Cason-Merendo et al v. Detroit Medical Center et al
Filing
793
OPINION and ORDER denying Defendant's 349 Sealed Motion to Exclude the Expert Testimony of Orley Ashenfelter. Signed by District Judge Gerald E. Rosen. (JOwe)
UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF MICHIGAN
SOUTHERN DIVISION
PAT CASON-MERENDA and
JEFFREY A. SUHRE,
Plaintiffs,
Case No. 06-15601
Hon. Gerald E. Rosen
Magistrate Judge Donald A. Scheer
v.
DETROIT MEDICAL CENTER, et al.,
Defendants.
____________________________/
OPINION AND ORDER DENYING DEFENDANTS’ MOTION TO
EXCLUDE THE EXPERT TESTIMONY OF ORLEY ASHENFELTER
At a session of said Court, held in
the U.S. Courthouse, Detroit, Michigan
on
April 22, 2013
PRESENT: Honorable Gerald E. Rosen
Chief Judge, United States District Court
I. INTRODUCTION
Through the present motion, certain of the Defendants seek to exclude the
testimony of Plaintiffs’ expert, Orley Ashenfelter, Ph.D., on the grounds that his
testimony fails to meet the admissibility standards set out in Fed. R. Evid. 702 and
Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct 2786 (1993).1
1
Specifically, the present motion was brought by Defendants Detroit Medical Center,
Henry Ford Health System, Mount Clemens General Hospital, Inc., William Beaumont Hospital,
and Trinity Health Corp. Since this motion was filed, Plaintiffs have reached settlements with
Defendants Mount Clemens General Hospital and William Beaumont Hospital, and the Court
has granted its preliminary approval of these settlements. In addition, Plaintiffs have more
recently reached settlements with Defendants Henry Ford Health System and Trinity Health
Although Defendants have advanced a number of challenges to Dr. Ashenfelter’s
proposed expert testimony as summarized in his expert and rebuttal reports, they argue
principally that Dr. Ashenfelter’s “benchmark” analysis rests upon an unreliable
methodology and is fatally undermined by unwarranted assumptions, oversimplifications,
and leaps in logic. In response, Plaintiffs contend that the benchmark analysis employed
by Dr. Ashenfelter has been widely used and accepted in antitrust suits, and that
Defendants’ various challenges to Dr. Ashenfelter’s implementation of this methodology
and his use and interpretation of data in the record are matters to be explored through
cross-examination at trial.
Defendants’ motion has been fully briefed by the parties. Having reviewed the
parties’ briefs and accompanying exhibits, the Court finds that the pertinent facts and
legal arguments are sufficiently presented in these written submissions, and that oral
argument would not aid the decisional process. Accordingly, the Court will decide
Defendants’ motion “on the briefs.” See Local Rule 7.1(f)(2), U.S. District Court,
Eastern District of Michigan. This opinion sets forth the Court’s rulings on this motion.
Corp., subject to the Court’s preliminary approval of these settlements.
Against this backdrop of settlement negotiations, Defendant Detroit Medical Center
evidently is the sole remaining party against which Plaintiffs are pursuing their claims in this
case. Nonetheless, because certain of these settlements are awaiting preliminary approval, the
Court will refer to the moving party as “Defendants” throughout the remainder of this opinion.
2
II. SUMMARY OF DR. ASHENFELTER’S PROPOSED EXPERT TESTIMONY
The underlying facts of this case have been thoroughly set forth in the Court’s
ruling on Defendants’ summary judgment motion, see Cason-Merenda v. Detroit Medical
Center, 862 F. Supp.2d 603, 606-23 (E.D. Mich. 2012), and need not be repeated here.
Briefly, the two Plaintiff registered nurses (“RNs”), Pat Cason-Merenda and Jeffrey A.
Suhre, allege that the Defendant health care institutions operating in the Detroit
metropolitan area have violated § 1 of the federal Sherman Act, 15 U.S.C. § 1, by
agreeing to regularly exchange compensation-related information among themselves in a
manner that has reduced competition among Detroit-area hospitals in the wages paid to
RNs.2 In pursuing this federal antitrust claim, Plaintiffs seek to recover on behalf of
themselves and a class of RNs employed by the eight Defendant hospitals.
A.
Dr. Ashenfelter’s Qualifications
As stated in his expert report, Dr. Orley Ashenfelter is the Joseph Douglas Green
Professor of Economics at Princeton University. He received his Ph.D. in economics
from Princeton in 1970, and he has received a number of awards and honors and authored
or edited many books, journals, and articles in the course of his academic career. Dr.
Ashenfelter has proffered expert reports or given expert testimony in several antitrust
2
Plaintiffs further allege that the Defendant health care providers have violated § 1 of the
Sherman Act by conspiring among themselves and with other local hospitals to hold down the
wages of RNs employed by these institutions. In a March 22, 2012 opinion and order, however,
the Court found that Defendants were entitled to summary judgment in their favor as to this
claim, see Cason-Merenda, 862 F. Supp.2d at 628-41, leaving Plaintiffs’ “rule of reason” claim
in Count II of their complaint as the sole antitrust claim going forward in this litigation.
3
cases and proceedings, including a similar nurse wage suit brought in the Northern
District of New York. See Fleischman v. Albany Medical Center, 728 F. Supp.2d 130,
145-50 (N.D.N.Y. 2010) (rejecting a defense challenge to Dr. Ashenfelter’s proposed
expert testimony in that case). In light of this extensive background, Defendants do not
question Dr. Ashenfelter’s qualifications to give expert testimony on economic issues of
relevance to this litigation.
B.
Dr. Ashenfelter’s Initial Expert Report
Dr. Ashenfelter begins his expert report by summarizing the allegations of
Plaintiffs’ complaint, and by assuming, as requested by Plaintiffs’ counsel, that Plaintiffs
can prove their allegations of (i) a conspiracy among the Defendant hospitals to depress
the compensation of their RNs,3 and (ii) an agreement among the Defendant health care
institutions to regularly exchange RN compensation data in a manner that resulted in
reduced competition among Defendants in the market for RNs and a corresponding
restraint in RN wages below competitive levels. Against this backdrop, Dr. Ashenfelter
provides his opinions on three questions posed by Plaintiffs and their counsel:
•
Whether it can be shown by common evidence that all or almost all
of the members of the [plaintiff] class were harmed by the
conspiracy?
•
What is the aggregate total lost compensation suffered by members
of the class and how can this aggregate sum be allocated across class
3
As noted above, the Court has determined as a matter of law that the record developed
during discovery fails to meet the evidentiary threshold for permitting a trier of fact to decide
whether this alleged conspiracy existed. See Cason-Merenda, 862 F. Supp.2d at 641.
4
members to reflect each class member’s individual losses?
•
Did the conspiracy lead to market power and anticompetitive
outcomes in the market for jobs as RNs in hospitals in the Detroit
[Metropolitan Statistical Area (“MSA”)]?
(Defendants’ Motion, Ex. A, Ashenfelter Report at ¶ 4.)
As to the first of these questions, Dr. Ashenfelter concludes that “all or almost all
of the class was harmed by the alleged conspiracy,” and that “this can be shown using
evidence that is common to the class.” (Id. at ¶ 11.) He gives three reasons for arriving at
this conclusion. First, Dr. Ashenfelter observes that at each Defendant hospital, the “class
members’ wages, and most other elements of compensation, were determined in a
common administered compensation system,” and he reasons that “[e]xplicit or implicit
cooperation among the defendants” in setting RN wages or sharing wage-related
information “would have caused adjustments to the compensation systems.” (Id. at ¶ 7.)
The adjustments to these compensation systems, in turn, “would have affected the
compensation of all or almost all individual class members.” (Id.)4
Next, Dr. Ashenfelter uses econometric analysis to “show[] that relative labor
supply to the various jobs in the [plaintiff RN] class is very elastic.” (Id. at ¶ 8.)5 From
this he infers that “even if the alleged conspiracy directly affected a subgroup of the class,
4
As noted in Plaintiffs’ response to Defendants’ motion, Defendants seemingly do not
challenge this aspect of Dr. Ashenfelter’s opinion.
5
Later in his report, Dr. Ashenfelter explains that “[e]conomists say that the relative
supply of workers to different jobs is ‘highly elastic’ if a small proportional change in relative
compensation levels for the jobs will result in a large proportional change in the relative supply
of workers to the jobs.” (Id. at ¶ 98.)
5
such as a plausible benchmark group, pay for all jobs held by members of the class would
have been depressed.” (Id. at ¶ 8.)
Third, Dr. Ashenfelter employs a benchmark analysis as a class-wide method for
both identifying and quantifying the impact of the alleged conspiracy upon members of
the plaintiff class. As he explains:
. . . [E]ach of the defendants makes extensive use of nurses supplied
by temporary agencies (“agency nurses”) to do work that would otherwise
be performed by members of the class. In a competitive labor market an
employer will pay each employee the amount that the employee adds to the
firm’s revenue. It is reasonable to infer that the value to the defendant
hospitals of the work performed by agency nurses is at least as high as the
fee paid by the hospital to the agency (otherwise it would not be profitable
for the hospital to employ the agency nurses). As a result, the fee paid to
the agency for an agency nurse is less than or equal to the marginal revenue
product of that RN in that job. Therefore, the fee provides a conservative
benchmark for the cost to a hospital of employing an RN in that job in a
competitive market.
(Id. at ¶ 9 (footnote with citation omitted).)
This benchmark analysis, resting upon the Defendant hospitals’ use of agency
nurses, forms much of the basis for Defendants’ various challenges to Dr. Ashenfelter’s
expert testimony. Dr. Ashenfelter explains that he has used this agency nurse benchmark
“to calculate what each member of the class would have earned in the absence of the
alleged conspiracy,” and that this calculation “show[s] that almost all of the members of
the class actually earned less than they would have in the ‘but-for’ world.” (Id. at ¶ 10.)
Accordingly, this analysis not only demonstrates, in Dr. Ashenfelter’s view, that “all or
almost all of the class was harmed by the alleged conspiracy and [that] this can be shown
6
using evidence that is common to the class,” but it also “provides a measure of each class
member’s lost earnings.” (Id. at ¶¶ 11-12.) Dr. Ashenfelter, therefore, relies on this
benchmark to answer both the first and second questions posed by Plaintiffs’ counsel, as
this analysis makes it “possible both to calculate the aggregate amount of damages and
allocate the total among individual class members in a reasonable way.” (Id. at ¶ 12.)
As Dr. Ashenfelter recognizes in his report, the benefits and services received by a
Defendant hospital through the use of an agency nurse are not precisely the same as those
provided by a regular hospital-employed RN, so that it is necessary to adjust the rates
paid for agency nurses to account for these differences. (See id. at ¶ 112.) Broadly
speaking, he makes two such adjustments in his initial report. First, he subtracts from his
agency fee benchmark certain human resources costs — including the cost of fringe
benefits and the cost of payroll taxes and worker’s compensation contributions — that the
hospital would have to pay for its own RNs, but that agencies pay on behalf of their
agency nurse employees who are brought in to perform RN work at the hospital. (See id.
at ¶¶ 112-14.) Next, he attempts to quantify the value to a hospital of using an agency
nurse to fill an RN vacancy on short notice, by looking to an instance in which Defendant
Oakwood Healthcare Inc. set up an “internal agency” that “employed RNs to work at
various Oakwood facilities as if they were working for an external agency.” (Id. at ¶
115.) Using data from this “internal agency” experience, Dr. Ashenfelter computes a
11.2 percent “premium” as representing the additional value provided by agency nurses
performing more temporary or flexible RN services. (Id.)
7
Finally, turning to the third question posed by Plaintiffs’ counsel, Dr. Ashenfelter
“conclude[s] that the alleged conspiracy would have provided the defendants with market
power and that this market power would have led to anticompetitive effects in the
market.” (Id. at ¶ 13.) In support of this opinion, Dr. Ashenfelter points to data which, in
his view, establish two anticompetitive effects in the relevant market, sub-competitive
compensation and sub-competitive levels of RN employment. (See id. at ¶¶ 13, 127-28.)
Based on the principle that “evidence of anticompetitive effects is also evidence of
market power because the anticompetitive effects could not exist in the absence of market
power,” he concludes that the Defendant hospitals must have exercised market power in
order to produce the anticompetitive outcomes he has observed. (Id. at ¶¶ 13, 126-27.) In
addition, Dr. Ashenfelter analyzes the Defendant institutions’ collective market share in
the relevant market — RN jobs at hospitals in the Detroit MSA — and opines that “if the
defendants were to act cooperatively they would have sufficient market share to depress
earnings and employment in the market.” (Id. at ¶¶ 14, 129.)
C.
Dr. Ashenfelter’s Rebuttal Report
In his rebuttal report, Dr. Ashenfelter seeks to address the various challenges
raised by Defendants’ several experts — Professor Daniel L. Rubinfeld, Dean Edward A.
Snyder, Professor Daniel S. Hamermesh, and Joseph Caracci, RN6 — to the analysis and
6
In separate motions that remain pending before the Court, Plaintiffs seek to exclude
certain of the opinions offered by defense experts Rubinfeld and Snyder. In addition, in an
opinion and order dated October 18, 2010, the Court granted Plaintiffs’ motion to exclude the
testimony of defense expert Caracci.
8
conclusions set forth in his initial report. Most notably, Dr. Ashenfelter’s rebuttal report
reflects two changes to his agency benchmark analysis and figures to account for the
points made by Defendants’ experts. First, Dr. Ashenfelter makes a further downward
adjustment to his agency nurse benchmark — in an amount less than $0.10 per hour of
nurse employment — to reflect additional human resources costs that the Defendant
hospitals incur in employing class members but do not pay for agency nurses. (See
Defendants’ Motion, Ex. B, Ashenfelter Rebuttal Report at ¶¶ 4, 39-40.) Next, he
increases his “flexibility premium” — a figure which, as noted earlier, is intended to
quantify the value to a hospital of employing agency nurses who are willing to work
“where and when they are needed,” (id. at ¶ 3) — to 18 percent (from its initial value of
11.2 percent), to account for the observation in Professor Rubinfeld’s expert report that
not all of the RNs in Oakwood’s “internal agency” were equally comparable to true
agency nurses. (See id. at ¶¶ 7, 64-67.) In light of these adjustments, Dr. Ashenfelter has
computed “total lost earnings for the class of $720.5 million as compared with $847.6
million in [his] first analysis of this question.” (Id. at ¶ 7.)
Roughly a month after his rebuttal report, Dr. Ashenfelter produced an errata sheet
that corrects some of the figures found in his rebuttal report. (See Defendants’ Motion,
Ex. F, Rebuttal Errata.) For example, Dr. Ashenfelter has further increased his
“flexibility premium” to 21.5 percent. (See id. at 1.) He has also lowered his overall
9
calculation of class damages from $720.5 million to $596.2 million. (See id. at 2.)7
III. ANALYSIS
A.
The Standards Governing Defendants’ Motion
In resolving Defendants’ challenge to the proposed testimony of Plaintiffs’ expert,
Dr. Orley Ashenfelter, the Court necessarily begins with the language of the pertinent
Federal Rule of Evidence, which provides:
If scientific, technical, or other specialized knowledge will assist the
trier of fact to understand the evidence or to determine a fact in issue, a
witness qualified as an expert by knowledge, skill, experience, training, or
education, may testify thereto in the form of an opinion or otherwise, if (1)
the testimony is based upon sufficient facts or data, (2) the testimony is the
product of reliable principles and methods, and (3) the witness has applied
the principles and methods reliably to the facts of the case.
Fed. R. Evid. 702.8 This Rule imposes upon the federal district courts a “basic
7
In their motion, Defendants point to a sur-rebuttal report prepared by one of their
experts, Professor Rubinfeld, as purportedly showing that Dr. Ashenfelter’s errata sheet makes
“not merely corrections in arithmetic, but wholesale changes in his methodology.” (Defendants’
Motion, Br. in Support at 8, 15-16.) In an order dated March 31, 2010, however, the Court
denied Defendants’ request for leave to file the Rubinfeld sur-rebuttal report upon which they
seek to rely in challenging Dr. Ashenfelter’s errata, finding that this proposed report was not
timely produced. In addition, Plaintiffs strenuously object to the notion that Dr. Ashenfelter’s
errata sheet incorporates any changes in methodology, and instead assert that this addendum
merely reflects some technical mathematical recalculations to correct an “inadvertent omission”
of a variable and a “computational error” in the results reported in the rebuttal report. (Plaintiffs’
Response Br. at 8-9 & n.10 (characterizing Defendants’ claim of a change in methodology as an
“outrageous” and “demonstrably baseless allegation”).) While the Court finds it unnecessary to
stray too far into this thicket, it would certainly appear that Dr. Ashenfelter’s two-page errata,
with its handful of changes to specific figures from his rebuttal report, would be a poor candidate
for a Trojan horse of methodological changes to his underlying analytical approach.
8
Since Defendants filed their motion, the language of this Rule has been amended, but the
advisory committee notes for these amendments state that “[t]hese changes are intended to be
stylistic only,” and that “[t]here is no intent to change any result in any ruling on evidence
admissibility.” Fed. R. Evid. 702, advisory committee notes to 2011 amendments.
10
gatekeeping obligation” to ensure that an expert’s proffered testimony is both relevant
and reliable. Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147, 119 S. Ct. 1167, 1174
(1999).
The Sixth Circuit has described a court’s inquiry under Rule 702 as governed by
three mandatory requirements and a non-exclusive list of additional considerations:
Parsing the language of the Rule, it is evident that a proposed
expert’s opinion is admissible, at the discretion of the trial court, if the
opinion satisfies three requirements. First, the witness must be qualified by
“knowledge, skill, experience, training, or education.” Fed. R. Evid. 702.
Second, the testimony must be relevant, meaning that it “will assist the trier
of fact to understand the evidence or to determine a fact in issue.” Id.
Third, the testimony must be reliable. Id. Rule 702 guides the trial court by
providing general standards to assess reliability: whether the testimony is
based upon “sufficient facts or data,” whether the testimony is the “product
of reliable principles and methods,” and whether the expert “has applied the
principles and methods reliably to the facts of the case.” Id. In addition,
Daubert provides a non-exclusive checklist for trial courts to consult in
evaluating the reliability of expert testimony. These factors include:
testing, peer review, publication, error rates, the existence and maintenance
of standards controlling the technique’s operation, and general acceptance
in the relevant scientific community.
In re Scrap Metal Antitrust Litigation, 527 F.3d 517, 528-29 (6th Cir. 2008) (internal
quotation marks and citations omitted). The Supreme Court has cautioned, however, that
the factors cited in Daubert “do not constitute a definitive checklist or test,” and that they
“may or may not be pertinent in assessing reliability, depending on the nature of the issue,
the expert’s particular expertise, and the subject of his testimony.” Kumho Tire, 526 U.S.
at 150, 119 S. Ct. at 1175 (internal quotation marks and citations omitted); see also In re
Scrap Metal, 527 F.3d at 529 (recognizing that “the Daubert factors are not dispositive in
11
every case and should be applied only where they are reasonable measures of the
reliability of expert testimony” (internal quotation marks and citation omitted)).
“[W]hether Daubert’s specific factors are, or are not, reasonable measures of reliability in
a particular case is a matter that the law grants the trial judge broad latitude to determine.”
Kumho Tire, 526 U.S. at 153, 119 S. Ct. at 1176.
As this Court observed in an earlier antitrust suit, a court “must remain mindful of
its limited gatekeeping role” under Rule 702. In re Northwest Airlines Corp. Antitrust
Litigation, 197 F. Supp.2d 908, 914 (E.D. Mich. 2002). In particular, the “rejection of
expert testimony is the exception rather than the rule, and the trial court’s role as
gatekeeper is not intended to serve as a replacement for the adversary system.” In re
Northwest Airlines, 197 F. Supp.2d at 913 (internal quotation marks and citation omitted).
Thus, Daubert emphasizes that “[v]igorous cross-examination, presentation of contrary
evidence, and careful instruction on the burden of proof are the traditional and appropriate
means of attacking shaky but admissible evidence,” and that “[t]hese conventional
devices, rather than wholesale exclusion . . . , are the appropriate safeguards where the
basis of [expert] testimony meets the standards of Rule 702.” Daubert, 509 U.S. at 596,
113 S. Ct. at 2798.
B.
Defendants’ Challenges to Dr. Ashenfelter’s Benchmark Analysis Go Only to
the Weight, and Not the Admissibility, of His Proposed Testimony, and
Therefore Must Be Left for the Trier of Fact to Resolve.
As observed earlier, the lion’s share of Defendants’ challenges to Dr. Ashenfelter’s
proposed expert testimony focus on his benchmark analysis, through which he looks to
12
the Defendant hospitals’ use of nurses supplied by outside agencies to both demonstrate
the impact of the alleged antitrust conspiracy upon the members of the plaintiff RN class
and calculate the damages incurred by the plaintiff class. Accordingly, the Court turns
first to Defendants’ various critiques of Dr. Ashenfelter’s benchmark analysis, and then
addresses the handful of remaining issues raised in Defendants’ motion.
As a threshold matter, Defendants do not dispute that the “benchmark” or
“yardstick” approach adopted by Dr. Ashenfelter in formulating his expert opinion is a
“well accepted” method of proving antitrust damages. Fishman v. Estate of Wirtz, 807
F.2d 520, 551 (7th Cir. 1986); see also Conwood Co. v. United States Tobacco Co., 290
F.3d 768, 793 (6th Cir. 2002); Home Placement Service, Inc. v. Providence Journal Co.,
819 F.2d 1199, 1205-06 (1st Cir. 1987). Indeed, this Court held in a prior antitrust suit
that an economic expert produced by the plaintiff air travelers would be permitted to
present his benchmark analysis to the trier of fact, both as a means of showing that the
defendant airlines had engaged in monopolistic pricing and as a measure of the damages
purportedly suffered by the plaintiff class as a result of the defendants’ allegedly
anticompetitive conduct. See In re Northwest Airlines, 197 F. Supp.2d at 922-30. Under
this benchmark approach, “the plaintiff’s experience in a hypothetical free market” that
would exist in the absence of the defendant’s antitrust violation is determined by
reference to “the experience of a comparable [participant] in an actual free market.”
Fishman, 807 F.2d at 551.
As this Court has recognized, the benchmark chosen by Plaintiffs and their expert
13
“must be sufficiently comparable to the market under consideration to permit the
conclusion that price differences are the product of antitrust violations, and not other
factors.” In re Northwest Airlines, 197 F. Supp.2d at 922; see also Home Placement
Service, 819 F.2d at 1206 (“Central to this so-called ‘yardstick’ approach . . . is the
requirement [that] the plaintiff identify a sufficiently comparable firm (the ‘yardstick’)
against which it can measure its quantum of damages.”). This requirement of sufficient
comparability, however, does not demand strict identity between the benchmark and the
hypothetical free market that Plaintiffs and their expert seek to describe, because “[t]he
vagaries of the marketplace usually deny us sure knowledge of what plaintiff’s situation
would have been in the absence of the defendant’s antitrust violation.” J. Truett Payne
Co. v. Chrysler Motors Corp., 451 U.S. 557, 566-67, 101 S. Ct. 1923, 1929 (1981).
“Markets need not be wholly identical to serve useful and reliable purposes . . . in
determining what a market would do in the absence of an antitrust violation,” In re
Northwest Airlines, 197 F. Supp.2d at 929, and Plaintiffs and their expert need only
produce sufficient evidence of comparability “as to permit a legitimate comparison by the
trier of fact,” Home Placement Service, 819 F.2d at 1206.
Against this backdrop, the admissibility of Dr. Ashenfelter’s benchmark approach
turns on two questions: (i) whether the fees paid by the Defendant hospitals for agency
nurses are sufficiently comparable to the wages paid to RNs employed by those hospitals,
such that the former may be used as a benchmark for what the latter would be in the
absence of Defendants’ alleged antitrust violation, and (ii) whether Dr. Ashenfelter’s
14
analysis sufficiently accounts for the ways in which agency fees differ from in-house RN
wages. The first of these questions need not be addressed at any length, because
Defendants have largely failed to suggest any reason why agency nurse fees should be
categorically ineligible for consideration as a benchmark for competitive RN wages.
Defendants’ only apparent argument on this point is that Dr. Ashenfelter’s reliance on
agency nurse fees is methodologically unsound “because he failed to investigate the
temporary nurse agency industry that is the source of his benchmark.” (Defendants’
Motion, Br. in Support at 10.) Yet, as the Court explained in an earlier ruling in this case,
“the internal operations of and costs incurred by a nurse staffing agency play no role in
Dr. Ashenfelter’s analysis.” Cason-Merenda v. Detroit Medical Center, No. 06-15601,
2010 WL 8583308, at *4 (E.D. Mich. Oct. 18, 2010). Rather, nurse agencies are merely a
“black box” in this analysis, into which Defendants pay sums of money in exchange for
workers who “have the same qualifications as [the RNs employed by the Defendant
hospitals] and work side-by-side with their employee nurses providing the same essential
care-giving services to their patients.” (Plaintiffs’ Response Br. at 14.) Accordingly,
Defendants have failed to explain the need for Dr. Ashenfelter to investigate the
temporary nurse agency industry, much less suggest how this lack of investigation
undermines his benchmark analysis.
Turning to the second question, Defendants contend as a general matter that Dr.
Ashenfelter has failed to make the “substantial adjustments” that purportedly are
necessary to ensure that his agency fee benchmark is “reasonably comparable” to the
15
wages that RNs would receive in the absence of Defendants’ allegedly anticompetitive
conduct. (Defendants’ Motion, Br. in Support at 10-11.) In advancing this argument,
Defendants begin with the broad assertion that Dr. Ashenfelter failed to “consider the
differences between the fee to an agency and a wage to a nurse,” but instead merely
“equated the two, ignoring the fact . . . that a fee to any agency, like a price paid to any
company, covers more than the wages of the nurse employed by the agency.” (Id. at 11.)
To the extent, however, that Defendants suggest that Dr. Ashenfelter made no
adjustments whatsoever in determining how an agency fee should compare to RN wages,
Plaintiffs correctly observe that this assertion “is, quite simply, untrue.” (Plaintiff’s
Response Br. at 15.) As noted above, Dr. Ashenfelter explicitly acknowledges in his
initial expert report that “in hiring members of the class, the defendants incur costs in
addition to the class members’ monetary compensation that they do not incur when
contracting to hire a nurse from an agency,” and he recognizes that it therefore is
necessary to “adjust these agency rates to arrive at a corresponding figure for hourly
earnings of members of the class.” (Defendants’ Motion, Ex. A, Ashenfelter Report at ¶
112.)
Nonetheless, Defendants maintain that the adjustments made by Dr. Ashenfelter
are inadequate in a number of respects. First, they contend that Dr. Ashenfelter failed to
thoroughly review the records of the Defendant hospitals in order to fully account for the
human resources costs these hospitals avoided by using agency nurses. As explained
above, however, Defendants’ argument on this point ignores the adjustments outlined in
16
Dr. Ashenfelter’s initial report to account for these costs, (see Ashenfelter Report at ¶¶
113-14), as well as the further adjustment made in his rebuttal report to reflect additional
human resources costs identified by Defendants and their expert, Professor Rubinfeld,
(see Defendants’ Motion, Ex. B, Ashenfelter Rebuttal Report at ¶¶ 36-40, 67 n.50).9
9
At various points in their motion, Defendants suggest that it was somehow “improper”
for Dr. Ashenfelter to use his rebuttal report as an opportunity to correct inaccuracies or
infirmities that Defendants and their experts had identified in his initial report. (See, e.g.,
Defendants’ Motion, Br. in Support at 12-13 & n.6, 15-16 (characterizing the rebuttal report as
an impermissible “do-over”).) Yet, as Plaintiffs observe, “it would be odd indeed if the law
prevented an expert from taking on board the suggestions for refinements put forward by another
expert commenting on his opinion.” (Plaintiffs’ Response Br. at 27.) Indeed, the Federal Rule
governing initial and supplemental discovery disclosures — including the disclosure of expert
testimony — expressly requires a party to “supplement or correct” such a disclosure “if the party
learns that in some material respect the disclosure . . . is incomplete or incorrect,” Fed. R. Civ. P.
26(e)(1), and Defendants do not contend that Plaintiffs and their expert failed to make the
necessary supplementation or correction within the established time limit for doing so.
To be sure, there can come a point that an expert’s supplemental submission amounts to
such a “dramatic, pointed variation” from his initial report as to exceed the permissible purpose
of supplementation or correction under Rule 26(e)(1). Keener v. United States, 181 F.R.D. 639,
641 (D. Mont. 1998); see also In re Ready-Mixed Concrete Antitrust Litigation, 261 F.R.D. 154,
159-60 (S.D. Ind. 2009) (striking an expert submission that was “not ‘supplemental,’ as
contemplated by Rule 26,” but instead “employ[ed] a host of new detailed analyses . . . . , none
of which was developed in the original [expert] report”). Under these circumstances, the courts
have recognized that overbroad “supplementation” under the guise of Rule 26(e)(1) would
undermine the purposes of the expert disclosure provisions set forth in subsection (a)(2) of the
Rule, which are intended to “prevent unfair surprise at trial” and “prevent[] experts from ‘lying
in wait’ to express new opinions at the last minute, thereby denying the opposing party the
opportunity to depose the expert on the new information or closely examine the expert’s new
testimony.” Minebea Co. v. Papst, 231 F.R.D. 3, 5-6 (D.D.C. 2005).
Dr. Ashenfelter’s rebuttal report does not trigger these concerns. It does not alter the
fundamental benchmark approach adopted in Dr. Ashenfelter’s initial report, nor does it advance
any new theories as to impact or damages. Rather, Dr. Ashenfelter has merely refined his
analysis and corrected or augmented some of his calculations in response to certain of the points
raised by Defendants’ experts upon their review of his initial report. The courts have explained
that the filing of such a supplemental report “that fully informs the recipient of the anticipated
testimony of the expert” without “belatedly send[ing] the case on a wholly different tack”
accomplishes the “very purpose” of Rule 26(e), which is “to prevent surprise at trial.” Talbert v.
17
While Defendants point to Professor Rubinfeld’s sur-rebuttal report as describing still
more costs that Dr. Ashenfelter should have deducted in arriving at his estimate of the
“but-for” wages that would have been paid to Plaintiffs in the absence of Defendants’
alleged antitrust violations, the Court has determined that this sur-rebuttal report was
untimely produced and cannot be considered. In any event, Dr. Ashenfelter’s efforts to
account for the Defendant hospitals’ human resources costs are “based upon facts in the
record,” and Defendants’ challenges to the completeness of Dr. Ashenfelter’s review of
this record and the accuracy of his resulting adjustments do not warrant the wholesale
exclusion of Dr. Ashenfelter’s testimony as unreliable, but instead are matters to be
“tested on cross-examination and subjected to further scrutiny and criticism by
Defendants’ own expert” at trial. In re Scrap Metal, 527 F.3d at 530-31; see also In re
Northwest Airlines, 197 F. Supp.2d at 927 (“[T]o the extent that Defendants and their
experts have applied a similar methodology and merely reached a different conclusion,
such a ‘battle of the experts’ must be resolved by the trier of fact.”).
Similarly, Defendants’ challenge to the accuracy of Dr. Ashenfelter’s “flexibility
premium” — an adjustment to the agency fee that is intended to account for the additional
value provided by agency nurses who perform more temporary or flexible RN services,
City of Chicago, 236 F.R.D. 415, 421, 424 (N.D. Ill. 2006); see also Crowley v. Chait, 322 F.
Supp.2d 530, 540 (D.N.J. 2004) (observing that “Daubert does not require that an expert’s
testimony be excluded simply because he admitted and corrected his own mistakes,” but that, to
the contrary, such error correction “strengthens the quality of the expert report”). Nor can
Defendants claim any prejudice from any “last minute” disclosures in Dr. Ashenfelter’s rebuttal
report, where they had the opportunity to depose him after the production of this report. Thus,
the Court rejects Defendants’ characterization of this rebuttal report as an improper “do-over.”
18
(see Ashenfelter Report at ¶ 115) — does not provide a basis for excluding Dr.
Ashenfelter’s proposed testimony as unreliable. As noted earlier, Dr. Ashenfelter’s initial
report estimated this flexibility premium as 11.2 percent, (id. at ¶ 115), but he revised this
figure to 18 percent in his rebuttal report, (see Ashenfelter Rebuttal Report at ¶ 67), and
then to 21.5 percent in the errata to his rebuttal report, (see Defendants’ Motion, Ex. F,
Errata Sheet at 1). Notably, this 21.5 percent figure exceeds the 19.5 percent flexibility
premium estimated by Defendant’s expert, Professor Rubinfeld, (see Defendants’ Motion,
Ex. H, Rubinfeld Report at ¶ 137 n.175), and incorporates the sole substantive critique
offered by Professor Rubinfeld to Dr. Ashenfelter’s calculation of this premium.
To be sure, Defendants point to errors in the statistical analysis through which Dr.
Ashenfelter sought to address Professor Rubinfeld’s critique — errors that Dr.
Ashenfelter acknowledged and attempted to correct in the errata to his rebuttal report —
and they further maintain that Dr. Ashenfelter’s estimated flexibility premium, even as
revised and increased in his rebuttal report and errata, still remains an “inadequate
deduction” that fails to adequate capture all of the differences between wages paid to RN
employees and fees paid for agency nurses. (Defendants’ Motion, Br. in Support at 16.)
Yet, as stated by Defendants’ own expert, Professor Rubinfeld, these are “data and
implementation flaws” that reflect Dr. Ashenfelter’s purported “fail[ure] to account
appropriately for deductions from the agency bill rate” paid by the Defendant hospitals
“for items such as agency margins or actual premium[s] earned by agency nurses relative
to internal pool nurses or permanent employees.” (Rubinfeld Report at ¶ 133 & n.163.)
19
Along the same lines, Defendants assert in their motion that these purported flaws in Dr.
Ashenfelter’s expert analysis have led him to “overstate[]” the damages allegedly suffered
by the plaintiff class. (Defendants’ Motion, Br. in Support at 16.) These quarrels with
the accuracy of Dr. Ashenfelter’s calculations, as opposed to his underlying benchmark
approach to estimating RN wages in the absence of Defendants’ alleged antitrust
violations, must be left for the trier of fact to resolve, and do not provide a basis for the
exclusion of Dr. Ashenfelter’s testimony under Rule 702. See Conwood Co., 290 F.3d at
794 (holding that an expert’s failure to consider additional variables identified as
important by an opposing party’s expert normally affects only the probative value of the
expert’s testimony, and not its admissibility).
Moving beyond the issue of the specific adjustments that Dr. Ashenfelter made (or
failed to make) to his agency fee benchmark, Defendants next contend that the “one size
fits all” nature of this benchmark disqualifies it as a reliable measure of RN wages in a
market free from Defendants’ alleged antitrust conspiracy. As Defendants observe, Dr.
Ashenfelter’s benchmark approach generates a single “but-for” wage figure
encompassing all nurses who worked at a given Defendant hospital in a given year. In
reality, however — and as Dr. Ashenfelter acknowledged at his deposition, (see
Defendants’ Motion, Ex. D, Ashenfelter 3/31/2009 Dep. at 342-43) — RN wages vary
widely within a hospital, or even a single department, and this presumably would remain
true in the “but-for” world in which the Defendant hospitals had not engaged in an
alleged antitrust conspiracy. In Defendants’ view, this disjunction between Dr.
20
Ashenfelter’s benchmark and the real world disqualifies Dr. Ashenfelter’s analysis as a
reliable account of the antitrust injury allegedly suffered by the plaintiff class, where this
analysis would “lead[] to bizarre and arbitrary results” such as, for example, the most
experienced nurses having the smallest damages. (Defendants’ Motion, Br. in Support at
18.)
Plaintiffs correctly point out, however, that this critique of Dr. Ashenfelter’s
benchmark analysis derives from its nature as an admittedly conservative estimate of
“but-for” RN wages. Because Dr. Ashenfelter’s benchmark rests upon the fees paid by
the Defendant hospitals for agency nurses, it measures only the value of “generic” nursing
services provided by these agency nurses, and Dr. Ashenfelter has acknowledged that this
may result in “understat[ing] the losses of experienced nurses” as compared to the losses
suffered by their less experienced counterparts. (Ashenfelter Rebuttal Report at ¶ 72; see
also Ashenfelter 3/31/2009 Dep. at 345-48.) As Plaintiffs observe, so long as Dr.
Ashenfelter is able to persuade the trier of fact that his benchmark provides a truly
conservative estimate of but-for RN wages — an assertion that Defendants, of course, are
free to challenge, both through cross-examination and through the testimony of their own
experts — this will suffice to establish that Defendants’ alleged antitrust violations had a
common impact on the members of the plaintiff class, even if this benchmark might not
accurately measure the precise harm suffered by each individual class members. It
follows that Dr. Ashenfelter’s testimony is admissible as to this issue of common impact.
Moreover, the Court agrees with Plaintiffs that this acknowledged imprecision in
21
Dr. Ashenfelter’s approach does not preclude him from testifying with sufficient
reliability as to the damages suffered by the plaintiff class. When questioned on this
subject at his deposition, Dr. Ashenfelter recognized that it was “certainly possible” to
construct a benchmark that attempted to more precisely measure the variance in “but-for”
wages paid to nurses with differing levels of experience, skill, and training who work at
the same Defendant hospital, but he opined that this approach would be less reliable due
to the “potential . . . arbitrariness of the assumptions” he would need to make and the
resulting introduction of “potential error that [he] wouldn’t be able to quantify.”
(Ashenfelter 3/31/2009 Dep. at 344-45.) Again, if the trier of fact accepts this testimony,
as well as Dr. Ashenfelter’s more general characterization of his approach as providing a
“lower bound estimate” of the losses suffered by the members of the plaintiff class, (id. at
345), the Defendant hospitals will not be called upon to pay damages in excess of the
harm inflicted on the plaintiff class as a result of their alleged antitrust violations. As
Plaintiffs aptly observe, Defendants’ stated concern over this purportedly “arbitrary”
feature of Dr. Ashenfelter’s analysis surely does not arise from any notions of
“unfairness” to some class members, such as highly experienced nurses, who “may
receive damages that underestimate their ‘true’ losses,” but instead seeks to ensure,
through the wholesale exclusion of Dr. Ashenfelter’s testimony, that the members of the
plaintiff class “can never recover anything for their injuries.” (Plaintiffs’ Response Br. at
31.) The Court finds nothing in Rule 702 or the case law that would mandate this result;
to the contrary, in In re Scrap Metal, 527 F.3d at 534, the Sixth Circuit upheld an
22
aggregate measure of damages in an antitrust suit that rested upon a “uniform-impact”
theory similar to that advanced by Dr. Ashenfelter here.
Next, Defendants challenge the reliability of Dr. Ashenfelter’s benchmark analysis
as purportedly dependent upon a theory of “collusive price discrimination” that is
“completely unsupported by the facts.” (Defendants’ Motion, Br. in Support at 20.) As
Plaintiffs point out in response, however, the phrase “collusive price discrimination”
appears nowhere in Dr. Ashenfelter’s initial or rebuttal reports, nor is it accurate to say
that his benchmark analysis “relies” on any such theory of “collusive price
discrimination.” Rather, Dr. Ashenfelter merely mentions the economic behavior of price
discrimination in his rebuttal report in response to a point raised by Defendants’ experts
— or, as Dr. Ashenfelter puts it, to explain “how it could be that the defendant hospitals
are willing to pay so much more for agency nurses than they are willing to pay their
regular RNs.” (Ashenfelter Rebuttal Report at ¶ 31.) This discussion of price
discrimination, then, is largely peripheral to Dr. Ashenfelter’s benchmark analysis, and it
should be left to the trier of fact to resolve this “battle of the experts” and determine the
adequacy of Dr. Ashenfelter’s explanation in response to the critique of Defendants’
experts.
Defendants next contend that Dr. Ashenfelter’s benchmark analysis rests on an
unfounded assumption that the Defendant hospitals make “extensive” use of agency
nurses, (Ashenfelter Report at ¶ 103), without any effort to confirm the validity of this
assumption. Yet, in the very next paragraph of his expert report, Dr. Ashenfelter cites
23
record evidence of the millions of dollars spent annually by the Defendant hospitals for
agency nurses and the large numbers of hours worked by agency nurses at the Defendant
hospitals during the relevant period. (Id. at ¶ 104.) While Defendants fault Dr.
Ashenfelter for failing to compare the Defendant hospitals’ use of agency nurses to the
use rate of agency nurses at hospitals elsewhere in the country, Plaintiffs correctly
observe that Dr. Ashenfelter’s benchmark analysis does not require that the word
“extensive” be construed as “more than in other cities.” (Plaintiffs’ Response Br. at 37.)
Rather, the evidence of Defendants’ significant use of agency nurses provides a sufficient
basis in the record for allowing the trier of fact to consider Dr. Ashenfelter’s benchmark
analysis and determine the weight it should be given.
Finally, Defendants assert that Dr. Ashenfelter’s admitted failure to test the
methodology underlying his benchmark analysis renders this analysis unreliable and
inadmissible. Yet, while the Daubert standard includes testing as one of the factors a
court may consider in determining the admissibility of an expert’s testimony, both the
Supreme Court and the Sixth Circuit have emphasized that the factors cited in Daubert
“do not constitute a definitive checklist or test,” and that “the Rule 702 inquiry [i]s a
flexible one.” Kumho Tire, 526 U.S. at 150, 119 S. Ct. at 1175 (emphasis in original)
(internal quotation marks and citations omitted); see also In re Scrap Metal, 527 F.3d at
529. More specifically, although a lack of testing may render an expert’s opinion
unreliable where the expert’s theory “easily lends itself to testing and substantiation,”
Dhillon v. Crown Controls Corp., 269 F.3d 865, 870 (7th Cir. 2001), the courts have
24
recognized that this Daubert factor may not apply as well, if at all, to the social sciences,
in which theories are less susceptible to “ideal experimental conditions and controls,”
United States v. Simmons, 470 F.3d 1115, 1123 (5th Cir. 2006); see also Isely v.
Capuchin Province, 877 F. Supp. 1055, 1065-66 (E.D. Mich. 1995) (permitting a
psychological expert to testify regarding repressed memory, despite the expert’s
acknowledgment that her theory of repressed memory “cannot be tested empirically”);
Voilas v. General Motors Corp., 73 F. Supp.2d 452, 461 (D.N.J. 1999) (observing that
outside the hard sciences, “theories are often not subject to testing or experimentation”
(internal quotation marks and citation omitted)).
In this case, Defendants evidently suggest that Dr. Ashenfelter could have tested
his benchmark approach by examining agency fees and nurse wages paid by hospitals
outside the Detroit metropolitan area. As Plaintiffs observe, however, Defendants’ own
expert, Professor Rubinfeld, attempted to perform such an analysis, and Dr. Ashenfelter
has rejected this analysis as “unreliable” in light of the lack of credible data as to “what
agencies charge in cities other than Detroit.” (Ashenfelter Rebuttal Report at ¶ 58.)
Although the trier of fact certainly need not credit Dr. Ashenfelter’s view that such a test
cannot be reliably conducted due to the limited available data as to agency fees paid by
hospitals in other markets, this nonetheless is another “battle of the experts” that is not
appropriate for resolution in the context of a Rule 702 motion to exclude an expert’s
testimony.
C.
Defendants’ Remaining Challenges to Dr. Ashenfelter’s Economic Analysis
25
Do Not Warrant the Exclusion of His Proposed Expert Testimony on These
Subjects.
As observed earlier, the bulk of Defendants’ challenges to Dr. Ashenfelter’s
proposed expert testimony are directed at the benchmark approach through which he
proposes to demonstrate the common impact of Defendants’ alleged antitrust violations
upon the members of the plaintiff class and the damages suffered by the class. Beyond
these critiques, however, Defendants also contend that other aspects of Dr. Ashenfelter’s
economic analysis are subject to exclusion under Rule 702 as unreliable. The Court
disagrees, and instead finds, once again, that these remaining challenges must be left for
resolution by the trier of fact.
As the first of these challenges, Defendants argue that Dr. Ashenfelter’s analysis of
the utilization rate of RNs in the Detroit metropolitan area should be excluded because of
his purported failure to account for alternative explanations for the low utilization rate
identified in his expert report. Dr. Ashenfelter opines in his report that this low utilization
rate in the Detroit area, as compared to “mean RN utilization in other [U.S.] cities,”
supports the conclusion that “utilization of RNs is below the competitive level” in the
Detroit metropolitan area, and thus “provides further confirmation that fewer RNs are
working in hospital jobs [in the Detroit area] than would be expected in the absence of the
alleged information exchange conspiracy.” (Ashenfelter Report at ¶¶ 128, 130, 144.) In
Defendants’ view, however, Dr. Ashenfelter has failed to identify a reliable basis for the
conclusion he draws from this utilization analysis, where the record reveals that at least
26
thirty other cities encompassed within this analysis have RN utilization rates lower than
Detroit’s. This suggests, according to Defendants, that these low utilization rates in
Detroit and several other cities might well be attributable to factors other than an alleged
antitrust conspiracy — unless, of course, one is prepared to assume that the hospitals in
these other cities have engaged in similar antitrust violations — yet Dr. Ashenfelter
acknowledges that he did not look into the possible causes of the low utilization rates in
other cities, and thus could not say whether factors apart from allegedly anticompetitive
conduct might be responsible for these low rates.
As Plaintiffs observe, however, this argument reads Dr. Ashenfelter’s RN
utilization study in isolation, rather than as part of a larger overall analysis in which Dr.
Ashenfelter relies upon both Detroit’s relatively low RN utilization rate and the
purportedly depressed RN wages paid by the Defendant Detroit-area hospitals to
conclude that the Detroit-area RN market suffers from each of “the two anticompetitive
effects that can be expected to result from an exercise of market power in a labor market.”
(Ashenfelter Report at ¶¶ 125-28.) This addresses, at least indirectly, the only alternative
explanation specifically identified by Defendants and their experts — namely, that low
RN utilization may also be attributable to “higher than competitive wages.” (Defendants’
Motion, Br. in Support at 7.) Because Dr. Ashenfelter’s finding of depressed RN wages
rules out this alternative, his utilization analysis is not subject to exclusion on the ground
that it fails to account for this alternative explanation for low RN utilization in the Detroit
area. In any event, and as observed earlier, any such purported failure to address and rule
27
out other possible causes of the injury allegedly suffered by the plaintiff class affects only
the weight, and not the admissibility, of Dr. Ashenfelter’s proposed expert testimony. See
Conwood Co., 290 F.3d at 794; see also Jahn v. Equine Services, PSC, 233 F.3d 382, 390
(6th Cir. 2000) (“The fact that several possible causes might remain uneliminated only
goes to the accuracy of the conclusion, not to the soundness of the methodology.”
(internal quotation marks, alteration, and citation omitted)).
Next, Defendants challenge Dr. Ashenfelter’s analyses of the relevant geographic
and product markets as based on insufficient investigation or data. Yet, as a threshold
matter, Plaintiffs point out — and the Court likewise has recognized in an earlier ruling in
this case — that “proof of actual detrimental effects . . . can obviate the need for an
inquiry into market power, which is but a surrogate for detrimental effects.” CasonMerenda, 862 F. Supp.2d at 648 (internal quotation marks and citation omitted). Because
Dr. Ashenfelter’s benchmark analysis, “if credited, serves as direct proof of a detrimental
impact upon the wages paid to RNs by the Defendant hospitals,” any purported
weaknesses in Dr. Ashenfelter’s analysis of the relevant market and of the Defendant
hospitals’ share of that market would not defeat Plaintiffs’ remaining rule-of-reason
claim. 862 F. Supp.2d at 648.
In any event, the Court finds that the purported weaknesses identified by
Defendants in Dr. Ashenfelter’s market analysis affect only the weight, and not the
admissibility, of Dr. Ashenfelter’s proposed testimony on this subject. Defendants do not
dispute that the elasticity analysis upon which Dr. Ashenfelter primarily relies, (see
28
Ashenfelter Report at ¶¶ 152-54), is an accepted method for defining a relevant
geographic market.10 While Defendants suggest additional factors Dr. Ashenfelter should
have considered in his market analysis, this sort of challenge is amenable to exploration
on cross-examination of Dr. Ashenfelter at trial. Similarly, to the extent that Defendants’
experts have proffered a different analysis and definition of the relevant geographic
market, Dr. Ashenfelter has in turn identified a number of purported flaws in this
analysis, (see Ashenfelter Rebuttal Report at ¶¶ 127-39), and the resulting “battle of the
experts” must be resolved by the trier of fact. Finally, as to Defendants’ contention that
Dr. Ashenfelter’s definition of the relevant product market is undermined by his failure to
properly consider whether this market should be narrower — e.g., broken into subclasses
of RNs in different hospital departments — or broader — e.g., expanded to include RNs
in non-hospital settings — Plaintiffs observe that the definition of a product market is
driven by “economic realities and industry practice.” Spirit Airlines, Inc. v. Northwest
Airlines, Inc., 431 F.3d 917, 933 (6th Cir. 2005). As this Court has previously
recognized, “to see that Dr. Ashenfelter’s exclusion of non-hospital RNs [from the
relevant product market] was reasonable, one need look no further than the Defendant
hospitals’ own conduct in commissioning and conducting wage surveys” that nearly
10
Defendants do take issue, however, with Dr. Ashenfelter’s failure to conduct an
econometric study of the elasticity of the supply of nurses. Yet, Dr. Ashenfelter testified at his
deposition that he chose not to do so because “there just wasn’t enough data . . . to do something
that would be transparent and credible,” (Ashenfelter 1/6/2009 Dep. at 149), and Plaintiffs point
out that “none of Defendants’ three economists have attempted any such analysis to support their
discussion of market definition,” (Plaintiffs’ Response Br. at 42).
29
always sought “information only on what other hospitals [we]re paying their hospital
nurses.” Cason-Merenda, 862 F. Supp.2d at 648-49 (emphasis in original) (internal
quotation marks and citation omitted).
Finally, Defendants challenge the admissibility of an “inverse elasticity” analysis
performed by Dr. Ashenfelter, from which he concludes that collusion to depress the
wages of a “substantial subgroup” within the RN class would result in lower wages for all
jobs in the class. (See Ashenfelter Report at ¶¶ 95-102.) Again, however, this challenge
rests in part on Dr. Ashenfelter’s purported failure to include certain variables in his
analysis, and such omissions affect only the probative value of this analysis. To the
extent that Defendants fault Dr. Ashenfelter for failing to consider cross-elasticity of the
supply of nurses among specific pairs of hospital departments — e.g., between the
medical-surgical and intensive care units, (see Defendants’ Motion, Br. in Support at 30)
— Plaintiffs correctly respond that it is irrelevant, for purposes of Dr. Ashenfelter’s
analysis, to know how many nurses would move from one specific department to another
in response to changes in wages; rather, it only matters whether nurses would move
generally among departments in response to this change.11 Dr. Ashenfelter cites a basis in
the record for his calculation of this “inverse elasticity” figure, (see Ashenfelter Report at
¶ 102), and Defendants remain free to cross-examine him as to other record evidence that
11
Plaintiffs further suggest that “it would require an unfathomable amount of data” to
conduct the specific department-pair analysis proposed by Defendants. (Plaintiffs’ Response Br.
at 43.)
30
would undermine this calculation. See McLean v. 988011 Ontario, Ltd., 224 F.3d 797,
801 (6th Cir. 2000) (emphasizing that “mere weaknesses in the factual basis of an expert
witness’ opinion bear on the weight of the evidence rather than on its admissibility”
(internal quotation marks, alteration, and citation omitted)).
IV. CONCLUSION
For the reasons set forth above,
NOW, THEREFORE, IT IS HEREBY ORDERED that Defendants’ April 24,
2009 motion to exclude the expert testimony of Orley Ashenfelter (docket #349) is
DENIED.
s/Gerald E. Rosen
Chief Judge, United States District Court
Dated: April 22, 2013
I hereby certify that a copy of the foregoing document was served upon the parties and/or
counsel of record on April 22, 2013, by electronic and/or ordinary mail.
s/Julie Owens
Case Manager, (313) 234-5135
31
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?