Food Lion, LLC v. Dean Foods Company

RECOMMENDED FOR FULL-TEXT PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 14a0001p.06
UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT
_________________
X
-
In re: SOUTHEASTERN MILK ANTITRUST
-
LITIGATION
_____________________________________ -
-
No. 12-5457
FOOD LION, LLC and FIDEL BRETO, on behalf ,>
-
Plaintiffs-Appellants, -
of himself and all others similarly situated,
-
-
-
-
v.
-
-
DEAN FOODS COMPANY, NATIONAL DAIRY
-
HOLDINGS, L.P., DAIRY FARMERS OF
AMERICA, INC., DAIRY MARKETING -
-
-
SERVICES, LLC, and SOUTHERN MARKETING
Defendants-Appellees. -
AGENCY, INC.,
-
N
Appeal from the United States District Court
for the Eastern District of Tennessee at Greeneville.
Nos. 2:07-cv-00188; 2:08-md-01000—J. Ronnie Greer, District Judge.
Argued: July 25, 2013
Decided and Filed: January 3, 2014
Before: ROGERS and COOK, Circuit Judges; VAN TATENHOVE, District Judge.*
_________________
COUNSEL
ARGUED: Neil K. Gilman, HUNTON & WILLIAMS LLP, Washington, D.C., for
Appellants. Paul H. Friedman, DECHERT LLP, Washington, D.C., for Appellees.
ON BRIEF: Neil K. Gilman, Richard L. Wyatt, Jr., Todd M. Stenerson, HUNTON &
WILLIAMS LLP, Washington, D.C., Gordon Ball, BALL & SCOTT, Knoxville,
Tennessee, R. Laurence Macon, AKIN GUMP STRAUSS HAUER & FELD LLP, San
Antonio, Texas, for Appellants. Paul H. Friedman, DECHERT LLP, Washington, D.C.,
*
The Honorable Gregory F. Van Tatenhove, United States District Judge for the Eastern District
of Kentucky, sitting by designation.
1
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 2
Carolyn Hazard Feeney, DECHERT LLP, Philadelphia, Pennsylvania, Steven R. Kuney,
WILLIAMS & CONNOLLY LLP, Washington, D.C., W. Todd Miller, BAKER &
MILLER PLLC, Washington, D.C., Jerry L. Beane, Kay Lynn Brumbaugh, ANDREWS
KURTH LLP, Dallas, Texas, for Appellees.
_________________
OPINION
_________________
GREGORY F. VAN TATENHOVE, District Judge. Dean Foods Company and
Suiza Foods Corporation were the two largest processed milk bottlers in the country in
2001. At that time, they announced plans to merge their operations, which the
Department of Justice approved subject to divestment of particular milk processing
plants. The merged company, now known as Dean Foods, is accused of violating
15 U.S.C. § 1 of the Sherman Antitrust Act by conspiring with a raw milk supplier/milk
processor and the purchaser of the divested processing facilities to divide markets and
restrict output. The district court granted summary judgment for Defendants, ruling that
Plaintiffs could not provide sufficient proof of injury, nor could they establish the
relevant antitrust geographic market, primarily because their expert’s testimony was
excluded. Two retailers of processed milk, Food Lion LLC and Fidel Breto, appeal both
of these conclusions. For the following reasons, we REVERSE and REMAND.
I
A
Prior to 2001, Dean Foods Company and Suiza Foods Corporation competed to
process and sell bottled milk to retailers. Suiza was the largest processor of milk in the
United States, and Dean Foods was the second largest. Both processors purchased their
raw milk from other entities. Dairy Farmers of America (“DFA”), a dairy farmer
cooperative, was Suiza’s primary supplier and business partner, owning almost 34% of
Suiza Dairy Group, which was a subsidiary of Suiza’s. Dean Foods obtained its raw
milk predominantly from independent farmers.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 3
Dean Foods and Suiza merged in 2001 under the name Dean Foods, hoping to
obtain “distribution efficiencies and economies of scale,” which would result in millions
of dollars in cost savings. As they began consolidating, certain agreements were
negotiated, with input from the Department of Justice, to avoid antitrust problems. To
secure financing for the merger, Suiza purchased DFA’s ownership interest in exchange
for cash, six dairy processing plants, and two contractual provisions related to DFA’s
ability to provide raw milk to the merging companies’ processing plants. One provision
promised DFA a specific sum of money if its supply contracts for plants previously
owned by Suiza were terminated within twenty years. The other provision stipulated
that Dean Foods would owe DFA liquidated damages if DFA were not provided an
opportunity to supply raw milk for the plants Dean Foods owned prior to the merger.
The six processing plants that DFA received from Suiza were quickly transferred
to a newly formed partnership called National Dairy Holdings (“NDH”), which DFA
partly owned. NDH was formed to compete with Dean Foods, and after it added five
more processing plants Dean Foods divested, it became the second largest milk bottler
in the southeast. NDH had four owners. Two owners were former Suiza executives, and
one was a former business partner of DFA’s chief executive officer. Together, they
owned a 50% equity interest. DFA owned the other 50% equity interest, and it
possessed the power to “veto any agreement that would substantially affect the operation
of NDH, contracts, or capital expenditures greater than $50,000, and the acquisition,
expansion or disposal of NDH’s facilities.” The Department of Justice’s Antitrust
Division approved NDH’s purchase and operation of the eleven plants.
These facts set the stage for the illegal conspiracy that Food Lion and Fidel Breto
(“Plaintiffs”), two retailers of processed milk, have alleged — a conspiracy in which
DFA serves as the keystone. With NDH as Dean Foods’ largest competitor, it would
stand to reason that if NDH were weakened, Dean Foods would enjoy a stronger position
in the marketplace for selling processed milk. Although DFA’s ownership stake
provides an obvious incentive to fully support NDH’s fledgling enterprise, DFA’s raw
milk supply agreements with the merged company create fertile soil for the development
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 4
of a conflict of interest. Supported by several disputed factual allegations, the essence
of Plaintiffs’ conspiracy claim is as follows:
NDH knowingly accepted ‘second best’ plants, operated those plants at
losses and eventually shuttered some of those plants in an unlawful
agreement with its competitor Dean/Suiza because, in return, its parent
company, DFA, received a commitment from Dean/Suiza to allow it to
supply raw milk to each Dean/Suiza bottling plant, including the pre-
merger Dean plants previously supplied by independent dairy farmers.
[Appellant Br. at 15.]
B
The Plaintiffs originally brought suit in the district court based on five claims,
alleging violations of Sections 1 and 2 of the Sherman Antitrust Act and Section 3 of the
Clayton Act. The district court granted summary judgment to the Defendants on Counts
II, III, and IV, but denied summary judgment on Counts I and V. After the close of
discovery, the Defendants again moved for summary judgment on several additional
grounds that had not been raised previously. The district court found the additional
arguments Defendants raised to be convincing and granted summary judgment in favor
of Defendants on Counts I and V, ruling that the Plaintiffs failed to meet the
requirements for establishing an antitrust violation under Section 1 of the Sherman
Antitrust Act. Plaintiffs now appeal the district court’s ruling on Count I.
In Count I, the Plaintiffs allege that the Defendants engaged in a conspiracy not
to compete, in violation of 15 U.S.C. § 1. For a plaintiff to successfully bring an
antitrust claim under Section 1 of the Sherman Act, the plaintiff must establish that the
defendant’s actions constituted an unreasonable restraint of trade which caused the
plaintiff to experience an antitrust injury. Expert Masonry, Inc. v. Boone County,
Kentucky, 440 F.3d 336, 342 (6th Cir. 2006) (quoting Crane & Shovel Sales Corp. v.
Bucyrus-Erie Co., 854 F.2d 802, 805 (6th Cir. 1988)). In its second summary judgment
opinion, the district court found that: 1) Plaintiffs failed to establish that the restraint on
trade was unreasonable, largely because the court excluded the testimony of Plaintiffs’
expert; and 2) that Plaintiffs failed to establish the requisite element of injury. In re
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 5
Southeastern Milk Antitrust Litig., 2:08-MD-1000, 2012 WL 1032797, at *6, 12 (E.D.
Tenn. Mar. 27, 2012). Those decisions are the subject of this review.
II
This Court reviews the district court’s grant of summary judgment de novo. Yet
it must be mindful that “[i]n this circuit, courts are generally reluctant to use summary
judgment dispositions in antitrust actions due to the critical ‘role that intent and motive
have in antitrust claims and the difficulty of proving conspiracy by means other than
factual inference.’” Expert Masonry, Inc. v. Boone County, Kentucky, 440 F.3d 336,
341(6th Cir. 2006) (quoting Smith v. N. Mich. Hosp., Inc., 703 F.2d 942, 947 (6th Cir.
1983)).
A
Unfortunately, there is no general agreement on the exact standards to use when
resolving antitrust cases. As much as we might wish that a precise process with clear
elements existed, antitrust cases in this circuit, and in others, apply various approaches
to adjudicating antitrust claims. There are some areas of consensus, however. A good
starting point is the statute itself. Section 1 of the Sherman Act states that “Every
contract, combination in the form of trust or otherwise, or conspiracy, in restraint of
trade or commerce among the several States, or with foreign nations, is declared to be
illegal.” 15 U.S.C. § 1. The plaintiffs must first show, therefore, that an agreement
between two or more economic entities exists since unilateral conduct would not violate
this statute. Nat’l Hockey League Players Ass’n v. Plymouth Whalers Hockey Club, 419
F.3d 462, 469 (6th Cir. 2005).
Next, because nearly every agreement between parties could be considered a
restraint of trade, the Supreme Court has limited Section 1 to apply only to
“unreasonable” restraints of trade. Nat’l Hockey League Players, 419 F.3d at 469 (citing
Nat’l Collegiate Athletic Ass’n v. Board of Regents of Univ. of Okla., 468 U.S. 85, 98
(1984)). Whether the restraint is “unreasonable” is determined by one of two approaches
— either the per se rule or the “rule of reason.” Id. at 469; Care Heating & Cooling,
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 6
Inc. v. American Standard, Inc., 427 F.3d 1008, 1012 (6th Cir. 2005). If the rule of
reason is used, plaintiffs must additionally show that the restraint produced
anticompetitive effects within the relevant product and geographic markets, while the
per se rule is reserved for restraints that are so clearly unreasonable that their
anticompetitive effects within geographic and product markets are inferred. Expert
Masonry, 440 F.3d at 342-43. Finally, regardless of which approach is used, the plaintiff
must also establish that the illegal conspiracy caused injury to the plaintiff. Id. at 342,
345 (quoting Crane & Shovel Sales Corp. v. Bucyrus-Erie Co., 854 F.2d 802, 805 (6th
Cir. 1988)); see also In re Cardizem CD Antitrust Litigation, 332 F.3d 896, 909 n.15 (6th
Cir. 2003) (noting that the per se rule does not negate the need for plaintiffs to establish
antitrust injury).
In the case at hand, the parties do not contest the first required element — the
existence of a conspiracy. The district court found enough evidence of a conspiracy for
Plaintiffs to persist past summary judgment on that element of their § 1 claim, and
accordingly, that issue is not challenged here. The dispute instead centers on whether
the district court erred in applying the rule of reason instead of the per se rule. Even if
the conspiracy at issue is not a per se violation, Plaintiffs maintain that proof of a
geographic market was unnecessary, and thus contest the district court’s finding that
they failed to provide sufficient evidence of an appropriate geographic market. Finally,
the parties also dispute the district court’s finding that Plaintiffs have not provided
sufficient evidence of antitrust injury.
B
As explained above, the Plaintiffs must present evidence showing that the
Defendants’ agreement “unreasonably restrains trade” in order to satisfy the second
requirement of an antitrust claim. Re/Max Intern., Inc. v. Realty One, Inc., 173 F.3d 995,
1012 (6th Cir. 1999). In the case at hand, the district court applied the rule of reason to
determine whether the alleged restraint was unreasonable, a decision which Plaintiffs
now contest. The district court’s decision to use the rule of reason is a question of law,
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 7
see Expert Masonry, 440 F.3d at 342, which we review de novo. See Associated Gen.
Contractors of Ohio, Inc. v. Drabik, 250 F.3d 482, 484 (6th Cir. 2001) (citation omitted).
1
A restraint may be deemed unreasonable “either because it fits within a class of
restraints that has been held to be ‘per se’ unreasonable, or because it violates what has
come to be known as the ‘Rule of Reason.’” FTC v. Indiana Fed’n of Dentists, 476 U.S.
447, 457-58 (1986) (quoting Chicago Bd. of Trade v. United States, 246 U.S. 231, 238
(1918)). The less common method of determining whether the restraint is unreasonable
is the per se rule. “Certain agreements, such as horizontal price fixing and market
allocation, are thought so inherently anticompetitive that each is illegal per se without
inquiry into the harm it has actually caused.” In re Cardizem CD, 332 F.3d at 907
(quoting Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752, 768 (1984)).
The per se rule should only be used when the restraint has “such predictable and
pernicious anticompetitive effect,” that there is “limited potential for procompetitive
benefit.” Id. (quoting State Oil Co., 522 U.S. at 10). Once applied, “no consideration
is given to the intent behind the restraint, to any claimed pro-competitive justifications,
or to the restraint’s actual effect on competition.” Id. at 906-07 (quoting Nat’l College
Athletic Ass’n v. Board of Regents, 468 U.S. 85, 100 (1984)). Applying this standard,
then, should be done reluctantly and infrequently, informed by other courts’ review of
the same type of restraint, and only when the rule of reason would likely justify the same
result. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877, 886-87
(2007) (citations omitted); see also Nat’l Hockey League Players, 325 F.3d at 718-19
(cautioning that the Supreme Court has described the per se rule as a “demanding” rule
that should be applied “only in clear cut cases”) (citations omitted).
Unless the restraint falls squarely into a per se category, the rule of reason should
be used instead. Expert Masonry, 440 F.3d at 343. Unlike the per se rule, the rule of
reason utilizes a burden-shifting framework that allows the court to “analyze the history
of the restraint and the restraint’s effect on competition.” Nat’l Hockey League Players,
325 F.3d at 718. First, the plaintiff must establish a prima facie case by showing five
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 8
elements: (1) a conspiracy (2) that produced anticompetitive effects; (3) that the scheme
“affected relevant product and geographic markets”; (4) that the conspiracy’s goal and
related conduct was illegal; (5) and that the restraint was the proximate cause of the
plaintiff’s antitrust injury. Expert Masonry, 440 F.3d at 343 (citing Care Heating &
Cooling, 427 F.3d at 1014). If a plaintiff passes over these hurdles, the burden then
shifts to the defendant to produce evidence that the restraint in question has
“procompetitive effects” that are sufficient “to justfi[y] the otherwise anticompetitive
injuries.” Expert Masonry, 440 F.3d at 343 (quoting Nat’l Hockey League Players,
325 F.3d at 718). Finally, if the defendant meets this burden, the plaintiff may still
prevail by showing that “any legitimate objectives can be achieved in a substantially less
restrictive manner.” Id. (quoting Nat’l Hockey League Players, 325 F.3d at 718)
(quotation marks omitted).
When determining whether to use the per se rule or the rule of reason, courts
must consider the type of restraint at issue — whether it is horizontal or vertical. Expert
Masonry, 440 F.3d at 344. An agreement “between competitors at the same level of the
market structure” is horizontal. Sancap Abrasives Corp. v. Swiss Indus. Abrasives, 19 F.
App’x 181, 191 (6th Cir. 2001) (quoting Crane & Shovel Sales Corp. v. Bucyrus-Erie
Co., 854 F.2d 802, 805-06 (6th Cir. 1988)). Horizontal restraints are considered to be
more threatening, and thus result in per se treatment more regularly. See Expert
Masonry, 440 F.3d at 344 (citing examples from cases). Vertical restraints —
agreements between parties “at different levels of the market structure, such as
manufacturers and distributors”— have more redeeming qualities (e.g., allowing for
distribution efficiencies) and are subjected to the rule of reason. Total Benefits Planning
Agency, Inc. v. Anthem Blue Cross & Blue Shield, 552 F.3d 430, 435 (6th Cir. 2008)
(quoting Leegin Creative Leather Products, 551 U.S. 877); see also Expert Masonry,
440 F.3d at 344-45.
2
After careful consideration, the district court concluded that the agreement at
issue in this case should be scrutinized under the rule of reason because “the essence of
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 9
the agreement alleged by the Plaintiffs is one between Dean in its role as a processor of
bottled milk and DFA in its role as a supplier of raw milk, and that the [ ] supply
agreements for raw milk are central to the completion of the alleged conspiracy.” In re
Southeastern Milk Antitrust Litig., 2012 WL 1032797, at *12. The “substantial vertical
elements” were too significant for the district court to agree with Plaintiffs that the
essence of the conspiracy was horizontal. [Id.]
Plaintiffs have previously characterized the conspiracy as a complex relationship
among DFA, Dean Foods, and NDH. That arrangement, however, necessarily involves
vertical elements in the relationship between DFA and Dean Foods. Yet now on appeal,
Plaintiffs contend that the conspiracy is much simpler than previously alleged, and urge
this Court to find that Defendants have committed a per se violation of the Sherman Act.
In support of this contention, Plaintiffs reassert that the Defendants’ agreement is
horizontal, arguing that “the conspiracy here is the agreement between Dean/Suiza and
NDH not to compete.” This argument, however, is unavailing. Plaintiffs should not be
able to change their characterization of the conspiracy midstream in order to gain a more
favorable outcome.
Plaintiffs concede that the reason NDH would agree to weaken itself is unclear
until NDH’s ownership structure is disclosed. Plaintiffs explain that NDH conspired
with Dean Foods only because DFA owned and controlled NDH, and because the
conspiracy served DFA’s purposes. Assuming Plaintiffs’ theory of their injury is true,
they suffered harm because of the result from the agreement between Dean Foods and
DFA. The conspiracy’s effect on the plaintiff, however, is not the sole means of
determining whether a restraint is horizontal or vertical. The agreement which causes
the effect is determinative. See Business Electronics Corp. v. Sharp Electronics Corp.,
485 U.S. 717, 730 n.4 (1988). For example, “a facially vertical restraint imposed by a
manufacturer only because it has been coerced by a ‘horizontal cartel’ agreement among
his distributors is in reality a horizontal restraint.” Id. In this case, the agreement was
between Dean Foods and DFA — Dean Foods agreed to buy the raw milk it needed from
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 10
DFA, thus creating a vertical relationship, in exchange for DFA’s hampering NDH’s
ability to effectively compete.
Plaintiffs’ explanation that NDH acted as it did because of a horizontal
agreement between Dean Foods and NDH has no logical basis because such an
arrangement would present an agreement whereby Dean Foods simply benefits and NDH
is harmed. Rather, DFA is the sine qua non for this conspiracy; NDH would compete
but for DFA and the non-price restrictions it allegedly imposed on NDH.
“Courts cannot act perfunctorily when distinguishing restraints that merit a per
se approach from those that deserve rule of reason analysis,” and the court may apply
the per se rule “only if a restraint clearly and unquestionably falls within one of the
handful of categories that have been collectively deemed per se anticompetitive.” Expert
Masonry, 440 F.3d 336, 343-44. Here, Plaintiffs have not alleged facts that would place
this situation into one of those limited categories. Rather, the restraint at issue appears
to involve a vertical relationship, thus requiring the Court to apply the rule of reason.
See Care Heating, 427 F.3d at 1013-14.
Moreover, even if the agreement is horizontal in the way Plaintiffs now claim,
applying the rule of reason is the default position and can be applied to horizontal
restraints as well if they do not fit into existing categories of per se violations. E.g.,
F.T.C. v. Indiana Fed’n of Dentists, 476 U.S. at 458-59 (applying the rule of reason to
a horizontal agreement where “the economic impact” of a business practice was “not
immediately obvious”). Indeed, the Sixth Circuit has an “automatic presumption in
favor of the rule of reason standard,” Care Heating, 427 F.3d at 1012 (citing Bus. Elecs.
Corp., 485 U.S. at 726), while the per se rule is reserved only for those infrequent
occasions of “clear cut cases” in which the trade restraint is “so unreasonably
anticompetitive that they present straightforward questions for reviewing courts.” Id.
at 1012.
Finally, summary judgment is only appropriate when there are no genuine issues
of material fact, and the moving party is entitled to judgment as a matter of law. Fed.
R. Civ. P. 56(c); Celotex Corp. v. Catrett, 477 U.S. 317, 323-25 (1986). Here, when
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 11
considering the allegations contained in the parties’ briefs, it appears that a factual
dispute exists as to the exact nature of the conspiracy, and as to whether it was obviously
anticompetitive. For instance, the Defendants have produced evidence that the
agreement at issue may have had procompetitive aspects, which would indicate that this
situation would not fall into the categories of per se unreasonable restraints on trade.
[E.g., Appellee Br. at 42 n.16 (presenting evidence that plant closures reduced costs and
rationalized assets, resulting in cost savings).] Another factual dispute exists as to
whether NDH was truly weakened due to the agreement between Dean Foods and DFA.
[Appellee Br. at 12-13 (referencing evidence of active competition between Dean and
NDH after the merger)]. Depending on the situation, full supply contracts such as those
involved in this case “may well be of economic advantage to buyers as well as to sellers,
and thus indirectly of advantage to the consuming public.” Standard Oil Co. v. United
States, 337 U.S. 293, 306 (1949). Additionally, the Department of Justice fully reviewed
and sanctioned the agreement at issue here, which presumably would not have occurred
if the agreement was a per se unreasonable restraint on trade. Therefore, especially at
the summary judgment stage, this is not a “clear cut” case of an obviously
anticompetitive trade restraint, and thus the district court was correct to apply the default
standard of the rule of reason.
3
As explained above, when applying the rule of reason analysis, plaintiffs
generally must establish the effect on the relevant geographic and product markets.
However, courts have recently begun to view the rule of reason in a broader manner in
certain cases. Plaintiffs contend that even if the Court applies the rule of reason to their
case, under the so-called “quick look” rule of reason analysis, they still should not be
required to prove geographic market because the adverse market effects are implied by
the obvious violation of the Defendants. Once the district court decided that the rule of
reason applied, it granted summary judgment to the Defendants, without addressing
whether a quick-look analysis might be appropriate. “[T]he alleged agreements
challenged by Plaintiffs ought to be subject to the rule of reason analysis, requiring that
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 12
Plaintiffs establish the relevant geographic antitrust market, something they cannot do.
For this reason, Defendants are entitled to summary judgment as to Count I of Plaintiffs’
complaint.” In re Southeastern Milk Antitrust Litig., 2012 WL 1032797, at *12.
Plaintiffs submit that this simple logic equation overlooks the recent deterioration of
clearly defined types of market analyses in favor of a more case-by-case approach.
Realcomp II, Ltd. v. FTC, 635 F.3d 815, 826 (6th Cir. 2011) (“The Court has moved
away from . . . reliance upon fixed categories and toward a continuum, within which the
extent of the inquiry is tailored to the suspect conduct in each particular case.”) (quoting
Polygram Holding, Inc. v. FTC, 416 F.3d 29, 34-34 (D.C. Cir. 2005)) (internal quotation
marks omitted).
This Court has characterized “quick look” analysis as a third type of category
arising from the blurring of the line between per se and rule of reason cases. See Expert
Masonry, 440 F.3d at 343. This less-rigid approach aligns with the Supreme Court’s
recognition of the value of the “quick look” approach as an abbreviated form of the rule
of reason analysis used for situations in which “an observer with even a rudimentary
understanding of economics could conclude that the arrangements in question would
have an anticompetitive effect on customers and markets.” Cal. Dental Ass’n v. FTC,
526 U.S. 756, 770 (1999). Applying this test is useful when the anticompetitive nature
of an agreement is so blatant that a detailed review of the surrounding marketplace
would be unnecessary. Id. at 769-70. In the same way that this analysis occupies
territory between the per se and rule of reason tests, so the burdens and presumptions do
as well. Once anticompetitive behavior is shown to a court’s satisfaction, even without
detailed market analysis, the burden shifts to the defendant who must justify the
agreement at issue on procompetitive grounds by providing some “competitive
justification” for the restraint at issue. Realcomp, 635 F.3d at 825.
Whatever tool is used to judge an agreement, “the essential inquiry remains the
same — whether or not the challenged restraint enhances competition.” Cal. Dental
Ass’n, 526 U.S. at 780 (quoting Nat’l College Athletic Ass’n, 468 U.S. at 104).
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 13
[T]here is generally no categorical line to be drawn between restraints
that give rise to an intuitively obvious inference of anticompetitive effect
and those that call for more detailed treatment. What is required, rather,
is an enquiry meet for the case, looking to the circumstances, details, and
logic of a restraint. The object is to see whether the experience of the
market has been so clear, or necessarily will be, that a confident
conclusion about the principal tendency of a restriction will follow from
a quick (or at least quicker) look, in place of a more sedulous one.
Cal. Dental Ass’n, 526 U.S. at 780-81.
Here, using the quick-look analysis, Plaintiffs do not necessarily need to show
geographic market evidence to defeat summary judgment. The district court did not
distinguish between the two types of rule of reason analysis as explained above — the
full rule of reason analysis and the quick-look form of analysis. See Cal. Dental Ass’n,
526 U.S. at 769- 70. Under the quick-look standard, the Plaintiffs have met their burden
of raising a genuine issue of material fact as to whether Dean Foods violated the antitrust
laws even without establishing the relevant geographic market. In applying the summary
judgment standard, the Court must review the facts and draw all reasonable inferences
in favor of the non-moving party. Logan v. Denny’s, Inc., 259 F.3d 558, 566 (6th Cir.
2001) (citing Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250 (1986)). Accordingly,
when construing the facts and record evidence in Plaintiffs’ favor, the alleged unlawful
conduct has obviously adverse, anticompetitive effects; and for purposes of summary
judgment, the district court should have at least considered the fact that a more detailed
market analysis may not have been required under these circumstances. See Realcomp,
635 F.3d at 825. While it is true that the vertical elements of the Defendants’ agreement
require a rule of reason analysis, the agreement between the horizontal competitors for
the express purpose of limiting competition between them could be viewed as a “facially
anticompetitive restraint,” and the district court should consider this possibility on
remand. Realcomp, 635 F.3d at 827. Even though Dean Foods’ alleged conduct is not
illegal per se, the evidence in the record and the allegations in Plaintiffs’ complaint are
sufficient to shift the burden to Dean Foods to present some procompetitive benefits of
the alleged conduct. See Expert Masonry, 440 F.3d at 343.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 14
C
Under a quick-look analysis, the Plaintiffs do not necessarily need to establish
either product or geographic1 market evidence in order to defeat summary judgment.
In that event, the exclusion of their expert’s testimony would no longer be relevant.
However, the district court may yet determine that a full rule of reason analysis is still
required, in which case Plaintiffs would not be able to establish the relevant market apart
from the testimony of Professor Froeb, whose testimony was excluded by the district
court. Thus, on remand, Professor Froeb’s testimony about geographic market may yet
return to prominence, and therefore we review the decision to exclude it.
In applying the rule of reason to the case at hand, the district court required the
Plaintiffs to establish and define the relevant geographic market, but also held that
Plaintiffs’ expert witness, Professor Luke Froeb, formed his opinion concerning
geographic market by using an unreliable method. In re Southeastern Milk Antitrust
Litig., 2012 WL 1032797, at *12. As a result, Plaintiffs lacked a required element of a
rule-of-reason claim, and Defendants were granted summary judgment. Id. Plaintiffs
contend that the exclusion of Froeb’s testimony was improper.
The district court’s decision to exclude Froeb under the standard required by
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993), is reviewed for abuse of
discretion. Ky. Speedway, LLC v. Nat’l Ass’n of Stock Car Auto Racing, 588 F.3d 908,
915 (6th Cir. 2009). “A district court abuses its discretion when it ‘applies the wrong
legal standard, misapplies the correct legal standard, or relies on clearly erroneous
findings of fact.’” Romberio v. UnumProvident Corp., 385 F. App’x 423, 428 (6th Cir.
2009) (quoting Schachner v. Blue Cross & Blue Shield of Ohio, 77 F.3d 889, 895 (6th
Cir. 1996)). This Court will not find an abuse of discretion unless it has a ‘definite and
firm conviction that the trial court committed a clear error of judgment.’” Id. (quoting
Miami Univ. Wrestling Club v. Miami Univ., 302 F.3d 608, 613 (6th Cir. 2002)). This
issue was clearly appealed and briefed extensively, and for the reasons explained below
1
The definition of product market is not at issue on appeal.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 15
the district court should not have excluded Froeb’s testimony. See Ky. Speedway,
588 F.3d at 918.
1
Admissibility of expert testimony is governed by Federal Rule of Evidence 7022
and informed by the seminal case applying Rule 702, Daubert v. Merrell Dow
Pharmaceuticals, Inc., 509 U.S. 579 (1993). In Daubert, the Supreme Court explained
that “[u]nlike an ordinary witness . . . an expert is permitted wide latitude to offer
opinions, including those that are not based on firsthand knowledge or observation.”
Daubert, 509 U.S. at 593. The Supreme Court also recognized that implicit in the rule
is a district court’s gatekeeping responsibility, “ensuring that an expert’s testimony both
rests on a reliable foundation and is relevant to the task at hand.” Id. at 597.
Geographic market is defined as “the region in which the seller operates, and to
which the purchaser can practicably turn for supplies.” Tampa Elec. Co. v. Nashville
Coal Co., 365 U.S. 320, 327 (1965); see E.I. DuPont de Nemours & Comp. v. Kolon
Indus., Inc., 637 F.3d 435, 442 n.2 (4th Cir. 2011) (explaining that this standard applies
to § 1 Sherman Act claims); see also Spirit Airlines, Inc. v. Northwest Airlines, Inc.,
431 F.3d 917, 932-33 (6th Cir. 2005). Outlining a geographic market entails mapping
an area “within which the defendant’s customers who are affected by the challenged
practice can practicably turn to alternative suppliers if the defendant were to raise its
prices or restrict its output.” Kolon Indus., 637 F.3d at 441-42 (citations omitted). This
process is fact-intensive and focused on the “commercial realities of the industry.”
Brown Shoe Co. v. United States, 370 U.S. 294, 336 (1962). These include
considerations of areas where products are marketed, or where the defendant perceives
that it is competing; any applicable regulatory standards; transportation costs and
challenges such as risk of spoilage, size, or weight; and “other factors bearing upon
2
“A witness who is qualified as an expert by knowledge, skill, experience, training, or education
may testify in the form of an opinion or otherwise if: (a) the expert's scientific, technical, or other
specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;
(b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles
and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case.”
Fed. R. Evid. 702.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 16
where customers might realistically look to buy the product.” Kolon Indus., 637 F.3d
at 442-43 (citations omitted). Finally, a geographic market must be sizeable enough to
be “economically significant.” Brown Shoe, 370 U.S. at 336-37.
The purpose of defining a geographic market is to reveal whether, or to what
extent, market power exists. Thompson v. Metro. Multi-List, Inc., 934 F.2d 1566, 1573
(11th Cir. 1991). Market power is defined as the ability to charge a supracompetitive
price — a price above a firm’s marginal cost. Herbert Hovenkamp, Federal Antitrust
Policy: The Law of Competition and Its Practice, §§ 3.1, 3.1a (4th ed. 2011). Such
power could cover the nation, but it may exist in much smaller areas as well. Compare
United States v. Grinnell Corp., 384 U.S. 563, 575 (1966) (finding a national market),
with United States v. Phila. Nat’l Bank, 374 U.S. 321, 361 (1963) (finding a four-county
region a market). In either case, a geographic market must be drawn to consist of the
smallest area of overlap of Plaintiffs’ and Defendants’ locations, and in which prices
could be increased without retailers turning to alternative suppliers or other milk
processors entering the area. See Hovenkamp, supra, § 3.6.
The hypothetical monopolist is a related concept. See U.S. Dep’t of Justice and
Fed. Trade Comm’n, 1997 Horizontal Merger Guidelines (“Merger Guidelines”); see
also FTC v. Whole Foods Market, Inc., 548 F.3d 1028, 1038 (D.C. Cir. 2008)
(acknowledging the use of this construct when examining a market); Ky. Speedway,
588 F.3d at 918 (same). Using the hypothetical monopolist as an entity that controls all
the suppliers in a proposed market,3 a question is posed: could a monopolist profit if it
imposed a “small but significant non-transitory increase in price” (“SSNIP”)? Id.
Typically, the increase that is posited is five percent. Ky. Speedway, 588 F.3d at 918.
If buyers in a defined area would respond to a small, lasting increase in price — a SSNIP
— by purchasing from another supplier, rendering the SSNIP unprofitable, the market
has been too narrowly defined. See FTC v. Tenet Healthcare Corp., 186 F.3d 1045,
3
“A market is defined as a product or group of products and a geographic area in which it is
produced or sold such that a hypothetical profit-maximizing firm, not subject to price regulation, that was
the only present and future producer or seller of those products in that area would likely impose at least
a ‘small but significant nontransitory’ increase in price, assuming the terms of sale of all other products
are held constant.” Merger Guidelines § 1.0.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 17
1053, n.11 (8th Cir. 1999) (citing Merger Guidelines § 1.21)). Similarly, a market is too
small if additional suppliers would enter a market in response to one firm’s price
increase. Id. at 1052 (citing Bathke, 64 F.3d at 346). In either of those cases, the
question about a SSNIP must be reconsidered and applied to a wider area.
This should continue only until buyers in the relevant market respond to a SSNIP
by purchasing regardless of the increase. Merger Guidelines § 1.0. Although the
Merger Guidelines, including the hypothetical monopolist, are useful and informative
for courts in analyzing some antitrust violation claims, Ky. Speedway, 588 F.3d at 918
(citing Whole Foods, 548 F.3d at 1038), they were written to “describe the analytical
process that the [Department of Justice and Federal Trade Commission] will employ in
determining whether to challenge a horizontal merger.” Merger Guidelines § 0.2.
2
The process for excluding Professor Froeb’s testimony involved two orders from
the district court and one from the magistrate judge. This circuitous path warrants
explanation. In August, 2010, the district court initially denied summary judgment to
Defendants on Count I — violation of 15 U.S.C. § 1 by conspiring to restrain trade —
but granted summary judgment on Counts III and IV, which alleged monopolization in
violation of 15 U.S.C. § 2. In re Southeastern Milk Antitrust Litig., 730 F. Supp. 2d 804,
825, 826, 828 (E.D. Tenn. 2010) on reconsideration in part, 2:08-MD-1000, 2012 WL
1032797 (E.D. Tenn. Mar. 27, 2012). Froeb’s evidence played no part in the district
court’s conclusion as to Count I, but it was the decisive factor in ruling against Plaintiffs
on Counts III and IV, both of which require a party to establish a geographic market.
Id. at 825-28.
Froeb had marked out the geographic market for Counts III and IV by looking
at regions where Dean Foods sells, and Food Lion buys, processed milk. He examined
the states of Georgia, North Carolina, and Virginia as possible markets, but individually
each state was too small for the imposition of a profitable price increase because
suppliers would prevent a price increase by shipping cheaper milk into the affected area.
A regional market including Georgia, North Carolina, Virginia, South Carolina, and the
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 18
eastern part of Tennessee, however, was found to be sizeable enough. In making that
determination, Froeb relied on estimates of transportation costs and elasticity of demand.
In reaching its conclusion, the district court did not evaluate Froeb’s methods
under Rule 702 and Daubert. Instead, while assuming that Froeb’s testimony was
reliable and relevant, the court identified four ways that Froeb’s methodology was not
in compliance with the Supreme Court’s requirements for discerning geographic market;
because of this, the court held that Plaintiffs failed to establish a genuine issue of
material fact on this required element. In re Southeastern Milk Antitrust Litig., 730 F.
Supp. 2d at 825. In support of this conclusion, the district court first cited two Supreme
Court cases for the proposition that raising a genuine issue of material fact cannot be
done solely through expert testimony unsupported by facts in the record. Id. at 28 (citing
Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 242 (1993)
and Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 581 n.5 (1986)).
Second, the court cited testimony from Froeb’s deposition, finding that he clearly
testified that the proposed geographic market was founded on something other than the
Tampa Electric standard — that is, the market area “in which the seller operates, and to
which the purchaser can practicably turn for supplies.”4 Id. (citing Tampa Electric,
365 U.S. at 327). Third, Froeb did not consider “commercial realities.” Id. Finally, the
court claimed that Froeb mistakenly premised the geographic market on only one
customer — Food Lion. Id. (citing Apani Southwest, Inc. v. Coca-Cola Enters., Inc.,
300 F.3d 620, 632-33 (5th Cir. 2002)).
Four months later, the magistrate judge assigned to this case issued an order
excluding Froeb’s geographic market testimony pursuant to Rule 702. The order
explained that Defendants had to prevail on their motion to exclude Froeb’s testimony
because the summary judgment decision previously issued by the district court
established the law of the case. This was most clearly true for Froeb’s testimony
pertaining to the monopoly claims, but the magistrate judge concluded likewise for the
4
It is unclear whether Froeb’s statement applies to the geographic market created for the
monopoly claims in Counts 3 and 4, the conspiracy claim in Count 2, or both.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 19
conspiracy claim, using different deposition testimony from Froeb to support this
conclusion. Froeb had testified that the same analytic framework was used to demarcate
the geographic markets for both claims, and the magistrate judge reasoned that if the
framework was unreliable for the monopoly claims, it must also be so for the conspiracy
claim.
This lengthy journey to exclusion finally culminated with the district court’s
order on Plaintiffs’ motion to reconsider the magistrate judge’s decision, in which the
court affirmed the magistrate judge’s decision to exclude Froeb’s testimony. The district
court was skeptical that it needed to rule on the same issue again, but it decided the
cautious approach was most prudent and considered Defendants’ Daubert motion anew,
but as an alternative holding to the prior order. Nevertheless, Froeb’s opinions were
excluded again on similar grounds as before. Once again, the district court found that
Froeb’s deposition testimony did not satisfy the Tampa Electric standard because he had
used the hypothetical monopolist construct improperly. The court reemphasized the lack
of consideration of commercial realities — e.g., information about Plaintiffs’ purchasing
behavior or pricing, how retailers in the prescribed markets currently obtain supplies, or
actual elasticity of demand — and the lack of reliance on facts in the record. Froeb’s
model, based only on Food Lion’s locations, was again disparaged according to the
principle that “a geographic market cannot ordinarily be defined by reference to a single
customer.” Id. (citing Apani Southwest Inc., 300 F.3d at 632-33).
3
The district court’s reasoning in its decision to exclude Froeb’s testimony rests
on an incomplete review of the facts and the application of incorrect legal standards. See
Romberio, 385 F. App’x at 428. First, the blanket exclusion of Froeb’s testimony was
not warranted by his alleged use of one customer when defining a market. Froeb defined
his markets differently for the monopoly claims and the conspiracy claim. For the
monopoly claims, Froeb clearly stated, “[m]y analysis is motivated by including regions
where Dean and Food Lion engage in the sale and purchase of milk.” [R. 1159-5 at 39.]
In contrast, for the conspiracy (or coordinated action) claim, “the candidate area should
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 20
include plants owned by all the alleged conspirators.” Id. at 33. Froeb proceeded to
draw a map encompassing the processing locations owned by Dean Foods, NDH, and
DFA while making assumptions about the locations of “customers.” The result was a
more expansive footprint than that which Food Lion occupies.5 This error began with
the magistrate judge’s order, which quoted Froeb’s deposition. The magistrate judge
focused on Froeb’s explanation that he used “the same analytic framework”; but the
judge did not refer to the surrounding context, in which Froeb explains that he is
struggling to answer the questions posed to him, in part because he is unsure of what the
questioner was asking. Clearly, the methods for creating geographic markets for the
monopoly claims and the conspiracy claim were built on different foundations.
Consequently, the exclusion of one does not necessitate the exclusion of the other.
Second, the requirement that an expert base his findings on facts in the record is
a proper legal proposition, but it was misapplied. In the district court’s order on
summary judgment, it emphasized that “[t]here is nothing in this record to illustrate that
Professor Froeb has based his opinions on evidence in the record; in fact, he appears to
admit that he did not do so, relying instead on a theoretical model he constructed for the
purpose of the analysis.” In re Southeastern Milk Antitrust Litig., 730 F. Supp. 2d at
825. In support of that criticism, the district court cited two Supreme Court decisions.
Id. In Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574 (1986), the
Court affirmed the lower court’s decision to exclude an expert report that utilized
assumptions about a competitor’s costs because the study used assumptions and
estimates as inputs that were “implausible and inconsistent with record evidence.” Id.
at 594 & n.19. In Brooke Group Ltd., the Supreme Court reiterated its prior message:
an expert’s opinion must use valid facts to be reliable. 509 U.S. at 242. Facts can be
undependable for numerous reasons, including actual information that is of poor quality,
5
Orders 5 and 7 cover all or part of fourteen states: Alabama, Arkansas, Florida, Georgia, Indiana,
Kentucky, Louisiana, Mississippi, Missouri, North Carolina, South Carolina, Tennessee, Virginia, and
West Virginia. Plaintiff Fidel Breto has one store located in Tennessee, and Food Lion has 1,300 stores
in 11 states: Delaware, Florida, Georgia, Kentucky, Maryland, North Carolina, Pennsylvania, South
Carolina, Tennessee, Virginia, and West Virginia. The states in Orders 5 and 7 in which Plaintiffs are
located are as follows: Florida, Georgia, Kentucky, North Carolina, South Carolina, Tennessee, Virginia,
and West Virginia.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 21
and contradictory facts present in the record. Id. (citing J. Truett Payne Co., Inc. v.
Chrysler Motors Corp., 451 U.S. 557, 564-65 (1981)). None of those reasons, however
was relied on by the district court in the case at hand. Rather, the sole reason for
excluding Froeb’s testimony was the absence of facts in the record. Thus, the cited
precedent does not adequately support the district court’s conclusion.
Froeb’s report is bereft of citations to an underlying document or report as he
opines on the relatively elementary economic concepts of competition between
processing plants and the benefits one firm could garner by eliminating competition. [R.
1159-5 at 15.] Following that explanation, Froeb extensively cited facts from
government studies, academic publications, and the record itself as he created a
geographic market. [See, e.g., id. at 19 n.34 (citing data showing demand for milk is
relatively insensitive to price); see also id. at n.35 (citing information in the record
showing the same result);6 id. at 25 n.48 (citing record evidence regarding
transportations costs; id. at 26 n.50 (same); id. at 26-27 n.51 (citing reports and record
evidence as inputs for factoring transportation costs).] In sum, expert reports must be
based on proper facts, but each of those facts does not have to occupy an independent
part of the record for an expert to be able to use them when crafting an opinion.
Third, lack of reliance on evidence in the record was combined with criticism
that “commercial realities” were not considered. Commercial realities should be
contemplated when a geographic market is being created. See Kolonn Indus., 637 F.3d
at 442-443. The district court was troubled by the absence of actual information from
Food Lion, such as Food Lion’s purchasing habits, where it actually sought out supplies,
and data about price elasticity. The hypothetical monopolist test and Tampa Electric
both require data based on actual circumstances, e.g., where a buyer and/or seller is
located. Both inquiries, however, also require estimates, and even discount the value of
data based on actual behaviors. The question about buyers in Tampa Electric, for
instance, focuses on where they can “practicably” turn for supplies — not where they
actually do. See Morganstern v. Wilson, 29 F.3d 1291, 1296-97 (8th Cir. 1994); Bathke
6
Price elasticity is noted as a foundational element in Froeb’s definition of a geographic market.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 22
v. Casey’s General Stores, Inc., 64 F.3d 340, 346 (8th Cir. 1995). Moreover, the
hypothetical monopolist construct requires speculation about a buyer’s likely reaction
to a supplier’s price increase. Merger Guidelines § 1.21. Quite obviously, the estimate
should be informed by actual evidence when possible, id., but actual evidence does not
have to be based on the particular parties’ behaviors. In other words, a nationwide
estimate of demand elasticity that is used to predict the reaction of retailers in the
southeast to a price increase would be a reliable method of calculation. Thus, the
hypothetical monopolist test does not require Froeb to use data from Food Lion in place
of data gleaned from a broader sample.
Furthermore, Froeb did not completely ignore commercial realities. He may
have neglected to include important facts; and those identified by the district court may
have more closely aligned his analysis with that explained in the Merger Guidelines. See
Merger Guidelines § 1.21 (“In considering the likely reaction of buyers to a price
increase, the Agency will take into account all relevant evidence, including, but not
limited to, the following: evidence that buyers have shifted or have considered shifting
purchases between different geographic locations in response to relative changes in price
or other competitive variables . . . .”). But actual inputs were considered, most notably
transportation costs and plant locations inside and outside of the proposed geographic
market. Including some facts while omitting others goes to the “accuracy of the
conclusions, not to the reliability of the testimony.” In re Scrap Metal Antitrust
Litigation, 527 F.3d 517, 530-31 (6th Cir. 2008) (quoting Jahn v. Equine Servs., PSC,
233 F.3d 382, 390-93 (6th Cir. 2000)).
Finally, the district court found that Froeb did not apply the Tampa Electric
standard when forming the geographic market. Quoting Froeb’s deposition testimony,
the district court found that he denied that his market analysis looked at the area “in
which the seller operates and to which the purchaser can practically turn for supplies.”
When a similar question was posed again, Froeb repeated his disclaimer that his model
was based on a different approach. Because Froeb’s version of the hypothetical
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 23
monopolist test was applied unconventionally — or at least purportedly so, based on his
deposition testimony — his opinion was categorized as unreliable.
At its most basic, the hypothetical monopolist construct requires selection of the
smallest area in which a SSNIP could be successfully imposed. For that construct to be
valuable in a case, the area at issue must encompass at least some of the locations of the
seller (Defendants) and the buyer (Plaintiffs), including where the buyer could turn for
supplies if prices increased. See Merger Guidelines §§ 1.21, 1.22. The availability of
suppliers that are actually alternatives is limited by the economic realities of the industry
at issue. See id. at § 1.21. Applied in that way, the hypothetical monopolist and the
Tampa Electric standard are practically equivalent: the hypothetical monopolist is “a
useful framework for organizing the factors the courts have applied in geographic market
definition.” 2 Earl W. Kinter et al., Federal Antitrust Law § 10.15 (2013).
Notwithstanding Froeb’s disclaimer, he also states in his deposition that he
“started with areas 5 and 7 because that seemed to include all three of the defendants.”
[R. 1159-5 at 27.] Fidel Breto’s one location was within that area, as were some of Food
Lion’s stores.7 Practicable alternative suppliers were also considered, 8 extending as far
away as 300 miles from Milk Orders 5 and 7 and including all “regulated, non-captive”
plants. Opposing counsel asked if a smaller market were contemplated, and Froeb
replied that the market he described in his report was the only one considered. In their
7
With regard to Plaintiffs’ locations, Froeb graphs Orders 5 and 7, including many locales lacking
any stores owned by Plaintiffs, and makes the following assumption:
Customers locate stores near population centers and locate distribution centers to
minimize the cost of distribution to stores. In the model, customer locations are
represented by a grid of 119 evenly-spaced customer locations within Order Nos. 5 and
7. The total measure of demand at each location is proportional to the census population
nearest that point.
[Id. at 32 n.60.] Froeb’s reason for doing this rather than mapping Plaintiffs’ locations is unclear, but his
more extensive (and, albeit, hypothetical) rendering may be reliable under the principle that the greater
includes the lesser. Froeb’s model includes the actual locations of Defendants’ (sellers) processing plants
and, arguably at least, the practicable alternatives for Plaintiffs (buyers). His inclusion of additional
(perhaps superfluous) buyers does not undermine his conclusion that prices could be manipulated. It
should not be inferred that we are opining on the correctness of Froeb’s conclusions, we merely note that
his method harmonizes with the Tampa Electric standard.
8
“The model specifically considers (1) the locations of plants and customers, (2) the elasticity of
demand for milk, determined by parameters that measure both demand and cost characteristics of the milk
industry, and (3) the costs of transporting bottled milk.” [R. 1159-5 at 31.]
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 24
response on appeal, Defendants criticize Froeb’s market delineation and suggest that the
market is “much smaller.” Multiple courts of appeal have held that market definition is
a question of fact. Kolon Indus., 637 F.3d at 442 (citing cases from the First, Second,
Third, Fifth, Ninth, and Tenth Circuits9). Accordingly, that question is better left for a
jury to decide.10
In conclusion, Froeb’s testimony should not have been excluded on the grounds
relied upon by the district court.
D
As explained above, regardless of whether the court uses the rule of reason or the
per se rule, antitrust plaintiffs must still prove that the restraint at issue caused them to
suffer an antitrust injury. Expert Masonry, 440 F.3d at 342, 343. In its second summary
judgment opinion, the district court found that Plaintiffs had failed to “allege an injury
of the kind which the antitrust laws are designed to prevent.” In re Southeastern Milk
Antitrust Litig., 2012 WL 1032797, at *6. The district court’s justification for this
conclusion was that Plaintiffs’ injury expert, Professor Ronald Cotterill, had conducted
an econometric analysis which partly attributed an increase in the price of milk to the
merger itself rather than to any anticompetitive conduct. Id. The Defendants had
previously filed a Daubert motion, objecting to Cotterill’s testimony; and the matter was
referred to the magistrate judge, who denied Defendants’ motion and ruled that
Cotterill’s testimony was admissible. The district court never formally ruled on the
9
Coastal Fuels of Puerto Rico, Inc. v. Caribbean Petroleum Corp., 79 F.3d 182, 196 (1st Cir.
1996); Todd v. Exxon Corp., 275 F.3d 191, 199 (2d Cir. 2001); Weiss v. York Hosp., 745 F.2d 786, 825
(3d Cir. 1984); Heatransfer Corp. v. Volkswagenwerk, A.G., 553 F.2d 964, 979 (5th Cir. 1977); Oahu Gas
Serv., Inc. v. Pac. Res. Inc., 838 F.2d 360, 363 (9th Cir. 1988); Westman Comm’n Co. v. Hobart Int’l, Inc.,
796 F.2d 1216, 1220 (10th Cir. 1986); see also Thompson v. Metro. Mutli-List, Inc., 934 F.2d 1566, 1573-
74 (11th Cir. 1991) (citing Graphic Products Distributors, Inc., v. Itek Corp., 717 F.2d 1560 (11th Cir.
1983)).
10
Plaintiffs urge us to hold that the district court erred when it relied on Froeb’s deposition
testimony more heavily than the contents of his expert report. Traveling down that line of argument is
unnecessary. Simply reiterating a long-standing and unremarkable principle is sufficient: the law
establishes burdens of persuasion, and parties must bear those burdens. Clear deposition testimony that
contradicts one’s own expert report may make bearing that burden more difficult, and that challenge may
grow more daunting when the testimony and report are related to a difficult legal issue. Ultimately, we
trust that courts vigorously endeavor to rule properly by reviewing the evidence put before them. The onus
is on the parties to advocate clearly.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 25
objection, but when granting summary judgment to Defendants on the injury element,
the wording of the opinion led the Plaintiffs to believe that the district court was
agreeing with the Defendants’ objection, contrary to the ruling of the magistrate judge.
Because of that conclusion, the Plaintiffs have argued that the district court erred in not
giving proper deference to the magistrate judge’s findings, and that this Court should
now review the decision concerning the admissibility of expert testimony for abuse of
discretion. See KY Speedway, 588 F.3d at 915 (citing Hardyman v. Norfolk & W. Ry.
Co., 243 F.3d 255, 258 (6th Cir. 2001)).
However, although the district court never explicitly addressed Defendants’
objections to Cotterill’s testimony, the summary judgment opinion strongly suggests that
the district court concurred with the magistrate judge. The court did not exclude
Cotterill’s testimony, but simply concluded that it “does not create a material issue of
fact.” In re Southeastern Milk Antitrust Litig., 2012 WL 1032797, at *6. Federal Rule
of Civil Procedure 56 on Summary Judgment contains similar language, and legions of
cases do likewise when discussing the summary judgment standard. See e.g., In re
Cardizem CD, 332 F.3d at 906. No doubt it would have been clearer for the district
court to explain with particularity why Defendants’ objections were not compelling.
Nevertheless, it is obvious that the district court considered Cotterill’s testimony in light
of the magistrate judge’s opinion and after independent examination of the evidence.
In re Southeastern Milk Antitrust Litig., 2012 WL 1032797, at *5-6. 11 Consequently,
we review the district court’s grant of summary judgment concerning the issue of
antitrust injury de novo. In re Cardizem CD, 332 F.3d at 905-06 (citation omitted).
Antitrust plaintiffs cannot survive motions for summary judgment without
adequately alleging an antitrust injury. Expert Masonry, Inc., 440 F.3d at 345. In
addition to having to show injury-in-fact and proximate cause, antitrust plaintiffs must
11
In the context of reviewing a magistrate judge’s decision as to a dispositive motion and after
a party’s objection, three other courts of appeal have held that a presumption should exist that a district
court properly reviewed the motion. United States v. Hamell, 931 F.2d 466, 468 (8th Cir. 1991)
(explaining that unless contrary evidence is presented, the appellate court should assume a district court
engaged in appropriate review); Brunig v. Clark, 560 F.3d 292, 295 (5th Cir. 2009) (“[W]e will not assume
that the district court did not conduct the proper review.”); Garcia v. City of Albuquerque, 232 F.3d 760,
766 (10th Cir. 2000) (“[N]either 28 U.S.C. § 636(b)(1) nor Fed. R. Civ. P. 72(b) requires the district court
to make any specific findings; the district court must merely conduct a de novo review of the record.”).
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 26
specifically establish “antitrust injury.” In re Cardizem CD, 332 F.3d at 909. It is not
enough to simply allege that an individual competitor suffered adverse effects from the
defendants’ contract or conspiracy. Care Heating & Cooling, Inc., 427 F.3d at 1014-15.
Rather, “[a]ntitrust injury is (1) injury of the type the antitrust laws were intended to
prevent and (2) injury that flows from that which makes defendants’ acts unlawful.” In
re Cardizem CD, 332 F.3d at 909 (quoting Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc.,
429 U.S. 477, 489 (1977)) (internal quotation marks omitted). The antitrust injury
requirement is critical because it ensures that “the injury should reflect the
anticompetitive effect” of the defendant’s actions. Brunswick Corp., 429 U.S. at 489.
This “ensures that a plaintiff can recover only if the loss stems from a competition-
reducing aspect or effect of the defendant’s behavior.” In re Cardizem CD, 332 F.3d at
909-10 (quoting Atlantic Richfield Co. v. USA Petroleum, 495 U.S. 328, 342-43 (1990)).
In the case at hand, the district court concluded that Plaintiffs had not created a
genuine issue of material fact as to either aspect of antitrust injury. In re Southeastern
Milk Antitrust Litig., 2012 WL 1032797, at *6. The court reasoned that Cotterill’s
multiple regression analysis was too simplistic. Id. Instead of measuring the injury
Defendants’ conspiracy inflicted, Cotterill merely discerned that after controlling for
natural cost increases, prices rose an additional 7.9% between 2002 and 2007. This time
period coincides with the timing of the merger, and the court accordingly concluded that
Cotterill’s calculations only revealed the impact of the merger, which was not contested.
Id.
In reaching that conclusion, the court relied on two facts. First, in Cotterill’s
deposition testimony he stated that the purpose of his calculation was “to analyze
whether in fact the creation of NDH and the assertion that there would be economies of
size and lower prices through efficiencies generated by that creation from January 1,
2002 going forward, whether that in fact was true or not.” Id., at *6 (quoting Cotterill
Depo. April 12, 2010, at 17). Second, during the Department of Justice’s review of the
merger, it created a model to estimate the potential merger’s price-effect. Id. In
Cotterill’s expert report, he explained that his model was similarly designed to that of
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 27
the Department of Justice, and consequently, the court concluded that Cotterill’s
regression analysis must have measured the effect of the merger as well. Id.
That conclusion, however, was based on flawed propositions, and summary
judgment was not warranted on the issue of injury. Although Cotterill made the
statement quoted above, he added — in the same sentence as the testimony was
transcribed — that he was also charged with discerning “whether in fact there is a
reliable economic model of collusion rather than independent self-interest that says, yes,
they did engage in actions that in fact are consistent only with collusive decisions by
Dean, DFA, and NDH, and [those] decisions resulted in elevated prices to the plaintiffs
in this case.” Cotterill Depo. at 17-18.12 Answering that question would expose the
precise sort of injury and causation that is required, especially when Plaintiffs must
benefit from all reasonable inferences. See Logan v. Denny’s, Inc., 259 F.3d 558, 566
(6th Cir. 2001) (citing Anderson v. Liberty Lobby, 477 U.S. 242, 255 (1986)).
The district court’s concerns regarding Cotterill’s regression analysis also do not
support summary judgment. A multiple regression analysis is useful in quantifying the
relationship between a dependent variable (e.g., the price of milk) and independent
variables (e.g., energy costs and/or demand factors). Wiesfeld v. Sun Chemical Corp.,
84 F. App’x 257, 261 n.3 (3rd Cir. 2004). This type of model can also “control for other
independent variables so as to isolate and identify the effect of a single independent
variable on the dependent variable.” Wiesfeld, 84 F. App’x at 261 n.3. The Department
12
The entire quote reads as follows:
As I understand it, I have been asked to analyze whether in fact the creation of
NDH and the assertion that there would be economies of size and lower prices through
efficiencies generated by that creation from January 1, 2002 going forward, whether that
in fact was true or not, and whether in fact there is a reliable economic model of
collusion rather than independent self-interest that says, yes, they did engage in actions
that in fact are consistent only with collusive decisions by Dean, DFA, and NDH, and
that decisions resulted in elevated prices to the plaintiffs in this case.
Now, that was a hypothesis, sir.
First of all, there is a story that can be told that supports the defendants—or the
plaintiffs. There is a story. There is a hypothesis that supports them.
There is a counter-hypothesis, that is what Dean and Suiza represented to
Justice, which in fact there [are] economies of size, there [are] efficiencies, we are going
to pass those on and the plaintiffs in this case are going to enjoy the lower prices. That
is what we looked at.
Cotterill Depo. at 17-18.
No. 12-5457 Food Lion, et al. v. Dean Foods Co., et al. Page 28
of Justice used a regression model to predict that post-merger prices would rise by 2.5%.
Cotterill used a similar model and found that post-merger prices actually increased by
7.9%. Use of that same widely-accepted model does not necessarily mean that the
increase was due to legal causes.
Cotterill’s model, as applied to the facts, reveals three conclusions which, taken
together, can be viewed as evidence of antitrust injury. First, it is clear that Plaintiffs
purchased processed milk from the Defendants. Second, Cotterill’s model indicates that
after the merger Plaintiffs were charged 7.9% more for milk than an econometric
analysis could justify. And third, the district court found that evidence indicated that
Dean Foods and NDH, due to the influence of DFA, conspired to avoid competing
vigorously. In re Southeastern Milk Antitrust Litig., 730 F. Supp. 2d at 815-16 (holding
that there was enough evidence of a conspiracy to deny summary judgment); In re
Southeastern Milk Antitrust Litig., 2012 WL 1032797, at *6 (“The Court has previously
held and now reaffirms that the evidence as a whole creates genuine issues of material
fact as to whether Defendants entered into the alleged agreement.”). This is precisely
the kind of injury that the Sherman Act was designed to prevent. In re Cardizem CD,
332 F.3d at 910-11 (quoting Associated Gen. Contractors v. Cal. State Council, 459 U.S.
519, 538 (1983)) (“Preventing that kind of injury was undoubtedly a raison d’etre of the
Sherman Act when it was enacted in 1890.”).
This conclusion also resolves the question of whether Plaintiffs’ injuries “flow
from that which makes defendants’ acts unlawful.” In re Cardizem CD, 332 F.3d at 909.
The In re Cardizem CD court explained that when competition is limited pursuant to an
agreement and customers are punished through higher prices, the injury clearly results
from anticompetitive conduct. 332 F.3d at 909. Accordingly, summary judgment was
not warranted based on the lack of antitrust injury.
III
For the aforementioned reasons, the district court’s opinion is reversed, and this
case is remanded for further proceedings consistent with this opinion.