' selective quotation from the decision in City of Moundridge v. Exxon Mobil Corp., 429 F.Supp.2d 117, 132 (D.D.C.2006), notably truncates the court's complete statement that a showing of parallel conduct is “the first element necessary in showing a conspiracy with conscious parallelism. ” (Emphasis added.)

In short, while evidence of parallel conduct provides one means by which a plaintiff in a § 1 case may make a circumstantial showing of concerted activity, the case law establishes no hierarchy of the sorts of circumstantial evidence a plaintiff must produce to withstand summary judgment. Rather, circumstantial evidence of any sort will do, provided that it demonstrates “business behavior which evidences a unity of purpose or a common design and understanding,or a meeting of the minds in an unlawful arrangement.” Wallace, 55 F.3d at 1168 (internal quotation marks and citations omitted). Accordingly, the Court rejects Defendants' contention that Plaintiffs' per se claim fails for lack of evidence of parallel conduct.³⁰

30 In light of this conclusion, the Court need not address Plaintiffs' contention that they have, in fact, presented evidence of parallel conduct.

3. Plaintiffs' Evidence in Support of Their Per Se Claim Fails to Give Rise to an Inference of Conspiracy Stronger Than the Inference of Independent Action.

Having disposed of Defendants' threshold challenge based on the absence of evidence of parallel conduct, the Court now returns to the more general evidentiary standards by which Plaintiffs' per se claim of conspiracy must be judged. As noted earlier, the analysis of this claim is governed by the overarching requirement that Plaintiffs' evidence must “reasonably tend[ ] to prove that [Defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective,” and must further “tend[ ] to exclude the possibility that [Defendants] were acting independently.” Monsanto, 465 U.S. at 764, 104 S.Ct. at 1471.

As discussed at length below, the Court believes that the record evidence and the inferences that may fairly be drawn from this record give rise to a very close question on Plaintiffs' per se claim, and that fair-minded jurists could reasonably disagree as to the resolution of this question. Certainly, Plaintiffs have produced a great deal of evidence which, if fully credited and granted generous inferences, could circumstantially support a finding of concerted, albeit sophisticated, interactions among the Defendant hospitals for the purpose of suppressing RN wages, at least over time. However, after thorough review and careful consideration of this substantial body of evidence—and given the fairly stringent and somewhat elevated standard Plaintiffs must meet to establish their per se claim of conspiracy to fix wages—the Court finds that the inference of independent action arising from the record is at least as strong as the inference of a conspiracy to depress RN wages. It follows that an award of summary judgment to Defendants is warranted on Plaintiffs' per se§ 1 claim.

At the outset of this inquiry, the Court acknowledges Plaintiffs' well-taken point that they need not produce evidence of a formal agreement among the Defendant hospitals in order to prove that these health care institutions engaged in a conspiracy in violation of § 1. See United States v. General Motors Corp., 384 U.S. 127, 142–43, 86 S.Ct. 1321, 1329, 16 L.Ed.2d 415 (1966). As one appellate court has observed, the courts are not “so naive as to believe that a formal signed-and-sealed contract or written resolution would conceivably be adopted at a meeting of price-fixing conspirators in this day and age;” to the contrary, “the typical price-fixing agreement is usually accomplished” in a manner that “can ordinarily only be proved by inferences drawn from relevant and competent circumstantial evidence, including the conduct of the defendants charged.” Esco Corp. v. United States, 340 F.2d 1000, 1006–07 (9th Cir.1965). Moreover, this antitrust suit, in its present summary judgment posture, is subject to the usual rule that Plaintiffs, as the non-moving parties, must be given “the benefit of all reasonable inferences” that may be drawn from the evidence, Smith Wholesale, 477 F.3d at 861, and “the question of what weight should be assigned to competing permissible inferences remains within the province of the fact-finder at trial,” Apex Oil Co. v. DiMauro, 822 F.2d 246, 253 (2d Cir.1987).

The Sixth Circuit has identified four factors that a court should consider in evaluating the circumstantial evidence offered in support of a § 1 conspiracy claim:

(1) whether the defendants' actions, if taken independently, would be contrary to their economic self-interest; (2) whether the defendants have been uniform in their actions; (3) whether the defendants have exchanged or have had the opportunity to exchange information relative to the alleged conspiracy; and (4) whether the defendants have a common motive to conspire.
Re/Max International, 173 F.3d at 1009. Accordingly, the Court addresses each of these factors in turn, as well as other considerations identified by the parties, and then returns to the overarching inquiry whether the entire body of circumstantial evidence produced by Plaintiffs, considered as a whole, “tend[s] to exclude the possibility of independent conduct.” Re/Max International, 173 F.3d at 1009.

(a) Evidence of Actions Taken by the Defendant Hospitals That Are Inconsistent with Their Independent Interests

As the first category of circumstantial evidence offered in support of their per se§ 1 claim, Plaintiffs contend that they have identified myriad instances and ways in which the Defendant hospitals have acted inconsistently with their independent pursuit of their respective self-interests. Upon reviewing the evidence offered by Plaintiffs in support of this proposition, the Court agrees that most (but not all) of it tends to give rise to an inference of conspiracy that has greater force than the opposing inference of independent action.

First and foremost, Plaintiffs naturally point to the substantial body of evidence revealing the Defendant hospitals' regular requests for and exchange of competitively sensitive information regarding the wages paid to their RN workforces, projected wage increases, and the pay philosophies underlying their RN wage determinations. As set forth above in the Court's recitation of the voluminous record, Plaintiffs have produced evidence of Defendants' repeated exchanges—through both direct contacts and third-party surveys—of (i) detailed, current information as to the wages paid to their RN workforces, ( see, e.g., Plaintiffs' Response, Exs. 46E, 46G, 157); (ii) their planned future pay increases, ( see, e.g., Plaintiffs' Response, Exs. 71, 120, 121), and (iii) their overall philosophies and targets in setting RN wages, ( see, e.g., Plaintiffs' Response, Ex. 74). A fair number of these exchanges violated one or more of the criteria set forth in the DOJ/FTC Guidelines, including the recommendations (i) that wage surveys should be managed by a third party, (ii) that survey participants provide information that is more than three months old, and (iii) that the survey data be reported in a “sufficiently aggregated” form “such that it would not allow recipients to identify the ... compensation paid by any particular provider.” DOJ/FTC Guidelines, Statement 6(A) at 63.

As Defendants correctly observe, the “exchange of price data”—or, here, wage data—“among competitors does not invariably have anticompetitive effects,” but may to the contrary “increase economic efficiency and render markets more, rather than less, competitive.” United States v. United States Gypsum Co., 438 U.S. 422, 441 n. 16, 98 S.Ct. 2864, 2875 n. 16, 57 L.Ed.2d 854 (1978); see also Wallace, 55 F.3d at 1169;In re Baby Food, 166 F.3d at 126. In Wallace, 55 F.3d at 1169 & n. 5, for example, the court recognized that the defendant banks “naturally are interested in surveying the market to determine what other banks are charging in order to make strategic competitive decisions,” and that the banks' publication of their fees also served “to advise customers of the charges they must pay.” In addition, the DOJ/FTC Guidelines expressly emphasize that the “safety zones” described therein are not to be taken “as defining the limits of joint conduct that is permissible under the antitrust laws.” DOJ/FTC Guidelines at 6.

Nonetheless, the brute fact remains that any pro-competitive objectives identified by the Defendant hospitals—and, as Plaintiffs observe, Defendants are long on generalities but short on specifics in this regard—could just as well have been achieved through exchanges that fell comfortably within the “safety zone” identified in the DOJ/FTC Guidelines. Plaintiffs state without contradiction that they do not seek to condemn the use of surveys in general, and they point to reputable, nationally recognized firms that “report the results of properly aged and highly aggregated data gathered from enormous numbers of participants.” (Plaintiffs' Response Br. at 28 & n. 28 (citing an example of a Watson Wyatt survey that “had hundreds of participants”).) Yet, rather than relying on such third-party surveys that fully comported with the DOJ/FTC Guidelines, the Defendant hospitals time and again sought and obtained RN wage-related information through direct contacts and surveys that fell short of the DOJ/FTC “safety zone” criteria. Defendants are notably silent as to how it was necessary to their independent competitive interests to proceed in this fashion, rather than confining their exchanges to the “safety zone.”

To the contrary, the manner in which the Defendant hospitals actually exchanged wage information, at least on many occasions, certainly appears to have been inconsistent with not only the DOJ/FTC Guidelines, but also each participating hospital's independent self-interest. There is no dispute that these hospitals compete with each other in the market for RNs. ( See Defendants' Motion, Br. in Support at 12 (acknowledging that “Defendants compete for nurses with each other ....”).) ³¹ Against this backdrop, while it is understandable that each Defendant hospital would like to know what its competitors are paying their RNs, it is far less clear why a hospital would respond to a competitor's request for this information regarding its own RN wages and pay practices. As the Supreme Court observed in the analogous context of information-sharing among sellers, “if one seller offers a price concession for the purpose of winning over one of his competitor's customers, it is unlikely that the same seller will freely inform its competitor of the details of the concession so that it can be promptly matched and diffused.” United States Gypsum, 438 U.S. at 456, 98 S.Ct. at 2883;see also In re Coordinated Pretrial Proceedings in Petroleum Products Antitrust Litigation, 906 F.2d 432, 450 (9th Cir.1990) (noting that the disclosure of “sensitive price information might be considered contrary to a firm's self-interest,” and thus could support a jury's finding of a “common understanding” among the companies sharing this information). Likewise, in this case, if one of the Defendant hospitals sought to recruit a competitor's RNs by raising its own wages,³² it would surely run counter to this purpose to notify the competitor about the increased wage rate. Yet, there is nary an example in the record of a Defendant hospital refusing a request for wage-related information on the ground that it was unwilling to disclose this competitively sensitive data.

31 The parties do dispute, however, whether there is a shortage of hospital nurses in the Detroit area. ( Compare Plaintiffs' Response Br. at 63–65 (citing evidence in support of the proposition that “during the class period, there was a shortage of bedside nurses in Detroit”) with Defendants' Reply Br. at 25 (asserting that Plaintiffs' claim of a nurse shortage rests solely upon the “discredited utilization analysis” of its expert).) The Court finds it unnecessary to weigh in on this subject, in light of its disposition of Plaintiffs' per se§ 1 claim on grounds independent from this question.

32 As Plaintiffs observe, this scenario appears to be more hypothetical than real. In particular, the record fails to disclose any instance in which a Defendant hospital used wage data gathered from its competitors as part of an explicit attempt to “outbid” other hospitals for RNs. Such conduct, of course, would lend considerable support to the claim that Defendants' information exchanges served a pro-competitive purpose. By the same token, the absence of such conduct tends to suggest that the Defendant hospitals were acting inconsistently with their own interests by failing to exploit an advantage provided through their knowledge of their competitors' rates of RN compensation.

Still less clear is what pro-competitive justification exists for exchanging information regarding projected future wage increases, as the Defendant hospitals occasionally did in this case. The court in In re Flat Glass, 385 F.3d at 369, viewed an analogous disclosure of a planned future price increase as “qualitatively different” from the type of information exchange it had deemed insufficient to establish a conspiracy in an earlier case, In re Baby Food. Similarly, the Second Circuit opined in Todd, 275 F.3d at 211, that “exchanges of future price information are considered especially anticompetitive.” See also DOJ/FTC Guidelines, Statement 6(B) at 64 (“Exchanges of ... future compensation of employees are very likely to be considered anticompetitive.”). Along the same lines, the Court is at a loss to identify—and Defendants do not suggest—a pro-competitive justification for the Defendant hospitals' willingness to disclose to each other information regarding their overarching pay philosophies and target rates of RN pay within the local market, as occurred in at least one instance. ( See Plaintiffs' Response, Ex. 74 (December 2004 nursing pay policy survey sponsored by Henry Ford and conducted by Linda Budd).) ³³

33 Plaintiffs observe that during the class period, none of the Defendant hospitals has deviated from the market percentile targets disclosed in this survey.

Certainly, it is telling that when Plaintiffs and their counsel invited various employees of the Defendant hospitals to suggest a pro-competitive justification for their exchange of information with their counterparts at other Detroit-area hospitals, none was able to do so. For example, Defendant Beaumont's director of compensation, David Misner, testified that he had “no idea why we would want to share” RN wage-related information with other hospitals. (Plaintiffs' Response, Ex. LL, Misner Dep. at 221.) Other employees testified that in their past experience working in other industries or parts of the country, they did not directly contact their competitors for wage information. ( See Plaintiffs' Response, Ex. DD, Kontos Dep. at 27–28 (stating that he did not contact competitors when employed in the automotive industry “[b]ecause of the antitrust laws”); Ex. QQ, Perez Dep. at 16–17). Still another employee testified that her employer, Defendant Trinity, shared its wage data in third-party surveys “in order to receive the results of the survey.” (Plaintiffs' Response, Ex. D, Barcome Dep. at 81.) As Plaintiffs observe, this sort of quid pro quo understanding supports an inference of concerted action rather than independent conduct.

Moreover, even to the extent that the Defendant hospitals utilized third-party surveys rather than direct contacts to obtain market wage data—thereby employing a practice that was at least potentially more compliant with the DOJ/FTC Guidelines and less suggestive of collusive action—Plaintiffs properly point to the “pretextual” nature of at least some of these surveys. In particular, several such surveys were conducted by Linda Budd, a former human resources employee at Henry Ford, who carried out the surveys and reported the results in a manner that Defendants could not have believed consistent with the truly independent third-party surveys contemplated by the Guidelines. As noted earlier, Ms. Budd occasionally permitted sponsoring hospitals to design their own surveys, decide which institutions to survey, and determine how the results should be presented. In addition, she initially reported her survey results with the participating hospitals identified by name, and even when she switched to letter codes, she continued to provide disaggregated results to the sponsoring institution (and sometimes to participating hospitals as well), and her letter codes occasionally were easily deciphered (or she herself disclosed the key). Under this record, just as there seemingly was no pro-competitive justification for the Defendant hospitals to disclose their wage data directly to their competitors, it is difficult to see why these hospitals thought it was in their independent self-interest to disclose this data to Ms. Budd, where competitors could so readily determine its source.

Defendants' various efforts to downplay the significance of this substantial body of evidence of competitively suspect information exchanges are unavailing. First, although Defendants note that most of the direct hospital-to-hospital contacts involved lower-level employees, they do not seriously contend that these employees were acting on their own or against the wishes of their superiors.³⁴ To the contrary, the higher-level decisionmakers at each of the Defendant hospitals have acknowledged that they considered market wage data in making RN wage determinations, and the record shows that at least some of this data was acquired through direct contacts, particularly in the early portion of the class period. Similarly, to the extent Defendants contend that such contacts were only “sporadic,” Plaintiffs have managed to compile quite a record of these “sporadic” contacts. In any event, many of these contacts merely filled the gaps between “numerous third party surveys,” which occurred so frequently that they “in effect, constantly provided [the Defendant hospitals] with current information concerning each others['] pay rates.” (Plaintiffs' Response Br. at 34–35 (recounting the number and frequency of surveys undertaken during the class period).)

34 As Plaintiffs observe, given the time and effort involved in responding to some of the more detailed surveys, ( see Plaintiffs' Response Br. at 12–13 n. 16) (citing evidence of the time spent by employees of the Defendant hospitals to complete surveys), it is highly doubtful that a lower-level employee would have undertaken this task unless directed to do so, and it is also doubtful that a hospital's senior management would look the other way or remain ignorant as its compensation staff engaged in this task.

Defendants' remaining quibbles with Plaintiffs' characterization of their information exchanges warrant little discussion. While Defendants point to their occasional use of masked and/or aggregated survey results, there is ample evidence of survey results reported to the Defendant hospitals in unmasked and/or disaggregated form. Likewise, although one or more of the Defendant hospitals at times commissioned surveys conducted by truly independent third parties— e.g., Defendant Beaumont's use of Sullivan Cotter after October of 2003—this does not require that the Court (or the trier of fact) disregard the significant evidence of direct contacts or the use of “third parties” in name only, such as Ms. Budd. Finally, while Defendants fault Plaintiffs for their occasional citation to information exchanges that occurred prior to the class period, Plaintiffs correctly observe that these exchanges may form the basis for a recovery, so long as it can be shown that they affected wage determinations made during the class period. See Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 338, 91 S.Ct. 795, 806, 28 L.Ed.2d 77 (1971). In any event, the record plainly is not confined to information exchanges that occurred prior to the commencement of the class period in December of 2002.

In sum, the record of information sharing in this case provides considerable evidence of conduct by the Defendant hospitals that was inconsistent with the independent pursuit of their own interests. Both the nature and extent of these information exchanges and the manner in which they were undertaken serve to distinguish this case from the decisions cited by Defendants, in which the courts found that exchanges of information were at least as consistent with the defendants' individual interests as with a shared design or purpose. In Wallace, 55 F.3d at 1169, for example, the Sixth Circuit held that the defendant banks' publication of their NSF fees served “a legitimate interest in providing customers relevant information regarding the fees associated with their checking accounts,” and nothing in that case suggests that the banks surveyed the fees charged by their competitors through methods that would have fallen outside the DOJ/FTC “safety zone”—or, for that matter, that they gathered anything beyond publicly available information.

The other case principally relied upon by Defendants, In re Baby Food, is also distinguishable. The court in that case found that much of the information sharing evidence produced by the plaintiffs amounted to nothing more than “sporadic exchanges of shop talk among field sales representatives who lack pricing authority.” In re Baby Food, 166 F.3d at 125. Although Defendants here make a similar effort to downplay purportedly “sporadic” exchanges of wage-related information among “low-level” employees, the Court already has explained that the record of Defendants' sharing of this information was hardly “sporadic,” but instead consisted of fairly regular surveys augmented by direct contacts. In addition, it is important to note that the “low-level” employees here generally worked in the Defendant hospitals' compensation departments, and were expressly charged with the tasks of gathering competitive compensation information and recommending wage increases. While these employees may have lacked ultimate wage-setting authority, the evidence shows that they were participants in the process leading up to the decisions made by those who possessed this authority, and that at least some of the data they gathered played a role in this process. Thus, it cannot be so readily said here, as in In re Baby Food, 166 F.3d at 125, that there is no “evidence that the exchange of information had an impact on pricing decisions.” ³⁵

35 The Court returns below to a more detailed assessment of the evidence produced by Plaintiffs that purports to establish the impact of the Defendant hospitals' information exchanges on their wage-setting decisions.

Next, while Plaintiffs' claims of conduct inconsistent with the Defendant hospitals' independent interests rest principally upon Defendants' willingness to share wage-relatedinformation, Plaintiffs further contend that Defendants acted contrary to their self-interests by purportedly agreeing to refrain from “active recruitment [of RNs] from other defendant hospitals, even when it [wa]s known that certain hospitals [we]re paying their nurses low wages.” (Plaintiffs' Response Br. at 80.) Yet, the one isolated example of this conduct cited by Plaintiffs does not support an inference of conspiracy rather than independent conduct. Specifically, Plaintiffs point to an occasion in 2003 when Oakwood and St. John agreed, following complaints from DMC, to abandon recruitment efforts directly targeted at DMC nurses. At the time, DMC was experiencing a severe financial crisis, and it contemplated closing two of its facilities, Detroit Receiving Hospital and Hutzel Women's Hospital, which resulted in “a number of nurses leaving or considering leaving” DMC's employ. (DMC's Motion, Ex. 2, Natale Decl. at ¶¶ 16, 18.) The evidence produced by Plaintiffs indicates that St. John and Oakwood initially viewed these developments as an opportunity to recruit DMC nurses, but that they backed off these efforts following complaints from DMC officials. ( See Plaintiffs' Response, Ex. 34.)

The Court does not view this isolated incident as evidencing any sort of broader agreement among the Defendant hospitals not to recruit each others' RNs. As revealed in one of the documents produced by Plaintiffs regarding this episode, two high-level St. John officials—the hospital's corporate director of workforce planning, Jim Flanegin, and its senior vice president of human resources, Dan Zuhlke—discussed DMC's complaints about St. John's recruiting effort and “came to the conclusion that at this time it is better for all of us in the long run that we support the DMC rather than attack them.” (Plaintiffs' Response, Ex. 34A.) The undisputed record shows that DMC operates several “safety-net” hospitals in the City of Detroit, and that it provides “the majority of care to the indigent, uninsured, and underinsured citizens of Detroit,” totaling “approximately $250 million of uncompensated care annually.” (Defendants' Motion, Ex. K.1, Duggan Decl. at ¶¶ 5, 8.)

Under these circumstances, Plaintiffs' claim of conduct against a hospital's self-interest cannot be evaluated solely by reference to the willingness of a competitor to “raid” DMC's RN workforce. Rather, the record indicates that other factors played a role in Oakwood's and St. John's decisions to “stand down” from recruiting DMC nurses. The evidence suggests, in particular, that it would be consistent with the other Defendant hospitals' self-interests to ensure that DMC's safety-net facilities remained available to serve indigent and uninsured patients who would otherwise seek uncompensated care at their facilities. Indeed, it is clear from the evidence produced by Plaintiffs that the other Defendant hospitals did not altogether abandon their efforts to recruit DMC nurses, but instead continued to employ “other [recruitment] resources” in lieu of direct, targeted appeals to DMC's RN workforce. ( See Plaintiffs' Response, Ex. 34B, 34D.) This incident, then, does not serve as an additional example of the Defendant hospitals acting contrary to their own self-interests.

(b) Evidence of Uniformity in Defendants' Actions

The next factor to consider in the Court's evaluation of Plaintiffs' circumstantial evidence of conspiracy is “whether the defendants have been uniform in their actions.” Re/Max International, 173 F.3d at 1009. In one respect, it can readily be said that the Defendant hospitals were uniform in their actions—namely, in their willingness to provide competitively sensitivecompensation-related information to other Detroit-area hospitals, whether through direct contacts or third-party surveys.

Yet, beyond this broad level of generality, the evidence of uniform conduct is far more mixed. As observed by Defendants—and as noted in the report of their expert, Daniel Rubinfeld—there was not “uniform circulation, participation in, or receipt of surveys” by the Defendant hospitals, and Defendants “used surveys in different ways and to different degrees in their compensation setting.” (Defendants' Motion, Ex. G, Rubinfeld Report at ¶ 16.) As examples of these disparate practices, Defendants point out that Trinity did not commission any surveys during the class period, and that DMC commissioned only one during this time frame, while St. John asked Linda Budd to conduct annual surveys and Beaumont used semi-annual surveys conducted by Sullivan Cotter.³⁶ Given the expense of these surveys, ( see Plaintiffs' Response Br. at 12 n. 15 (citing evidence that the surveys cost “anywhere from a few hundred dollars to well over $1000 per survey”)), it surely would have behooved Defendants to pool their resources in furtherance of any purported scheme to depress RN wages, yet they did not do so. In addition, Defendants differed in their use of direct contacts to obtain wage-related information from other Detroit-area hospitals—as observed earlier, Henry Ford largely ceased this practice in 2003, while Beaumont continued to permit such direct contacts as warranted by the particular circumstances ( e.g., when seeking information for only a single job).

36 In addition, Mount Clemens operated on an altogether different schedule, seeking survey data in advance of its union negotiations approximately every three years.

Next, Defendants point out that Plaintiffs have not shown “that any two hospitals utilized the same set of surveys and information about the market to set their RN compensation.” (Defendants' Motion, Br. in Support at 15.) For example, in the instances noted by Plaintiffs where Beaumont cited market data to reduce the magnitude of wage increases—or, on one occasion, to determine that a previously anticipated pay increase would not be given—the record indicates that Beaumont relied exclusively on the surveys it had sponsored or conducted. ( See Plaintiffs' Response, Exs. 99, 104.) More generally, the record is bereft of evidence of any common pool of market data being used by two or more Defendant hospitals in making their RN compensation decisions. To be sure, and as this Court has previously recognized, “minor variances in the particular means ... chosen” by each Defendant hospital to set its RN wages do not defeat an inference of conspiracy, because “an agreement need not dictate every conceivable aspect of each conspirator's behavior in order to violate § 1.” In re Northwest Airlines, 208 F.R.D. at 199. Yet, the inference of conspiracy Plaintiffs seek to draw here is certainly weakened by the evidence that each Defendant hospital gathered and used its own unique set of market information as it carried out its separate wage analysis and determined its own rates of RN compensation.

Finally, and most significantly, the record reflects a decided lack of uniformity in the end results of each Defendant hospital's wage-setting process. Defendants' expert, Dr. Rubinfeld, has concluded upon his analysis of the underlying wage data that “there are substantial differences in the rate of change, levels, and forms of compensation among the Defendant hospitals.” (Defendants' Motion, Ex. G, Rubinfeld Report at ¶ 17.) To emphasize this point, Defendants cite two graphs preparedby Dr. Rubinfeld illustrating the different annual rates of change in the wages paid by the Defendant hospitals to their RN workforces. ( See Defendants' Motion, Br. in Support at 23, 24.) The first of these graphs shows that in 2005–06, for example, the annual change in mean total hourly RN wages ranged from 1.2 percent (for Mount Clemens) to 10.0 percent (for Trinity's St. Joseph–Macomb Hospital). ( See Rubinfeld Report, App. G at 5.) ³⁷ The second graph, which maps the annual change in mean base wages for RNs, shows that in 2005–06, for example, the annual increases ranged from 2.05 percent (for Mount Clemens) to 10.23 percent (for Trinity's St. Mary–Livonia Hospital). ( See id., App. G at 39.) ³⁸ In general, it is difficult to look at these graphs (or the underlying data from which they are derived) and discern a strong pattern or clear-cut uniformity in the wage-setting decisions made by the Defendant hospitals during the class period.

37 If one compares the underlying total hourly wage rates paid by the Defendant hospitals, rather than the annual rates of change, the rates for 2006 range from $30.64 (for one of the Henry Ford hospitals) to $38.66 (for one of the Oakwood facilities). ( See id., App. G at 16.)

38 Again, looking at the underlying mean base wages paid by the Defendant hospitals, the hourly rates for 2006 range from $27.80 (at one of the Henry Ford facilities) to $35.00 (at one of the Oakwood locations). ( See id., App. G at 48.)

To be sure, Plaintiffs seek to refute this apparent evidence of disparate wage-setting decisions by pointing to a graph prepared by their expert, Orley Ashenfelter, that Plaintiffs view as “demonstrat[ing] parallel compensation behavior among the Defendants.” (Plaintiffs' Response Br. at 93 (citing Defendants' Motion, Ex. C, Ashenfelter Report, Fig. 1).) In particular, this graph purports to show that the mean wage rates paid by the Defendant hospitals during the class period followed roughly similar trajectories. Yet, the underlying data reflected in this graph is the very same data relied upon by Defendants' expert, Dr. Rubinfeld, in the graphs reproduced in Defendants' brief. Dr. Rubinfeld's more specific year-by-year, hospital-by-hospital analysis of the underlying data reveals the key point obscured in Dr. Ashenfelter's much more general presentation of this data—namely, that while there might have been an overall, fairly modest upward trend in the wage rates paid by each of the Defendant hospitals during the class period, this overall trend was formed through individual, periodic wage determinations that varied significantly from one hospital to another (and from one year to the next within the same hospital system). Under analogous circumstances, the court in In re Baby Food, 166 F.3d at 129, discounted an expert's use of “trend lines” as “revealing nothing more than that the transaction prices tended to increase over time.” Accordingly, the Court finds that the evidence of the Defendant hospitals' uniformity of action—or lack thereof—tends to favor an inference of independent rather than coordinated conduct.

(c) Evidence of Defendants' Exchange of Information Relative to Their Alleged Conspiracy

The next factor identified by the court in Re/Max International, 173 F.3d at 1009, as relevant to the evaluation of circumstantial evidence of a § 1 conspiracy is “whether the defendants have exchanged or have had the opportunity to exchange information relative to the alleged conspiracy.” As already discussed, there is ample evidence in this case of the exchange of compensation-related information that would have enabled the Defendant hospitals to monitor each others' wage determinations and thereby sustain the wage-fixing conspiracy alleged by Plaintiffs.

To be sure, Defendants correctly observe that “[t]he exchange of [wage] data and other information among competitors does not invariably have anticompetitive effects,” but rather may “in certain circumstances increase economic efficiency and render markets more, rather than less, competitive,” United States Gypsum, 438 U.S. at 441 n. 16, 98 S.Ct. at 2875 n. 16;see also Continental Cablevision of Ohio, Inc. v. American Electric Power Co., 715 F.2d 1115, 1119 (6th Cir.1983) (“[I]n the absence of a purpose or effect to restrain competition, or some other evidence of an actual agreement to restrain competition, we hold that the exchange of price data does not offend § 1 of the Sherman Act.”). Yet, as previously noted, at least some of the information exchanges that occurred here trigger heightened concerns of anticompetitive impact. In particular, some of these exchanges involved disaggregated or unmasked (or both) competitor wage data, some occurred through direct contacts between competitors, and some involved current wage information or, in some cases, even future projected wage increases. Moreover, Defendants have failed to suggest why information exchanges that lacked these troublesome features, and that instead remained within the “safety zone” articulated in the DOJ/FTC Guidelines, would not have sufficed to achieve the business objectives they have identified as driving their compensation decisions. Under these circumstances, this factor tends to support the inference that the Defendant hospitals were engaged in an unlawful conspiracy, as opposed to pursuing their own independent interests.

(d) Evidence of Defendants' Common Motive to Conspire

The fourth and final factor identified by the Sixth Circuit in Re/Max International, 173 F.3d at 1009, is “whether the defendants have a common motive to conspire.” As a general matter, such a common motive is clear in this case, where Plaintiffs state without contradiction that RN wages are the largest single item in the Defendant hospitals' budgets. More generally, Defendants acknowledge that “[a]ll companies have an incentive to lower input prices.” (Defendants' Reply Br. at 32.) Yet, as Defendants point out, the evidence of common motive here is not especially compelling, given that “[p]rofit is always a motivating factor in the conduct of a business.” In re Baby Food, 166 F.3d at 134. Thus, while this factor perhaps tilts slightly in favor of Plaintiffs' claim of conspiracy, it cannot be said that it has a particularly strong “tend[ency] to exclude the possibility of independent conduct.” Re/Max International, 173 F.3d at 1009.

(e) Other Forms of Circumstantial Evidence Identified by the Courts as Indicative of an Unlawful Conspiracy

Beyond the four factors cited in Re/Max International, Plaintiffs address a number of other factors that the courts have considered in analyzing circumstantial evidence of a § 1 conspiracy.³⁹ First, Plaintiffs address the factor of plausibility, which can alter the evidentiary burden they must shoulder to give rise to a permissible inference of collusive action. As the Supreme Court has explained, “if the factual context renders [the plaintiffs'] claim implausible—if the claim is one that simply makes no economic sense—[the plaintiffs] must come forward with more persuasive evidence to support their claim than would otherwise be necessary.” Matsushita, 475 U.S. at 587, 106 S.Ct. at 1356;see also In re High Fructose Corn Syrup Antitrust Litigation, 295 F.3d 651, 661 (7th Cir.2002) ( “More evidence is required the less plausible the charge of collusive conduct.”). Here, however, where the motive to conspire is straightforward and fairly evident, and where Plaintiffs have alleged a “garden-variety [wage]-fixing conspiracy” among a number of competitors who stand on relatively equal footing, Plaintiffs' per se claim “involves no implausibility,” In re High Fructose Corn Syrup, 295 F.3d at 661, and they need not make the “more persuasive” showing that Matsushita demands when a claim “simply makes no economic sense.”

39 As noted by Plaintiffs, the courts sometimes refer to these considerations as “plus factors” which, when considered along with evidence of parallel conduct, can give rise to an inference of conspiracy sufficient to withstand summary judgment. See, e.g., Wallace, 55 F.3d at 1168.

Next, Plaintiffs point to certain structural characteristics of the relevant market which, in their view, would make a wage-fixing conspiracy feasible, including the high “switching costs” associated with a nurse changing jobs, the low elasticity of the supply of RNs, and the frequent sharing of wage information that would enable a quick response to any “cheating” by a co-conspirator. Defendants, in contrast, question the factual basis for the claims of Plaintiffs and their experts concerning high switching costs and low elasticity of supply. More importantly, Defendants contend that the characteristics of the relevant market tend to undercut rather than support an inference of conspiracy, where this market is not highly concentrated, the RN workforce is heterogenous, and the Defendant hospitals likewise are heterogenous purchasers of nursing services, with each hospital having different needs for these services that varied over time. Defendants further cite their different pay structures, their distinct timetables for wage decisions, and the disparate components of each Defendant hospital's RN compensation package as making it difficult to coordinate compensation decisions and depress the wages of RNs across-the-board, since information sharing and collusion as to one component of RN wages, such as the base wage rate, would not necessarily prevent effective competition as to other aspects of the overall nurse compensation package.

Plaintiffs next assert that the inference of conspiracy is strengthened by the evidence of opportunities to meet and collude, where the record discloses meetings attended by the chief executive officers of the four largest Defendant hospital systems—DMC, Henry Ford, St. John, and Oakwood—and where other health care industry groups provided forums to meet and discuss compensation-related information and to identify contacts from whom to obtain this information. In response, however, Defendants note the absence of evidence that their chief executive officers actually discussed any issues concerning nurse pay at their face-to-face meetings—to the contrary, those who have testified have denied that any such discussions occurred. Similarly, there is no evidence that the discussions among lower-level employees at other health care industry meetings had any impact on any Defendant's subsequent wage determinations.

Finally, Plaintiffs argue that further support for an inference of conspiracy can be found in Defendants' apparently knowing transgressions of the “safety zone” criteria set forth in the DOJ/FTC Guidelines. As recounted earlier, the record discloses that during the course of the class period, nearly every Defendant hospital became aware, if it was not previously aware, that direct contacts to obtain a competitor's wage data did not comport with the Guidelines, and that some of the third-party surveys in which they participated also failed to comply with one or more of the Guidelines' “safety zone” criteria. This awareness, however, did not lead to full compliance; instead, employees in Defendants' compensation departments continued to exchange wage information, albeit perhaps on a more sporadic and ad hoc basis. As Plaintiffs observe, such willful disregard for the DOJ/FTC Guidelines lends at least some support to the inference that Defendants were engaged in a conspiracy to depress RN wages, since it evidences a willingness to engage in information-sharing activity that has been identified in the Guidelines as having potentially anticompetitive effects.

(f) Plaintiffs' Circumstantial Evidence Viewed in Its Totality

Having now addressed each of the several categories of circumstantial evidence offered by Plaintiffs in support of their per se§ 1 claim, the Court now considers whether this body of evidence, viewed as a whole, meets the rather stringent standard governing § 1 claims. As noted, the Supreme Court has instructed that this evidence, viewed in a light most favorable to Plaintiffs, must “reasonably tend[ ] to prove that [Defendants] had a conscious commitment to a common scheme designed to achieve an unlawful objective,” and must further “tend[ ] to exclude the possibility that [Defendants] were acting independently.” Monsanto, 465 U.S. at 764, 104 S.Ct. at 1471. The record produced by Plaintiffs, in other words, must give rise to an inference of conspiracy that is stronger than the competing inference of independent action. Riverview Investments, 899 F.2d at 483.

As previously indicated, the record here makes this question indeed a close one, but the Court ultimately concludes that Plaintiffs have not met this standard. Out of the four factors identified by the Sixth Circuit as especially pertinent to this inquiry, see Re/Max International, 173 F.3d at 1009, the one that most strongly supports an inference of conspiracy is the evidence of actions taken by the Defendant hospitals that seemingly ran counter to their independent interests—most notably, their willingness to regularly share competitively sensitive wage data. On the other hand, the factor that most significantly undercuts this inference is the evidence of the lack of uniformity in the Defendant hospitals' wage-setting decisions. These two factors, in the Court's view, are largely determinative of the question whether Plaintiffs' per se§ 1 claim can withstand summary judgment.

Upon juxtaposing the evidence concerning these two most critical factors, the Court cannot say that the inference of conspiracy is stronger than the inference of independent action. The record leaves no room for doubt that each Defendant hospital was highly interested in learning about the wage rates and pay practices of other Detroit-area hospitals as it went about the process of establishing its own RN compensation package. Indeed, it was necessary for Defendants to obtain information about the prevailing RN wage rates in the Detroit-area market, given each Defendant hospital's acknowledgment that it set a target rate of RN compensation by reference to this prevailing market rate— e.g., Defendant Beaumont set a target at the 75th percentile of the highest rate paid in the market, while Defendant Henry Ford sought to pay its RNs at the 66th percentile of this market. All are agreed that Defendants would not run afoul of § 1 if they independently pursued these wage-setting objectives by resort to information obtained in accordance with the “safety zone” criteria set forth in the DOJ/FTC Guidelines. Yet, the record reveals numerous examples of wage-related information exchanges that failed to satisfy these criteria.

As discussed earlier, the Court readily agrees with Plaintiffs that these repeated exchanges of RN wage data outside the Guidelines' “safety zone” could permissibly be viewed by the trier of fact as running counter to the Defendant hospitals' independent interests. Nonetheless, under the totality of the record presented here, the Court finds that this evidence of conduct contrary to Defendants' self-interests gives rise to only a modest inference of conspiracy. The record discloses a number of explanations for Defendants' exchange of wage-related data outside the Guidelines' “safety zone,” including: (i) the continued adherence to historical practice, such as Beaumont's quarterly survey of nurse compensation from 1989 until the fall of 2003; (ii) quid pro quo understandings that if a given request for information was honored, the recipient hospital was more likely to provide its own information if requested at a later date; (iii) expediency, with compensation personnel determining in particular cases that additional information would be desirable regarding a specific position or component of an overall compensation package in advance of an immediately forthcoming wage-setting decision; and (iv) individual ad hoc decisions to provide information, such as Linda Budd's provision of disaggregated survey data to her “alma mater,” Henry Ford. What the record does not disclose, however, is any shared understanding or agreement that these exchanges of data were to be used to ensure uniformity among the Defendant hospitals in their RN wage-setting practices, rather than as part of each Defendant hospital's independent process of establishing its own RN compensation package.

To be sure, Plaintiffs point out that they need not produce “smoking gun” evidence of such a shared understanding or agreement, and that it is enough if the record gives rise to an inference that Defendants were operating under such a coordinated arrangement. Any such inference, however, must overcome the competing inference that arises from the evidence of the Defendant hospitals' actual use of the wage-related information they exchanged among themselves during the relevant period. As discussed earlier in the Court's survey of the evidence concerning uniformity (or non-uniformity) in Defendants' actions, this wage data was compiled and put to use in a series of discrete, individualized RN compensation decisions, with Defendants' decisionmaking processes varying in (i) their degree of sponsorship of and participation in surveys, (ii) their reliance on direct contacts to obtain wage-related information, and (iii) their selection of surveys and market information to consider in setting RN compensation. Moreover, the outcomes of these discrete decisionmaking processes do not reveal any discernible pattern of coordinated wage-setting, but instead reflect material differences in the rates at which each Defendant hospital's RN wages changed from year to year. In short, the record of each Defendant's RN wage-setting is sufficiently individualized and idiosyncratic to undermine any reasonable inference that the Defendant hospitals acted with a unity of purpose.

In this limited respect, then, the Court recognizes the force of Defendants' point that Plaintiffs' per se§ 1 claim is seriously weakened, albeit not outright defeated, by the absence of evidence of parallel conduct. As discussed earlier, the Court does not read the relevant case law as mandating a showing of parallel conduct in order to sustain a § 1 conspiracy claim. Yet, under the record presented here, it surely is more difficult to arrive at a reasonable inference of collective wage-fixing if the Defendants hospitals did not, in fact, engage in parallel conduct or reach parallel results in their compensation decisions for their respective RN workforces. Although Defendants need not have marched in lock-step in their gathering and use of competitor wage data and in their setting of RN wages, the evidence here simply features too wide a disparity among the Defendant hospitals' processes for determining RN compensation and the outcomes of these processes to support a reasonable inference of a conspiracy in violation of § 1.

In the end, Plaintiffs' per se§ 1 claim rests principally on an admittedly significant body of evidence of regular sharing of wage-related information that seemingly ran counter to the Defendant hospitals' independent self-interests, and arguably undermined their ability to compete against each other in their formulation of RN compensation packages. Under this record, it is fair to say that Defendants could have confined their information exchanges to the “safe harbor” of the DOJ/FTC Guidelines, and thereby achieved all of their legitimate, pro-competitive objectives without disclosing unmasked, competitively sensitive wage data to other local hospitals. Even accepting, however, that Defendants' information exchanges were sub-optimal from an economic and competitive standpoint, the courts have cautioned against “second-guessing” of business judgments, “even where the evidence concerning the rationality of the challenged activities might be subject to reasonable dispute.” In re Citric Acid Litigation, 191 F.3d 1090, 1101 (9th Cir.1999). The record here fails to establish that Defendants traveled beyond the realm of independent (albeit economically questionable) business practices and into the impermissible territory of coordinated wage-fixing, such that the inference of conspiracy is stronger than the competing inference of independent action. It follows that Defendants are entitled to summary judgment in their favor on Plaintiffs' Count I claim of a per se§ 1 violation.

C. Plaintiffs Have Produced Sufficient Evidence to Withstand Summary Judgment as to the Causal Element of Their “Rule of Reason” Claim.

The next challenge advanced in Defendants' joint summary judgment motion is directed at both of Plaintiffs' claims— i.e., their Count I claim of a per se antitrust violation through an agreement to fix wages, and their Count II “rule of reason” claim that the Defendant hospitals conspired among themselves to exchange wage information in a manner that harmed competition by depressing RN wages.⁴⁰ Defendants point out—and Plaintiffs do not dispute—that under either of these theories, “Plaintiffs must prove antitrust injury, which is to say injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants' acts unlawful.” Brunswick Corp. v. Pueblo Bowl–O–Mat, Inc., 429 U.S. 477, 489, 97 S.Ct. 690, 697, 50 L.Ed.2d 701 (1977); see also Atlantic Richfield Co. v. USA Petroleum Co., 495 U.S. 328, 341–45, 110 S.Ct. 1884, 1893–95, 109 L.Ed.2d 333 (1990) (confirming that a private plaintiff must make this same showing of antitrust injury— i.e., that the plaintiff has been “adversely affected by an anticompetitive aspect of the defendant's conduct”—in order to recover damages under a per se theory). The Sixth Circuit has emphasized that this “causal link” between an antitrust violation and antitrust injury “must be proved as a matter of fact and with a fair degree of certainty.” Shreve Equipment, Inc. v. Clay Equipment Corp., 650 F.2d 101, 105 (6th Cir.1981). Defendants maintain that Plaintiffs have failed as a matter of law to make this showing. Although the question again is a close one, here the Court disagrees.

40 In light of the Court's decision to grant summary judgment in Defendants' favor on Plaintiffs' per se claim, it is not necessary to address any additional challenges to this claim. Nonetheless, because Defendants' present challenge as to the element of causation applies equally to Plaintiffs' per se and “rule of reason” claims, and because (as will be seen) the analysis is precisely the same as to either of these theories of recovery, the discussion that follows need not be strictly confined to Plaintiffs' “rule of reason” claim.

It is clear, as a threshold matter, that Plaintiffs have not pursued one of the possible avenues for establishing a causal link between an antitrust violation and antitrust injury— i.e., proof through expert testimony. One of Plaintiffs' experts, Dr. Ashenfelter, testified that he made no effort to draw any conclusions as to “whether a particular information exchange led to a suppression of wages or not,” that he had not “actually used the [information] exchanges to study whether—what the effect of the exchange was on wages,” and that he had not “made that connection.” (Defendants' Motion, Ex. Z, Ashenfelter Dep. at 26.) Plaintiffs' other expert, Gregory Vistnes, similarly testified that he had not “attempted to make a direct causal link between the information exchange and the resulting conduct of the hospitals,” that he “was not looking at whether or not there were actual effects from any of the communications,” and that the “magnitude of the effect [of the information exchanges] and whether the effect left wages above or below what the competitive price or wage should have been, was not an issue that I was looking at.” (Defendants' Motion, Ex. AA, Vistnes Dep. at 213, 240, 252.)

Next, it is equally clear that the requisite causal link cannot be inferred solely from the fact that Defendants exchanged wage data, even if—as the evidence indicates—a number of these exchanges fell outside the “safety zone” articulated in the DOJ/FTC Guidelines. As noted earlier, “[t]he exchange of [wage] data and other information among competitors does not invariably have anticompetitive effects.” United States Gypsum, 438 U.S. at 441 n. 16, 98 S.Ct. at 2875 n. 16. The Guidelines themselves reiterate this point:

Participation by competing providers in surveys of prices for health care services, or surveys of salaries, wages or benefits of personnel, does not necessarily raise antitrust concerns. In fact, such surveys can have significant benefits for health care consumers. Providers can use information derived from price and compensation surveys to price their services more competitively and to offer compensation that attracts highly qualified personnel.
DOJ/FTC Guidelines, Statement 6 at 61. Likewise, Plaintiffs' expert, Dr. Vistnes, testified that “as a general matter an information exchange in this type of a context could be pro[ ] competitive and lead to ... either [an] increase or [a] reduction in wages,” and he further recognized that “a fact-based investigation of any case could show that conditions falling outside those [DOJ/FTC] safety zones are unlikely to cause competitive harm.” (Defendants' Motion, Ex. AA, Vistnes Dep. at 200; see also Ex. Z, Ashenfelter Dep. at 34–35 (agreeing that in the absence of collusion, information exchanges can be procompetitive).)

Nonetheless, Plaintiffs correctly note that they may forge the necessary link between Defendants' alleged antitrust violations and antitrust injury without expert testimony, and they contend that the record in this case is sufficient for the trier of fact to infer such a connection. As to the starting point of this causation analysis—the antitrust violations—Plaintiffs have, of course, alleged violations of two sorts: (i) a conspiracy to depress RN wages, and (ii) a conspiracy to exchange wage-related information in a manner that harmed competition by depressing RN wages. As is evident from the descriptions of these two alleged violations, these claims rest upon a common theory of antitrust injury—namely, that the wages paid by the Defendant hospitals to their RN workforces have been held below the competitive level that would prevail in the absence of Defendants' antitrust violations.⁴¹ The question, then, is whether the evidentiary record would permit the trier of fact to infer a causal connection between the two sorts of antitrust violations alleged by Plaintiffs and the antitrust injury identified by their expert. As discussed below, while there are obstacles in the evidentiary paths by which Plaintiffs seek to traverse the gap between a claimed antitrust violation and antitrust injury, the Court nonetheless concludes that the record is sufficient to allow the trier of fact to hear Plaintiffs' evidence and decide the issue of causation.

41 This theory of antitrust injury, in turn, rests upon a “benchmark” analysis performed by Plaintiffs' expert, Dr. Ashenfelter, who has computed a true “competitive” wage for RNs by reference to the fees paid by the Defendant hospitals for agency nurses. As a separate basis for an award of summary judgment in their favor, Defendants challenge Dr. Ashenfelter's agency fee benchmark analysis on a number of grounds. The Court, however, views this subject as more amenable to resolution in the context of Defendants' separate motion to exclude the expert testimony of Dr. Ashenfelter, which raises the same arguments (among others). Accordingly, the Court declines to address Defendants' challenges to Dr. Ashenfelter's expert analysis in the present opinion, but instead will address these matters (as necessary) in a later ruling.

First, Plaintiffs contend that the record in this case discloses a practice among the Defendant hospitals of regular exchanges of wage data “on demand,” exchanges of a nature and frequency that both the courts and Plaintiffs' expert, Dr. Vistnes, have deemed likely to result in depressed, sub-competitive wages. The Court already has recounted the substantial evidence of the Defendant hospitals' regular exchanges of unmasked and disaggregated compensation-related information, both through direct contacts and through third-party surveys that failed to comport with the DOJ/FTC Guidelines in one or more respects. What is more, Plaintiffs have produced evidence—also summarized earlier—that this wage data, while often gathered by lower-level employees, was routinely passed up to the hospitals' executive leadership for use in making wage determinations. Indeed, as Plaintiffs aptly observe, “[i]t could hardly be otherwise,” where “[t]he hospitals would hardly have spent thousands of dollars a year to commission surveys and devote[d] extensive resources to completing these surveys if they never used them,” and where “hospital executives would not have ordered employees to spend time contacting other hospitals to gather information that was not to be used.” (Plaintiffs' Response Br. at 95 (citations omitted).) Thus, while Defendants take issue with Plaintiffs' characterization of the evidence as establishing a “policy of on-demand information exchanges,” ( id. at 99), the Court finds that the record would permit a trier of fact to reach this conclusion.

Against this evidentiary backdrop, Plaintiffs point to case law which, in their view, recognizes the reasonableness of an inference that such “on-demand” exchanges of wage information will result in depressed, sub-competitive wages. First, Plaintiffs note the Supreme Court's recognition in United States Gypsum, 438 U.S. at 457, 98 S.Ct. at 2883, that “[r]egardless of its putative purpose, the most likely consequence of any such agreement [among competitors]to exchange price information would be the stabilization of industry prices.” The Court explained that this likely anticompetitive effect arises from the fact that “each seller would know that his price concession could not be kept from his competitors,” so that “no seller participating in the information-exchange arrangement would ... have any incentive for deviating from the prevailing price level in the industry.” United States Gypsum, 438 U.S. at 457, 98 S.Ct. at 2883. Similarly, in a case where, as here, the plaintiffs alleged that the defendants participated in regular surveys of the compensation paid to hospital employees—in that case, medical residents—the court stated:

[I]t is reasonable to infer that the annual nature of the survey provides the institutional defendant co-conspirators the information they need to keep compensation levels depressed; each institutional defendant is able to set compensation levels consistent with the previous year knowing that the levels will be checked against those of the coming year with the next Survey.
Jung v. Association of American Medical Colleges, 300 F.Supp.2d 119, 167 (D.D.C.2004). Finally, beyond this recognition of their theory of causation in the case law, Plaintiffs point to the opinion of their expert, Dr. Vistnes, that information exchanges of the sort engaged in by the Defendant hospitals here “can facilitate coordination and soften competition,” which in turn “can affect prices” (or, here, wages). (Defendants' Motion, Ex. L, Vistnes 3/20/2009 Expert Report at 45, 53.)

As Defendants point out, however, this case law and Dr. Vistnes's expert opinion share a common infirmity—each addresses Plaintiffs' proffered theory of causation in the abstract, without specifically identifying the sort of evidence that would establish that a particular exchange of wage or price information actually caused softened competition or wage (or price) stabilization in a given case. Most notably, in United States Gypsum, 438 U.S. at 456, 98 S.Ct. at 2883, the Court prefaced its above-cited discussion of the possible anticompetitive effects of information exchanges by stating that it was addressing this matter only “[a]s an abstract proposition,” based on “economic theory and common human experience.” Similarly, the decision in Jung, 300 F.Supp.2d at 167, arose in the context of a motion to dismiss, so the court had no occasion to address the sort of proof the plaintiffs would ultimately need to produce in support of their claim that the alleged information exchange in that case “facilitated the stabilization of medical resident compensation.” Finally, and as noted earlier, Dr. Vistnes's theory of softened competition is just that, a theory—by his own admission, he did not “attempt[ ] to make a direct causal link between the information exchange and the resulting conduct of the hospitals,” nor did he consider whether the effect of Defendants' information exchanges “left wages above or below what the competitive price or wage should have been.” (Vistnes Dep. at 213, 252.)

Yet, while the discussions in United States Gypsum and Jung might have been limited to abstract principles, the procedural posture here enables the Court to consider these principles in light of a voluminous record which, as stated, could be viewed as evidencing a practice of on-demand exchanges of wage data. As thoroughly recounted earlier in this opinion, the Defendant hospitals regularly reached out to each other to obtain specific information about their local competitors' wage rates and practices, and Defendants have acknowledged that their compensation decisions were based, at least to some extent, on considerations of their relative standing or wage percentile in the local market. This record establishes,at a minimum, the necessary background conditions to trigger the concern that was recognized in United States Gypsum and Jung only as an abstract or hypothetical proposition—namely, that each Defendant could comfortably hold the various elements of its RN compensation package to the minimum level needed to meet its objectives, secure in the knowledge that it did not need to outbid local competitors whose wage rates and practices were unknown. In addition, the Court has accepted, for present purposes, the assertion of Plaintiffs and their expert, Dr. Ashenfelter, that RN wages at the Defendant hospitals have in fact been held below competitive levels. Under a comparable record, the court in another RN wage-suppression suit found that the plaintiffs' theory of causation in that case, resting largely upon Dr. Vistnes's expert opinion as to the competition-softening effect of information exchanges, was “persuasive for inferring causation” and sufficient for the plaintiffs to withstand summary judgment. Fleischman, 728 F.Supp.2d at 164–65.

Even assuming that more is needed to remove Plaintiffs' theory of causation from the realm of abstraction, the Court is satisfied that the record here gives rise to issues of fact as to causation that a trier of fact should be permitted to resolve. In particular, Plaintiffs have produced evidence suggesting that wage-related data exchanged among the Defendant hospitals—data which, as discussed, was sometimes shared through direct contacts or through surveys that did not satisfy the DOJ/FTC “safety zone” criteria—was actually relied upon and brought to bear in Defendants' decisions to reduce elements of their RN compensation packages below the levels that were contemplated before this data became available. Plaintiffs cite, for example, the instances (discussed earlier) in which Beaumont executives expressly referenced market surveys of RN wages in explaining decisions to award smaller-than-recommended pay increases. The record also includes the testimony of DMC's chief executive officer, Mike Duggan, explaining that a smaller than recommended merit increase for 2006 was motivated by a desire not to “raise [employee] wages unnecessarily,” and stating more generally that DMC “tr[ies] to raise in every specialty what is necessary to keep us competitive and keep our positions filled,” but “tr[ies] not to pay more than that.” (Plaintiffs' Response, Ex. O, Duggan Dep. at 53.) Similarly, Plaintiffs point to a set of “speaker's notes” accompanying a St. John salary increase proposal for fiscal year 2003/04, indicating that disaggregated, unmasked wage range data that had been very recently obtained from other Defendant hospitals was used as a basis for recommending that the maximum rate for St. John's nurses be set just below the next lowest maximum range. ( See Plaintiffs' Response, Ex. 91; see also Exs. 92, 93 (another set of speaker's notes showing that competitor RN wage rates were used in St. John's March 2004 process of setting its RN compensation, with the resulting increase set at the minimum level of the 3.5–to–5.0 percent increase under consideration).) ⁴²

42 In light of these and other examples cited by Plaintiffs, the Court fails to see how Defendants can tenably claim that Plaintiffs have not “present[ed] any evidence ... that wages would have been higher in the absence of the information exchanges challenged in their brief.” (Defendants' Reply Br. at 45.)

To be sure, this body of evidence suffers from weaknesses that undercut the inferences Plaintiffs wish to draw from it. First, it is highly anecdotal, shedding light on only a handful of the many RN compensation decisions made by the Defendant hospitals during the class period. As discussed earlier, Defendants and their expert have produced graphs that summarize the net effect of these decisions on RN wages, and these graphs reveal a significant degree of variance among the Defendant hospitals in their rates of change and underlying wage rates during the class period.⁴³ Thus, even making the most of the examples cited by Plaintiffs, they still face a difficult challenge in persuading a trier of fact to infer that RN wages in the market as a whole were depressed as a result of the exchange of wage-related data among the Defendant hospitals. While it may be true, as Plaintiffs assert, that the requisite causal connection need not be established through an exhaustive compilation of “specific example[s] of ... particular wage cut[s] that can be linked to ... specific exchanges of information,” (Plaintiffs' Response Br. at 99), there surely are limits to how much a trier of fact will be willing, or even permitted, to extrapolate from a modest number of wage-limiting decisions to market-wide setting of wages at sub-competitive levels.

43 Plaintiffs, of course, have submitted a graph prepared by their own expert, Dr. Ashenfelter, that purports to show greater uniformity in Defendants' wage-setting decisions and resulting wage rates over the class period. This “battle of the experts” must be left for the trier of fact to resolve.

Next, and more generally, Plaintiffs face a challenge in ensuring that the requisite finding of causation rests upon conduct that violates antitrust law, as opposed to conduct that does not. As explained, exchanges of wage data are not per se unlawful, and Plaintiffs' own expert, Dr. Vistnes, has acknowledged that an exchange of wage data may be pro-competitive, even if it results in a wage reduction. ( See Vistnes Dep. at 200.) Moreover, the law is clear that a private plaintiff in an antitrust case may recover damages only for losses attributable to “a competition- reducing aspect or effect of the defendant's behavior.” Atlantic Richfield, 495 U.S. at 344, 110 S.Ct. at 1894. Thus, Defendants correctly observe that Plaintiffs can recover only by showing that the sub-competitive wage levels identified by Dr. Ashenfelter were caused by a “competition- reducing aspect or effect” of the Defendant hospitals' exchange of wage data.

Nevertheless, while all of this may give rise to thorny issues of proof, the Court declines to adopt Defendants' proposed solution— i.e., a ruling as a matter of law that Plaintiff cannot make this showing. Taking Defendants' argument to its logical conclusion, a plaintiff seemingly could never prevail in a wage-suppression (or parallel pricing) suit based on a theory of information sharing, because even if a group of competitors shared all of their confidential business data in making wage (or price) decisions—or, indeed, convened in the same conference room and reached their decisions at the same time—the plaintiff still would be left to somehow identify and separate out the particular aspects of this information-sharing arrangement that were unlawful, and then prove that these decisions would have been different if the competitors had engaged in only a “permissible” exchange of data. Even assuming one could arrive at a settled understanding of the lawful versus unlawful aspects of an exchange of wage (or price) data— e.g., by adopting the DOJ/FTC Guidelines, with their preference for masked, aggregated, and aged data—there would still remain the difficulty of forging a specific causal link between an anti-competitive wage (or price) decision and an unlawful component of the data exchange that preceded it.

The Court does not read the pertinent case law as imposing such an insurmountable standard of proof. As revealed in the foregoing discussion of this law, the courts view certain types of information exchanges as triggering concerns of anticompetitive effects, and the DOJ/FTC Guidelines likewise speak to these concerns. The record in this case features ample evidence of exchanges of wage data that implicate these concerns and exceed the Guidelines' “safety zone,” and Plaintiffs also have produced at least some evidence that the data gathered in these exchanges was used in reaching decisions that had an unfavorable (or less favorable) impact upon nurse compensation. Finally, Plaintiffs have proffered the opinion of their expert, Dr. Ashenfelter, that the RN workforces at the Defendant hospitals are being paid at sub-competitive levels. This record, viewed as a whole, serves to blunt Defendants' criticism that Plaintiffs seek to try their case in the abstract— i.e., that Plaintiffs have argued only that “causation in an information exchange case is theoretically possible,” but have failed to present facts that would support “a concrete theory of causation in this case.” (Defendants' Reply Br. at 44.) Having reviewed the record here in light of the relevant case law and governing legal standards, the Court finds that Plaintiffs have produced sufficient evidence from which a trier of fact could infer the requisite causal link between an antitrust violation and antitrust injury.

D. Plaintiffs Need Not Offer a Detailed Analysis of the Relevant Market in Order to Pursue Their “Rule of Reason” Claim.

Finally, in a challenge directed solely at Plaintiffs' Count II “rule of reason” claim, Defendants argue that this claim is defeated by Plaintiffs' failure to establish the relevant market within which to evaluate their allegations of anticompetitive effects. In particular, Defendants challenge the assertion of Plaintiffs and their expert, Dr. Ashenfelter, that the relevant market in this case consists only of RNs employed at hospitals in the Detroit metropolitan area, and not RNs who work for non-hospital employers such as physicians' offices, ambulatory surgery centers, nursing homes, and public health agencies. In response, Plaintiffs contend that they need not produce a detailed or formal market analysis in this case, where they have produced direct proof of the anticompetitive effects of Defendants' exchange of wage data. As discussed below, the Court finds that Plaintiffs have the better of the argument on this issue.

Under a “conventional rule-of-reason approach,” a court must “engage in a thorough analysis of the relevant market and the effects of the restraint in that market.” Realcomp II, Ltd. v. Federal Trade Commission, 635 F.3d 815, 825 (6th Cir.2011). As both sides agree, however, even in cases where a challenged practice is subject to scrutiny under the “rule of reason,” rather than subject to condemnation as per se unlawful, the “practical difference” between these two analytical approaches “is sometimes negligible,” Re/Max International, 173 F.3d at 1013, and the courts occasionally find that a “less searching Rule of Reason analysis” is sufficient under the facts of a particular case, In re Northwest Airlines, 208 F.R.D. at 206. The Supreme Court has acknowledged “[t]he truth ... that our categories of analysis of anticompetitive effect are less fixed than terms like ‘per se,’ ‘quick look,’ and ‘rule of reason’ tend to make them appear.” California Dental Association v. Federal Trade Commission, 526 U.S. 756, 779, 119 S.Ct. 1604, 1617, 143 L.Ed.2d 935 (1999). Of particular pertinence here, the Supreme Court has explained that “[s]ince the purpose of the [rule of reason] inquiries into market definition and market power is to determine whether an arrangement has the potential for genuine adverse effects on competition, proof of actual detrimental effects, such as a reduction of output, can obviate the need for an inquiry into market power, which is but a surrogate for detrimental effects.” Federal Trade Commission v. Indiana Federation of Dentists, 476 U.S. 447, 460–61, 106 S.Ct. 2009, 2019, 90 L.Ed.2d 445 (1986) (internal quotation marks and citation omitted).

In Plaintiffs' view, they have offered such proof of “actual detrimental effects” here, and thus need not propound a detailed market analysis. Specifically, they point to Dr. Ashenfelter's benchmark analysis of the fees paid by the Defendant hospitals for agency nurses. This analysis, if credited, ⁴⁴ serves as direct proof of a detrimental impact upon the wages paid to RNs by the Defendant hospitals. Moreover, Plaintiffs have identified a sufficient basis for attributing this detrimental impact to Defendants' exchange of information regarding the compensation paid to their RN workforces, where the case law, see, e.g. Todd, 275 F.3d at 211–13, the DOJ/FTC Guidelines, and the opinion of Plaintiffs' expert, Dr. Vistnes, all lend support to the proposition that information exchanges of the sort engaged in here can be expected to soften competition and have an anticompetitive effect on the wages paid to RNs.⁴⁵ Accordingly, the Court finds that Plaintiffs' “rule of reason” claim is not defeated by the absence of a detailed analysis of the relevant market and Defendants' power within that market. See Fleischman, 728 F.Supp.2d at 163–64 (concluding under a similar record that a detailed market analysis was unnecessary).

44 Again, Defendants argue separately that this benchmark analysis is subject to exclusion, but the Court leaves this argument for another day.

45 To be sure, and as discussed above, the Court recognizes that Plaintiffs face difficulties of proof in showing that Defendants' exchanges of wage data actually did have an anticompetitive effect on RN wages, and that this effect was brought about by unlawful aspects of these exchanges. As explained, this evidentiary obstacle arises from the difficulty of isolating the impact of the unlawful, versus the lawful, aspects of these information exchanges. Yet, the detailed market analysis sought by Defendants would not aid Plaintiffs (or the trier of fact) in resolving this difficulty, but instead would address a different question—namely, whether Defendants' exchanges of information perhaps might not raise concerns of anticompetitive effect because, for example, any tendency of these exchanges to depress RN wages would be offset or overcome by the ability of RNs to pursue employment in a non-hospital setting. As indicated below, the Court is satisfied that the more abbreviated market analysis offered by Plaintiffs sufficiently addresses this question.

Next, to the extent Defendants argue that Plaintiffs must at least “show the rough contours of a relevant market,” Republic Tobacco Co. v. North Atlantic Trading Co., 381 F.3d 717, 737 (7th Cir.2004), the Court is satisfied that Plaintiffs and their expert have met this more relaxed standard of “rule of reason” analysis. As Plaintiffs observe, to see that Dr. Ashenfelter's exclusion of non-hospital RNs was reasonable, one need look no further than the Defendant hospitals' own conduct in commissioning and conducting wage surveys and in their direct exchanges of compensation-related information. In particular, “Defendants have commissioned literally scores of surveys to assist them in setting the wages for their hospital nurses and engaged in hundreds of phone calls and e-mails to help with that process, and those surveys and contacts always seek information only on what other hospitals are paying their hospital nurses.” (Plaintiffs' Response Br. at 126.) Although Defendants point to an exception or two that seem to prove the rule, and they further cite some evidence that they have competed with non-hospital employers in their recruitment of nurses, there is no basis in the record for believing that Plaintiffs' exclusion of non-hospital RNs and employers from the relevant market has significantly distorted the analysis of the anticompetitive effects of Defendants' exchange of wage data under the rule of reason.

IV. CONCLUSION

For the reasons set forth above,

NOW, THEREFORE, IT IS HEREBY ORDERED that the motion for summary judgment brought by Defendants Henry Ford Health System, Mount Clemens General Hospital, Inc., William Beaumont Hospital, and Trinity Health Corp. (docket # 376) is GRANTED IN PART, with the moving Defendants awarded summary judgment in their favor as to Count I of Plaintiffs' Third Corrected Class Action Complaint, and DENIED IN PART, as to the moving Defendants' challenge to Count II of Plaintiffs' Third Corrected Class Action Complaint. Likewise, IT IS FURTHER ORDERED that Defendant Detroit Medical Center's motion for summary judgment (docket # 358) is GRANTED IN PART, with respect to Count I of the complaint, and DENIED IN PART, with respect to Count II of the complaint.

Finally, in light of the tentative settlement reached between Plaintiffs and Defendant Mount Clemens General Hospital, IT IS FURTHER ORDERED that Defendant Mount Clemens General Hospital's renewed motion for summary judgment (docket # 355) is DENIED WITHOUT PREJUDICE.