Standard Iron Works v. Arcelormittal et al

Filing 577

MEMORANDUM Opinion and Order: Plaintiffs' Motion for Class Certification is granted in part and denied in part. See Order for further details. Signed by the Honorable James B. Zagel on 9/9/2015. Mailed notice(ep, )

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF ILLINOIS
EASTERN DIVISION
In Re: Steel Antitrust Litigation,
THIS DOCUMENT RELATES TO ALL
DIRECT PURCHASER ACTIONS:
Standard Iron Works v. ArcelorMittal et al.,
Case No. 08-cv-5214
No. 08 C 5214
Judge James B. Zagel

Wilmington Steel Processing Co., Inc. v.
ArcelorMittal, et al., Case No. 08-cv-5371
Capow, Inc. d/b/a Eastern States Steel v.
ArcelorMittal, et al., Case No. 08-cv-5633
Alco Industries, Inc. v. ArcelorMittal, et al.,
Case No. 08-cv-6197
Gulf Stream Builders Supply, Inc. v.
ArcelorMittal, et al., Case No. 10-cv-4236

MEMORANDUM OPINION AND ORDER
Plaintiffs Standard Iron Works, Alco Industries, Inc., Gulf Stream Builders Supply, Inc.,
Wilmington Steel Processing Co., and Capow, Inc. d/b/a Eastern States Steel (collectively
“Plaintiffs”), on behalf of itself and all others who purchased steel products directly from
domestic steel producers, Defendants ArcelorMittal S.A. and ArcelorMittal USA LLC, United
States Steel Corporation, Nucor Corporation, Steel Dynamics, Inc., and SSAB Swedish Steel
Corporation (collectively “Defendants”), from April 2005 through December 2007, moves to
certify a class action against Defendants for an alleged multi-year antitrust conspiracy to reduce
the production of steel products in the United States in violation of Section 1 of the Sherman Act,
15 U.S.C. § 1. 1 In particular, Plaintiffs allege that Defendants’ top executives, which collectively

1

This action is instituted under §§ 4 and 16 of the Clayton Act, 15 U.S.C. §§ 15 and 26 to recover treble damages,
amongst other fees and costs, for injuries sustained by Plaintiffs and members of the putative class for Defendants’

1

operated 14 steel mills and received sales of $100 billion during the relevant period, conspired to
restrict the output of their steelmaking furnaces (i.e., “raw steel” production, according to
Plaintiffs) to raise the price of finished steel goods purchased by the class.
For the following reasons, I grant, in part, and deny, in part, Plaintiffs’ motion for class
certification. 2
I.

Motion for Class Certification

Direct Purchaser Plaintiffs moved for class certification on May 24, 2012. Defendants
filed an Opposition to Plaintiffs’ Motion for Class Certification on February 28, 2013, which
Plaintiffs replied to on October 15, 2013. An evidentiary hearing then took place on March 5,
March 6, and April 11, 2014, at which the parties presented evidence and argument relating to
both class certification and Defendants’ Daubert motions to exclude Plaintiffs’ experts.
Plaintiffs offered the testimony of Dr. John L. Solow, Professor of Economics at the
University of Iowa, an expert on “economics, industrial organization and econometrics,” Dr.
Jeryl K. Wright, an expert in “steel making, steel processing, steel plant processed layouts, steel
customer service and quality system management,” and Dr. James T. McClave, an expert in
“statistics and econometrics.”
Defendants offered the testimony of their experts, Dr. Jerry A. Hausman, the MacDonald
Professor of Economics at the Massachusetts Institute of Technology, as an expert in “economics
and econometrics,” and two steel industry executives, Mr. Glenn A. Pushis, Vice President and
General Manager of SDI’s Flat Roll Division, and Ms. Lisa Roundabush, U.S. Steel’s Managing
Director, Product Quality North American Flat Rolled Operations.

alleged violations of § 1 of the Sherman Act, 15 U.S.C. § 1.
2
As in all cases, as this case proceeds, further information may come to light or a change in circumstances may
occur which requires modification of the methodology employed to effectively and efficiently resolve the pending
issues related to this matter.

2

By agreement of the parties, Defendants offered the testimony of Dr. Brian G. Thomas,
C.J. Gauthier Professor of Mechanical Engineering and Director of the Continuous Casting
Consortium at the University of Illinois at Urbana-Champaign, which was submitted through an
affidavit and a Revised Appendix to his report, as an expert in metallurgy, steel processing, and
steel manufacturing. Defendants also submitted the expert report of Ms. Jennifer A. Hillman, a
former Commissioner at the United States International Trade Commission, on international
trade law and international trade remedies.
Based on a rigorous review of the complete record, the Court denied Defendants’ motion
to exclude the testimony of Plaintiffs’ experts after finding that the expert opinions of Drs.
Solow, Wright, and McClave were based on reliable methodology generally accepted in the
relevant scientific community and were relevant to the task at hand for the purposes of Daubert.
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 580, 113 S. Ct. 2786, 2790, 125 L. Ed. 2d
469 (1993). That the opinions of the experts in this case are, in part or whole, valid as well as
reliable is yet to be decided.
II.
A.

Legal Standard

Class Certification Requirements

I have broad discretion in determining whether to certify a class, see Donovan v. Estate of
Fitzsimmons, 778 F.2d 298, 306 (7th Cir.1985), but I may not reach the merits of the case when
considering certification. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 177–178, 94 S.Ct.
2140, 2152–53, 40 L.Ed.2d 732 (1974). I am to interpret Rule 23 liberally to permit class actions.
See In re Folding Carton Antitrust Litigation, 75 F.R.D. 727, 731 (N.D.Ill.1977). Calkins v. Fid.
Bond & Mortgage Co., No. 94C5971, 1998 WL 719569, at *1 (N.D. Ill. Oct. 8, 1998). In moving
for class certification, Plaintiffs bear the burden of establishing that all the requirements of Rule
23(a)—numerosity, commonality, typicality, and adequacy of representation—and, additionally,
3

one of the requirements of Rule 23(b) are met. General Telephone of the Southwest Co. v.
Falcon, 457 U.S. 147, 161, 102 S.Ct. 2364, 2372–73, 72 L.Ed.2d 740 (1982); Lilly v. Ford
Motor Co., No. 00 C 7372, 2002 WL 507126, at *1 (N.D. Ill. Apr. 3, 2002). Under Rule
23(b)(3), a class should be certified where (1) common questions predominate over individual
questions, and (2) a class action is superior to other adjudicative methods. Miller v. Gen. Motors
Corp., No. 98 C 2851, 2003 WL 168626, at *1 (N.D. Ill. Jan. 26, 2003).
It “may be necessary for the court to probe behind the pleadings before coming to rest on
the certification question,” and that certification is proper only if “the trial court is satisfied after
a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied.” Wal-Mart Stores,
Inc. v. Dukes, 131 S. Ct. 2541, 180 L. Ed. 2d 374 (2011) (quoting General Telephone Co. of
Southwest v. Falcon, 457 U.S. 147, 160-161 (1982).
III.
A.

Discussion

Rule 23(a) Requirements
1.

Numerosity

To satisfy the numerosity requirement, “the class must be so numerous that joinder of all
members is impracticable.” Fed. R. Civ. P. 23(a)(1). Plaintiffs’ class is comprised of thousands
of class members. Plaintiffs allege that there are 5,585 direct purchasers, with more than half of
the class having fewer than 50 transactions during the relevant class period. Schmidt v. Smith &
Wollensky LLC, 268 F.R.D. 323, 326 (N.D.Ill. 2010) (“a class of more than 40 members is
generally believed to be sufficiently numerous”); Riordan v. Smith Barney, 124 F.R.D. 60, 62
(N.D.Ill. 1989) (citing cases certifying classes with 10-23 members); Swanson v. American
Consumer Industries, 415 F.2d 1326, 1333 (7th Cir. 1969) (finding forty class members sufficient

4

to certify a class). The Court is satisfied that numerosity, which is uncontested, is satisfied and
joinder is impracticable.
2.

Commonality

Commonality is satisfied if “there are questions of law or fact common to the class.” Fed.
R. Civ. P. 23(a)(2). A common nucleus of operative facts exists between the named plaintiffs and
the class, such that commonality is met, when a defendant has “engaged in standardized conduct
toward members of the proposed class.” Keele v. Wexler, 149 F.3d 589, 594 (7th Cir. 1998);
Saltzman v. Pella Corp., 257 F.R.D. 471, 475 (N.D.Ill. 2009) aff’d 606 F.3d 391 (7th Cir. 2010).
While factual variations among the class grievances do not defeat a class action, Patterson v.
General Motors Corp., 631 F.2d 476, 481 (7th Cir. 1980), Plaintiffs must identify an issue of fact
or law whose resolution “is central to the validity of each” class member’s claim. Wal-Mart
Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551, 180 L. Ed. 2d 374 (2011). The common issue or fact
“must be of such a nature that it is capable of classwide resolution—which means that
determination of its truth or falsity will resolve an issue that is central to the validity of each one
of the claims in one stroke.” Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551, 180 L. Ed.
2d 374 (2011).
Plaintiffs’ theory is that Defendants, steel producers, engaged in a conspiracy to
manipulate the supply and/or price of steel products sold in the United States. Plaintiffs will offer
evidence common to the class to establish the identity of the participants, the duration of the
conspiracy, and the acts carried out in furtherance of the conspiracy. The central issues before
the court will be whether the alleged conspiracy violated Section 1 of the Sherman Act; whether
the conduct of Defendants impacted the supply and/or prices of steel products sold in the United
States, and whether any related price changes caused injury to Plaintiffs and other class

5

members. The Court will also be required to measure the extent of class-wide damages, if any,
and assess whether Defendants have any defenses to potential liability. Resolution of these issues
is essential to every class members’ claims. Commonality is satisfied.
Defendants’ argument that the putative class is comprised of thousands of steel
purchasers, varying in size and product type purchased, and that impact and damages are not
common questions to the entire class, but require individualized proof by each class member, are
issues for a predominance inquiry. Commonality is satisfied.
3.

Typicality

Under Rule 23(a), typicality is satisfied if “the claims or defenses of the representative
parties are typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a). The typicality
requirement “directs the district court to focus on whether the named representatives’ claims
have the same essential characteristics as the claims of the class at large.” Retired Chi. Police
Ass’n City of Chicago, 7F.3d 584, 597 (7th Cir. 1993). In the antitrust context, a “plaintiff’s claim
is typical if it arises from the same event or practice or course of conduct that gives rise to the
claims of other class members,” De La Fuente v. Stokely-Van Camp, Inc., 713 F.2d 225, 232 (7th
Cir. 1983), and “plaintiffs and all class members alleg[e] the same antitrust violations by
defendants.” In re Rubber Chemicals Antitrust Litig., 232 F.R.D. 346, 351 (N.D. Cal. 2005)
; see also In re Urethane Antitrust Litig., 251 F.R.D. 629, 640 (D. Kan. 2008) aff'd, 768 F.3d
1245 (10th Cir. 2014)(“[t]ypicality refers to the nature of the claims of the representative, not the
individual characteristics of the plaintiff”). While class claims must be “fairly encompassed” by
the claims of the named plaintiffs, the typicality requirement does not require that all class
members suffer the same injury as the named representative. General Tel. Co. of the Northwest,
Inc. v. Equal Employment Opportunity Commission, 446 U.S. 318, 330 (1980); Id (at Fuente).

6

“The issue of typicality is closely related to commonality and should be liberally construed.”
Saltzman, 257 F.R.D. at 480 (citing Keele, 149 F.3d at 595).
Here, the named plaintiffs, Standard Iron Works, Wilmington Steel Processing, Co. 3,
Alco Industries, Inc., Capow, Inc. d/b/a Eastern States Steel, and Gulf Stream Builders Supply,
Inc., like all other members of the proposed class, bought steel products directly from
Defendants during the alleged conspiracy period and subsequently brought claims under the
Sherman Act alleging that Defendants conspired to restrict steel production and raise the price of
those steel products during the relevant period. While Defendants contend that the named
representatives’ claims depend on individualized facts and distinct evidence that would be
properly presented at individual trials, the claims of each named representative and class
members are based on the same legal theory and arise from the same course of conduct (i.e.
conspiring to restrict steel supply); therefore, typicality is satisfied under Rule 23(a).
4.

Adequacy

Rule 23(a)(4) requires that the “representative parties will fairly and adequately protect
the interests of the class.” Fed. R. Civ. P. 23(a)(4). The adequacy requirement is satisfied where
the named representative (1) has retained competent counsel, (2) has a sufficient interest in the
outcome of the case to ensure vigorous advocacy, and (3) does not have interests antagonistic to
those of the class. Saltzman v. Pella Corp., 257 F.R.D. 471, 480 (N.D. Ill. 2009) aff'd, 606 F.3d
391 (7th Cir. 2010). The adequacy inquiry under Rule 23(a)(4) serves to uncover conflicts of
interest between named parties and the class they seek to represent. Id. (citing Amchem Products,

3

Parties dispute whether Wilmington Steel Processing, Co. retained possession of its antitrust claims after it sold
“substantially all of its assets” to first, Concord Steel in September 2006 and then, new business assets to VR Laser
Services USA, Inc. in May 2008. Because Wilmington Steel did not expressly transfer its antitrust claims, it retains
its antitrust claims. See Franco v. Conn. Gen. Life Ins. Co., 818 F. Supp. 2d 792, 812 (D.N.J. 2011); Gulfstream III
Associates, Inc. v. Gulfstream Aerospace Corp., 995 F.2d 425, 440 (3d Cir. 1993) (“any assignment of antitrust
claims, as a matter of federal common law, must be an express assignment”).

7

Inc. v. Windsor, 521 U.S. 591, 625, 117 S.Ct. 2231, 138 L.Ed.2d 689 (U.S.1997)). The burden is
on the party opposing class certification to demonstrate that representation will be inadequate.
Robin v. Doctors Officecenters Corp., 686 F.Supp. 199, 203 (N.D.Ill. 1988)(citing Lewis v.
Curtis, 671 F.2d 779, 788 (3d. Cir.), cert denied, 459 U.S. 880, 103 S.Ct. 176, 74 L.Ed.2d 144
(1982).
Defendants argue that Plaintiffs’ allegations that steel service centers, such as “Metals
Service Center Institute,” were among the forums in which Defendants made relevant
conspiratorial statements representing a conflict with a significant component of the putative
class. Defendants contend that Plaintiffs cannot represent the class members that they accused of
conspiring with Defendants. Plaintiffs do not, however, allege that either the MSCI trade
association or any class member that controls a steel service center is an alleged co-conspirator
in this case. To the extent Defendants argue that certain service centers may have benefitted from
production cuts, they are nevertheless entitled to recover 100% of the overcharge imposed by the
conspiracy, just as all other direct purchasers in the proposed class. Illinois Brick Co. v. Illinois,
431 U.S. 720, 746, 97 S. Ct. 2061, 2075, 52 L. Ed. 2d 707 (1977).
There is no conflict between the named Plaintiffs’ and class members’ interests; rather,
they are aligned in establishing Defendants’ liability under the antitrust laws and maximizing
class-wide damages. Additionally, the named Plaintiffs and putative class are represented by
experienced counsel who have litigated this matter vigorously and competently. Thus adequacy
is satisfied under Rule 23(a)(4).
B.

Rule 23(b) Requirements
To satisfy Rule 23(b) and obtain class certification on grounds that “questions of law or

fact common to class members predominate over any questions affecting only individual

8

members, and that a class action is superior to other methods for fairly and efficiently
adjudicating the controversy,” Plaintiffs must show that common proof will predominate with
respect to each of the elements of their claims. Fed. R. Civ. P. 23(b)(3). The following four
factors that are pertinent to finding predominance and superiority: (1) the interest of members of
the class in individually controlling the prosecution or defense of separate actions; (2) the extent
and nature of any litigation concerning the controversy already commenced by or against
members of the class; (3) the desirability or undesirability of concentrating the litigation of the
claims in the particular forum; and (4) the difficulties likely to be encountered in the
management of a class action. Fed.R.Civ.P. 23(b)(3); Lilly v. Ford Motor Co., No. 00 C 7372,
2002 WL 507126, at *1 (N.D. Ill. Apr. 3, 2002).
1.

Predominance

The predominance criterion, requiring that common questions of law in this case
predominate over any question affecting only individual members, is similar to Rule 23(a)’s
commonality requirements, but even more demanding. Comcast Corp. v. Behrend, 133 S. Ct.
1426, 1432, 185 L. Ed. 2d 515 (2013) . When “common questions represent a significant aspect
of [a] case and…[they] can be resolved for all members of [a] class in a single adjudication,”
there is clear justification for handling the dispute on a representative rather than on an
individual basis. Messner v. Northshore University HealthSystem, 669 F.3d 802, 815 (7th
Cir.2012)(quoting 7AA Wright & Miller, Federal Practice & Procedure § 1778 (3d ed.2011));
Joseph v. Gen. Motors Corp., 109 F.R.D. 635, 641 (D. Colo. 1986). Determining what issues
predominate requires a qualitative assessment, rather than a simply quantitative determination
made “by counting noses,” that is to say, identifying whether there are more common issues or
more individual issues, regardless of relative importance. Parko, 1086. So, every issue in the

9

case need not be common; the question, rather, is whether substantial common issues
predominate. See Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 815 (7th Cir. 2012);
In re NASDAQ Mkt.-Makers Antitrust Litig., 169 F.R.D. 493, 517 (S.D.N.Y. 1996)
(predominance is satisfied unless “it is clear that individual issues will overwhelm the common
questions”).
Analysis of predominance under Rule 23(b) “begins, of course, with the elements of the
underlying cause of action.” Messner, 669 F.3d at 815 (citing Erica P. John Fund, Inc. v.
Halliburton Co., 563 U.S.804, 131 S.Ct. 2179, 2184, 180 L.Ed.2d 24 (2011)). The essential
elements of Plaintiffs’ cause of action arising under § 1 of the Sherman Act, 15 U.S.C. § 1 are
“(1) a violation of antitrust law…; (2) individual injury, or impact, caused by that violation; and
(3) measurable damages.” Reed v. Advocate Health Care, 268 F.R.D. 573, 581 (N.D.Ill. 2009)
(citing In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 311 (3d. Cir. 2008)).
Plaintiffs must affirmatively show how common evidence and a single, reliable
methodology will prove an element on a simultaneous, class-wide basis. Parko v. Shell Oil Co.,
739 F.3d 1083, 1085-86 (7th Cir. 2014); Butler v. Sears, 727 F.3d 796 (7th Cir. 2013). It is now
indisputably the role of the district court to scrutinize the evidence before granting certification,
even when doing so “requires inquiry into the merits of the claim.” In re Rail Freight Fuel
Surcharge Antitrust Litig.-MDL No. 1869, 725 F.3d 244, 253 (D.C. Cir. 2013) 133 S.Ct. at 1433.
After the Court examines “the substantive elements of plaintiffs’ cause of action and inquire[s]
into the proof necessary for the various elements,” it “inquire[s] into the form that trial on these
issues would take.” Simer v. Rios, 661 F.2d 655, 672 (7th Cir. 1981). In many cases alleging a
violation of antitrust laws, predominance is a test “readily met.” Amchem, 521 U.S. at 625.

10

(a)

15 U.S.C. § 1: a violation of antitrust law

Here, it is undisputed that whether Defendants engaged in conspiracy in violation of
antitrust law is a question common and central to the entire class’s claims. Any trial in this
matter will focus overwhelmingly on common proof of conspiracy—witness testimony,
documents, email and phone records, economic evidence, and other evidence relating to
Defendants’ conduct. In particular, Plaintiffs plan to prove causation by introducing the
statements of Defendants’ senior executives regarding its plan to “exercise discipline” and cut
supply over the relevant period, which, as Defendants’ executives subsequently stated at private
trade association meetings, caused higher prices and profits for the industry as a whole.
Plaintiffs also intend to offer the expert testimony of economist Dr. John Solow as
common proof of conspiracy. Dr. Solow, using structure-conduct-performance (“SCP”) analysis,
identified a number of structural features that are recognized as facilitating effective collusion,
including industry concentration, barriers to entry, lack of close substitutes, product
standardization, and trade associations. These factors suggest, in Dr. Solow’s opinion, that even
though not all finished steel is the same, Defendants compete principally on the basis of price in
the sale of the various finished steel goods which is an economic factor that motivates and
facilitates collusion. Defendants argue that Dr. Solow’s SCP analysis and opinions on the
structure of the steel industry merely shows that the market is not inconsistent with coordination
amongst firms. This objection goes to the weight that should be accorded to Dr. Solow’s opinion
and is not a basis for striking his opinion as unreliable. Dr. Solow’s testimony is admissible and
is common evidence that is relevant to establishing the element of conspiracy.
Plaintiffs offer evidence that is common to all class members and would prove or
disprove a violation of the antitrust law on a class-wide basis. The question of whether

11

Defendants engaged in a conspiracy is an issue that is central to this case and has the effect of
satisfying the prerequisite of predominance in Rule 23(b)(3). Whether Defendants’ conduct
constitutes conspiracy in violation of antitrust law is a question appropriate for resolution on a
class-wide basis.
(b)

15 U.S.C. §1: impact of the violation of antitrust law

Still, other issues in antitrust actions besides conspiracy have also been found to
predominate. The core of this dispute—as in many antitrust disputes—is the second element, the
antitrust impact, of the alleged conspiracy; in other words, did the alleged conspiracy cause
higher prices in the marketplace? The antitrust injury requirement allows a plaintiff to recover
only if the plaintiff has suffered a loss that stems from a competition-reducing aspect of the
defendant's behavior. In re Urethane Antitrust Litig., 251 F.R.D. 629, 634 (D. Kan. 2008) aff'd,
768 F.3d 1245 (10th Cir. 2014).
At the class certification stage, Plaintiffs’ burden is not to prove the element of antitrust
impact itself, but only to “demonstrate that the element of antitrust impact is capable of proof at
trial through evidence that is common to the class rather than individual to its members.”
Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1430 (2013) (emphasis added). While impact or
damages must be determined by a common methodology or evidence for all members of the
class, Rule 23(b)(2) does not require uniformity in impact or identical damages. Messner, 669
F.3d at 815 (finding that “common proof of damages for class members…is not required.”);
Butler, 727 F.3d at 819 (stating that “requir[ing] that every member of the class have identical
damages” would “drive a stake through the heart of the class action device.”).
To prove antitrust impact and the accompanying damages, Plaintiffs intend to offer the
expert testimony of Drs. Solow, Jeryl Wright, and James McClave to show that the structure of

12

the steel industry was ripe for a successful supply cartel and that raw steel production cuts caused
class-wide impact in the form of increased prices. To further demonstrate class-wide impact,
Plaintiffs intend to offer pricing data that shows that prices of all major categories of steel
products followed similar trends during the relevant period and resulted in widespread impact, as
well as econometric analysis that serves as proof that class members paid higher prices for steel
products than they would have paid in a competitive market. This evidence, Plaintiffs contend,
will prove that its class members suffered damages on a common basis as a result of Defendants’
conduct.
Defendants oppose class certification based on its position that Plaintiffs cannot prove
antitrust impact (i.e. whether the conspiracy caused an increase in prices) and damages using
common proof at trial. Specifically, Defendants argue that both Dr. Solow’s and Dr. McClave’s
methodologies have significant flaws that render them incapable of determining the impact of the
conspiracy on a class-wide basis and damages. Defendants further argue that, in any case, these
questions, which are critical to this case, require individual proof by each class member. This
Court has previously considered the expert testimonies of Drs. Solow, Wright, and McClave and
determined that they are all admissible as reliable and relevant under Daubert. Nonetheless,
expert testimony should never be “uncritically accepted as establishing a Rule 23 requirement,”
Messner, In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 311 (3d. Cir. 2008), and the
Court considers each of Defendants’ objections to Plaintiffs’ experts in turn.
(i) Common evidence of impact
(a) Supply-side Substitutability
Parties differ in even their basic understanding of how the steel industry operates. First,
Dr. Solow defined the relevant antitrust market from the supply side and explained that a

13

coordinated restriction in the supply of raw steel products would cause class-wide price effects
“under the basic laws of supply and demand.” Based on supply-side economic theory, Dr.
McClave and Dr. Solow opined that “even if two products are completely different from the
consumer’s standpoint, if they are made by the same producers an increase in the price of one
that is not cost-justified will induce producers to shift production from the other product to this
one in order to increase their profits by selling at a supracompetitive price.” Blue Cross & Blue
Shield United of Wisconsin v. Marshfield Clinic, 65 F.3d 1406, 1410-11 (7th Cir. 1995), as
amended on denial of reh'g (Oct. 13, 1995).
It is well-established that the steel industry does not sell a single commodity substance
called “raw steel.” Rather, it sells “a large variety of product types” with different chemical
properties, including “sheets, bars, and other shapes, of varying thickness, rigidity, heat
resistance, strength, ductility, weight and so forth.” Richard A. Posner, Antitrust Law 65 (2d ed.
2001). Still, Plaintiffs argue that the steel for virtually all of the finished products in the proposed
class, which includes a variety of different steel products, comes from common furnaces and can
be made by multiple, overlapping producers.
Dr. Solow explained in his reports and testimony the rule that, when the price of steel
products made in common or substitutable furnace facilities are connected, a cartel reducing raw
steel production would impact the price of all products made in curtailed furnaces and the price
of their supply-side substitutes. For example, Dr. Solow explained: “if B producers can costlessly
switch production to product A in a short time and can readily distribute the resulting output,
they will constrain the prices of A firms in virtually the same way as another A firm.”
This holds true even when steel products are “not close substitutes for each other on the
demand side[;] two products produced interchangeably from the same production facilities are

14

presumptively in the same market.” See generally Areeda, Hovenkamp, & Solow, Antitrust Law,
¶¶ 560-61 (3rd ed. 2007). As further evidence that there is at least some substitutability between
steel products, Plaintiffs point to the explanation of Dr. Hausman, Defendants’ rebuttal expert:
“[A] certain grade of steel has to meet ASTM standards, and it is all going to be the same, and
that is a perfect substitute. It is nondifferentiated. Products like that compete only on price.” Dr.
Solow concludes that because the prices of all steel products within the class definition are
linked in the same economic market on the supply side, common economic proof can be used to
show that prices for finished steel rose significantly and across the board after the production
cuts—across all major products and Defendants.
Defendants, however, dispute the very premise that supply-side substitutability can be
applied to steel products based on the realities of the steel industry and argue that Dr. Solow
should have defined the market for the steel industry from the demand side. First, Defendants
contend that steel prices vary widely because finished steel is largely custom made, with
differences in steel chemistry, product specifications, and end-uses. No defendant or mill makes
all products, and most mills focus on a relatively small number of products; moreover, preparing
a steel mill’s furnace to switch production of products is time-consuming and expensive.
Defendants contend that this demonstrates that, even if there had been a capacity constraint,
furnaces and steel mills would not have been able to switch production to high profit margin
steel products, even if it was profitable.
Markets may be appropriately defined from either the supply side (Plaintiffs’ and Dr.
Solow’s approach) or the demand side (Defendants’ and Professor Hausman’s approach), as long
as it can be used to determine the power of the relevant market. Rebel Oil Co. v. Atl. Richfield
Co., 51 F.3d 1421, 1438 (9th Cir. 1995). Dr. Solow testified that there is direct, undisputed

15

evidence that at least some substitution is possible and occurred. Dr. Jeryl Wright also opined
that industry furnace production is flexible and substitutable. He testified that “reducing primary
production capability” (i.e. cutting furnace production) “impacts every product that comes
through that mill.” While there are undoubtedly barriers to switching production, for our current
purposes, I find that there are also a wide range of products that may have been produced from
common furnaces that make supply substitutability an appropriate way to define the antitrust
market and measure impact.
(b) “But for”Production Levels
Plaintiffs’ theory is that Defendants’ executives conspired to cut overall steel production
in the United States twice within a 33-month period, which caused an increase in the price of
steel products. Even if some steel products are substitutable, Defendants argue that Plaintiffs
have failed to measure “but for” production levels, i.e. what production levels would have been
“but for” the alleged cartel to restrict supply. While Dr. Solow did not conduct statistical analysis
on production levels, he reviewed industry production data and direct evidence that production
was curtailed, interpreted a “Production Cut” chart created by Plaintiffs’ attorneys, testified
about what actual production levels were, and opined on the “but for” production levels.
Defendants argue that the “Production Cut” chart referenced by Dr. Solow had major
flaws, such as omitting major product categories, like structural beam and failing to account for
whole sections of the products encompassed by the proposed class and members. Defendants
also argue that any alleged production cuts would not have equally applied to all products within
Plaintiffs’ proposed definition and, consequently, would not have impacted the price of all
products equally. In response, Dr. Solow opined that the proper comparison is not whether the
sale of each and every steel product decreased over time during the cartel period, but whether

16

overall steel production was lower than it otherwise would have been had the coordinated
reductions in furnace output not taken place. Unconvinced, Defendants concluded that Plaintiffs
had not offered a class-wide method of proving at trial that the production of each proposed class
product decreased below competitive levels during the class period and failed to identify “but
for” production levels to link their liability theory to their damage estimate.
In Comcast, the Court declined to reject the Third Circuit majority’s conclusion that “[a]t
the class certification stage we do not require that [p]laintiffs tie each theory of antitrust impact
to an exact calculation of damages, but instead that they assure us that if they can prove antitrust
impact, the resulting damages are capable of measurement and will not require labyrinthine
individual calculations,” Behrend v. Comcast Corp., 655 F.3d 182, 203-04 (3d Cir. 2011) rev'd,
133 S. Ct. 1426, 185 L. Ed. 2d 515 (2013). Rather, the Court held that the plaintiffs, who had
failed to establish a method for proving common impact, could not rest on such assurances. The
Court further recognized the “inherent difficulty of identifying a ‘but-for world,’” and stated that
it did “not require that damages be measured with certainty, but rather that they be demonstrated
as “a matter of just and reasonable inference.” Behrend, 655 F.3d at 203 (citing Story Parchment
Co. v. Paterson Parchment Paper Co., 282 U.S. 555, 563, 51 S.Ct. 248, 75 L.Ed. 544 (1931).
Though Plaintiffs’ “but for” production data cannot be considered precise, I find it
reasonable that, if Plaintiffs are able to show antitrust impact (an increase in prices) on a class
basis, the resulting damages—which requires production data—are capable of measurement on a
class basis.
(c) Dr. McClave’s Model to Prove Common Impact
If allowed to proceed to trial, Plaintiffs must establish that the injury it suffered (in the
form of increased prices) from the violation of the antitrust laws is measurable. See Hydrogen

17

Peroxide, 552 F.3d at 311; see also Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259
F.3d 154, 188 (3d Cir. 2001) (“Proof of injury (whether or not an injury occurred at all) must be
distinguished from calculation of damages (which determines the actual value of the injury).”)
To proceed as a class, Plaintiffs must offer a common methodology for proving that prices
increased on a class-wide, rather than individual, basis.
Plaintiffs offer Dr. McClave’s model to measure the impact of the alleged conspiracy on
the price of steel products (the dependent variable) by controlling for relevant demand and cost
factors (the explanatory variables). Dr. McClave included in his model one explanatory demand
variable, a weighted average index of demand for construction and automobile manufacturing,
and one explanatory cost variable, the cost of scrap—which he determined to be the single most
important price driver in the industry. McClave’s model used a pre-conspiracy benchmark period
(February 2002 to March 2005), the proposed class period, during which the conspiracy
occurred, (April 2005 to December 2007), and a post-conspiracy benchmark period (January
2008 to December 2009).
Dr. McClave divided steel products into 25 product categories to reduce transaction data
from 30 million to approximately 6,700 transactions and then ran separate regression tests for
two broad categories of steel products, flat products and long products. Using multiple regression
analysis, Dr. McClave calculated a single, average estimated overcharge (i.e. estimated antitrust
impact) that is aggregated across all class members within each of the flat (5.6%) and long (7%)
categories and does not calculate individual damages (i.e. overcharges for any particular
customer). To verify its data in response to Dr. Hausman’s critique, Dr. McClave used a
customer-specific model which added customer identifiers and assessed impact on an individual
basis.

18

Rule 23 requires the court to take a hard look at the soundness of statistical models that
purport to show predominance. In re Rail Freight Fuel Surcharge Antitrust Litig.-MDL No.
1869, 725 F.3d 244, 255 (D.C. Cir. 2013). Defendants believe the Dr. McClave’s model suffers
from fatal flaws that render his results unreliable and unable to prove impact. For instance,
Defendants explain, the true bottleneck for many producers of steel was not the furnace but the
mills, so even a restriction on the supply at the furnace would not have impacted the production
of steel at many existing mills, or the price. Rather, Defendants argue that the dummy variable in
Dr. McClave’s model picked up factors that affect production and price and improperly
attributed the impact of those factors to the alleged conspiracy. Defendants point to an increased
demand on steel from China as a cause of increased prices, as well as an increase in the cost of
zinc.
Additionally, Defendants claim that, as in Rail Freight, Plaintiffs included fixed-price
long term contracts which were negotiated at prices higher than Dr. McClave’s estimated
competitive price. 725 F.3d at 253. Dr. Hausman argued that, “[t]he core thing that’s wrong with
the model is, he doesn’t take into account that this very high proportion of steel is sold by
contract…and you need to take that into account in your model.” Defendants, relying on
Comcast Corp. v. Behrend, argue that because Plaintiffs’ model cannot calculate the impact of
solely the alleged conspiracy without these external factors, it cannot be used to prove common
impact. 133 S.Ct. 1426, 1430 (2013).
A model that “identifies damages that are not the result of the wrong” is an impermissible
basis for calculating class-wide damages. Comcast, 133 S.Ct. at 1434; Butler v. Sears, Roebuck
& Co., 727 F.3d 796, 800 (7th Cir. 2013). In Comcast, Dr. McClave designed a regression model
comparing actual prices with hypothetical prices that would have prevailed but for the

19

petitioner’s allegedly anticompetitive conduct. Petitioners offered four theories of antitrust
impact, three of which were ultimately rejected by the district court. Dr. McClave’s regression
model was designed to calculate impact and damages resulting from “the alleged anticompetitive
conduct as a whole,” not just the theory of liability accepted by the court. The Court
determined—even without objection by Comcast to Dr. Clave’s model, testimony, or expert
report—that Dr. McClave’s model failed to attribute damages to any one particular theory of
anticompetitive impact. This, the Court concluded, resulted in a model that would identify
damages that were not the result of the wrong. With no common methodology for evaluating
class-wide impact, the Court found that questions of individual damage calculations would
inevitably overwhelm questions common to the class and rejected class certification.
Here, unlike Comcast, Dr. McClave’s model is offered to measure the impact of the
conspiracy. Plaintiffs do not have multiple theories of liability, but one distinct theory that the
price of steel products increased directly as a result of Defendants’ conspiracy to restrict supply.
While calculations need not be exact, Story Parchment Co., 282 U.S. 555, 563 (1931), a model
that does not even attempt to identify and isolate “damages” that are not the certain result of the
wrong cannot be used to establish that damages are susceptible of measurement across an entire
class for purposes of Rule 23(b)(3). Comcast, 133 S. Ct. at 1433; Butler, 727 F.3d at 799. Dr.
McClave’s model has narrowly identified the wrong (i.e. the conspiracy to restrict production)
and “translat[ed] [] the legal theory of the harmful event into an analysis of the economic impact
of that event.” Federal Judicial Center, Reference Manual of Scientific Evidence, 432 (3d ed.
2011) (emphasis added).
While it is likely true that some of the external factors raised by Defendants also had an
impact on the cost and price of steel, “the Supreme Court and this Circuit have confirmed on a

20

number of occasions that the selection of the variables to include in a regression analysis is
normally a question that goes to the probative weight of the analysis rather than to its
admissibility. Manpower, Inc. v. Ins. Co. of Pennsylvania, 732 F.3d 796, 808 (7th Cir. 2013)
(citing Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986) (reversing
lower court's exclusion of regression analysis based on its view that the analysis did not include
proper selection of variables)).
In any case, the model itself, Dr. McClave contends, has considered these factors and is
statistically sound with R-squared values, the most common measure of a model’s ability to
explain industry data, of 92-95%. This, he contends, indicates that his model is able to accurately
explain the impact of the conspiracy on steel production and price. FJC Reference Guide, at 355.
Dr. Hausman disputes that McClave’s model is reliable and properly used here because it
failed three tests that assess statistical reliability: the Chow test, the F-test, and the Hausman test.
First, Dr. Hausman contends that McClave’s model failing the Chow test indicates that there
were “structural changes” between periods. A structural change occurs when an increase in cost
leads to an increase in price, but the relationship doesn’t remain consistent over that period of
time; rather than accurately assessing the relationship between cost and price, Dr. McClave’s
model averages and estimates the relationship. Defendants explain that the structural break was
likely caused by Dr. McClave’s improper inclusion of the Great Recession in the post-conspiracy
benchmark. Defendants argue that the structural break showed false positives, which they claim
is proof that Dr. McClave’s results aren’t reliable.
There are different accepted methods to account for the reliability of serial correlations
and models, Fed. R. Evid. 702, Advisory Committee Note (“expert testimony cannot be excluded
simply because the expert uses one test rather than another, when both tests are accepted in the

21

field and both reach reliable results”), and this Court is satisfied that Dr. McClave’s multiple
regression methodology is reliable and admissible.
While multiple regression analysis, by definition, makes use of averages to identify the
underlying trends in a particular industry or data series, Federal Judicial Center, Reference Guide
on Multiple Regression (3d ed. 2012) at 334 (“Multiple regression…is a method in which a
regression line is used to relate the average of one variable—the dependent variable—to the
values of other explanatory variables.”) and is generally considered an appropriate tool for
estimating antitrust damages on a class-wide basis, see, e.g., In re Scrap Metal Antitrust Litig.,
527 F.3d 517, 532-34 (6th Cir. 2008); In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d
651, 660-61 (7th Cir. 2002) . See In re Foundry Resins Antitrust Litig., 242 F.R.D. 393, 411
(S.D. Ohio 2007) (characterizing multiple regression models as “reasonable damage
methodologies”), this Court must consider whether multiple regression is properly used here to
prove impact of the conspiracy across the broad range of products and producers represented in
the class. In re Flat Glass Antitrust Litig., 191 F.R.D. 472, 485-87 (W.D. Pa. 1999)
(“[t]here is no dispute that when used properly multiple regression analysis is one of the
mainstream tools in economic study and it is an accepted method of determining damages in
antitrust litigation”); In re Urethane Antitrust Litig., 237 F.R.D. 440, 452 (D. Kan. 2006)
(“damages are also likely susceptible to class-wide proof” using multiple regression analysis); In
re Linerboard Antitrust Litig., 305 F.3d 145, 153-55 (3d Cir. 2002) (same).
The realities of the steel industry reveal that it is unlikely that class members were either
impacted at the same levels or suffered the same damages. Still, class certification will not be
precluded simply because a class includes persons who have not been injured by the defendant’s
conduct. Kohen v. Pac. Inv. Mgmt. Co. LLC, 571 F.3d 672, 677 (7th Cir. 2009); Messner, 669

22

F.3d at 823 (quoting Schleicher v. Wendt, 618 F.3d 679, 687 (7th Cir. 2010). This is because
such an issue goes to the likelihood a class member will win on the merits; “[t]he chance, even
the certainty, that a class will lose on the merits does not prevent its certification.” Id. Even if I
can determine that there was some impact, the inevitable conclusion remains: Dr. Solow’s and
Dr. McClave’s methodology cannot be used, on a common basis, to measure accurately what
impact, if any, each member of the class experienced as a result of the alleged conspiracy. Even
if Plaintiffs determined that the overall price of steel increased, individual inquiries into the
damages each class member incurred would be required and predominate over issues common to
the class.
(d)

Efficiency of a class action

In determining whether predominance is satisfied, courts also consider the efficiency of a
class action as an alternative to individual suits. Butler v. Sears, Roebuck & Co., 727 F.3d 796,
800 (7th Cir. 2013) (finding that because the basic claim hinged on a question common to the
entire class, and the only individual issues concerned the amount of harm to particular class
members, it was more efficient for the principal issue to be resolved in a single proceeding than
for it to be litigated separately in hundreds of different trials). In particular, when adjudication of
questions of liability common to the class will achieve economies of time and expense, the
predominance standard is generally satisfied even if damages are not provable in the aggregate.
See Advisory Committee’s 1966 Notes on Fed. R. Civ. Proc. 23, 28 U.S.C.App., p. 141 (“[A]
fraud perpetrated on numerous persons by the use of similar misrepresentations may be an
appealing situation for a class action, and it may remain so despite the need, if liability is found,
for separate determination of the damages suffered by individuals within the class.”); 7AA
Wright, Miller, & Kane, § 1781, at 235-237.

23

Should the Court certify a class, the core issues for the Court to resolve would be
Defendants’ conduct and three elements of the claim (conspiracy, causation, and aggregate
damages). This Court has already determined that conspiracy may be proven by common
evidence. Whereas in the instance case, there may be differences among the members as to the
amount of damages incurred, it does not mean that a class action would be inappropriate. Rather,
the question of damages can be severed from that of liability and tried on an individual basis. In
re Allstate Ins. *665 Co., 400 F.3d 505, 508 (7th Cir.2005); Carnegie v. Household Int'l, Inc.,
376 F.3d 656, 661 (7th Cir.2004)) (“[I]f common questions do not predominate over the
individual questions so that class certification of the entire action is warranted,” a court may use
Rule 23(c)(4) to “isolate the common issues ... and proceed with class treatment of these
particular issues.”).
Allowing split proceedings furthers the Rule 23 purpose of promoting judicial economy
since the main issue will be tried only once, rather than for each class member, and damage
claims only need be determined in the event liability is found. § 1781 Class Actions in Which
Common Questions Predominate Over Individual Questions—Antitrust Actions, 7AA Fed. Prac.
& Proc. Civ. § 1781 (3d ed.).
2. Superiority
The last prerequisite for class certification is that the class action be “superior to other
available methods for the fair and efficient adjudication of the controversy.” Fed.R.Civ.P.
23(b)(3); Joseph v. Gen. Motors Corp., 109 F.R.D. 635, 642 (D. Colo. 1986). Manageability is
one of the key factors courts consider when determining superiority of a proposed class action.
Windham v. Am. Brands, Inc., 565 F.2d 29, 65 (4th CIr. 1977), cert. denied, 435 U.S. 968 (1978)
(it is a “firmly established principle that the issue of manageability of a proposed class action is

24

always a matter of ‘justifiable and serious’ concern for the trial court”)(citation omitted). Courts
should take a “practical, on-the-ground assessment of whether the proposed efficiencies from a
class action will outweigh the administrative problems and inefficiencies expected to ensure.”
Hamilton v. O’Connor Chevrolet, Inc., No. 02 C 1897, 2006 WL 1697171, at *13 (N.D.Ill. June
12, 2006). The more claimants there are, the more likely a class action is to yield substantial
economies in litigation, for purposes of determining whether a class action was superior to other
methods for fairly and efficiently adjudicating a controversy. Fed. Rules Civ. P. 23(b)(3); Butler.
I find that certifying the class on the issue of conspiracy and leaving resolution on the
question of impact and damages on an individual basis to be the most efficient and appropriate
course of action in this matter.
IV.

Conclusion

For the following reasons, I grant, in part, and deny, in part, Plaintiffs’ Motion for Class
Certification. Plaintiffs are certified as a class for the purpose of determining whether Defendants
engaged in a conspiracy in violation of 15 U.S.C. § 1. I decline to certify a class on the issues of
impact and damages.

ENTER:

James B. Zagel
United States District Judge
DATE: September 9, 2015

25