McAllister v.St. Louis Rams, LLC

Filing 358

MEMORANDUM AND ORDER..IT IS HEREBY ORDERED that this matter (4:16cv189) shall be REMANDED to the Circuit Court for the City of St. Louis, Missouri.IT IS FURTHER ORDERED that the defendants motion to dismiss (#49) is DENIED as moot. Signed by District Judge Stephen N. Limbaugh, Jr on 3/21/18. (MRS)

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF MISSOURI
EASTERN DIVISION
RONALD MCALLISTER,
Plaintiff,
v.
THE ST. LOUIS RAMS, LLC,
Defendant.

)
)
)
)
)
)
)
)
)
)

No. 4:16-CV-172 SNLJ
No. 4:16-CV-262
No. 4:16-CV-297
No. 4:16-CV-189

MEMORANDUM AND ORDER
This matter consists of four consolidated cases all pertaining to the St. Louis Rams
football team’s move from Missouri to California in 2016. This matter is before the Court
on plaintiff James Pudlowski’s motion to reconsider the Court’s order denying plaintiff’s
motion to remand (#48) and Pudlowski’s subsequent affidavit in support of his arguments
in favor of remand (#127).
I.

Background
Plaintiffs filed their first complaint in state court, and the defendant removed the

case to this Court citing diversity jurisdiction generally1 and, in the alternative, “minimal
diversity” under the Class Action Fairness Act, 28 U.S.C. § 1332(d)(2), (d)(5) (“CAFA”).
Plaintiff filed a First Amended Complaint (“FAC”) and with it a motion to remand to
state court. The FAC attempted to clarify that the class was comprised only of Missouri
citizens, rendering any “minimal diversity” arguments inaccurate for CAFA purposes.
Plaintiff also argued that even if CAFA jurisdiction were present, then the “local

1

It appears that the parties agree that standard diversity jurisdiction, 28 U.S.C. § 1332(a), is not
present here.
1

controversy exception” applied to require remand because more than two-thirds of the
plaintiff class are Missouri citizens. See 28 U.S.C. § 1332(d)(4)(A).
Defendants responded that the original petition created minimal diversity because
the class definition used the term “Missouri resident,” leaving open the possibility that a
non-Missouri citizen who was also a Missouri resident could be a member of the class
and thus provide diversity of citizenship. The defendants argued that the Court should
not consider the FAC, which revised the class definition to include only Missouri citizens
who were Missouri citizens at the time of the petition’s filing.
Plaintiffs replied and pointed out that defendants had not met their burden of proof
to establish CAFA jurisdiction because defendants had not identified any person who
created minimal diversity. Defendants sought and received permission to file a SurReply with attached affidavits from Missouri residents who were not Missouri citizens.
This Court did not consider the affidavits and remanded the matter to state court, but the
Eighth Circuit subsequently (and without ruling on any substantive matter) instructed this
Court to consider the affidavits before reaching a ruling on the remand issue (#55).
On remand from the Eighth Circuit, the plaintiffs filed a Sur-Response urging the
Court to consider its FAC definition of the class and citing Ninth Circuit law. Defendants
responded to the Sur-Response. This Court denied the motion to remand and held that
the FAC could not be considered and that minimal diversity exists.
Plaintiffs requested reconsideration of the Court’s order because, they say, the
Court did not consider the “local controversy exception” or the “home state exception”
to CAFA jurisdiction. Defendants responded that plaintiffs abandoned those arguments
because they had not been raised since the plaintiffs’ very first brief on the motion to
remand. Plaintiffs responded that they had not abandoned those arguments; rather, those
2

arguments were only relevant if the Court found “minimal diversity” under CAFA, which
plaintiff disputed from the beginning. Plaintiffs further note that because it was
defendants’ burden to establish CAFA jurisdiction, the burden was not on plaintiffs to
prove the exceptions were effective unless defendants met their burden.
The Court’s Memorandum and Order held that defendants had established each of
the three CAFA jurisdictional elements: (1) minimal diversity among the parties (that is,
any class member and any defendant are citizens of different states); (2) at least 100 class
members; and (3) an amount in controversy of more than $5 million. 28 U.S.C. §
1332(d)(2), (d)(5); Westerfeld v. Indep. Processing, LLC, 621 F.3d 819, 822 (8th Cir.
2010). The Court also held that the class definition in the first petition, not the FAC,
applied because the pre-removal complaint is the relevant document for determining
whether removal was appropriate. (#64 at 5.) Plaintiff points out that the original
petition explicitly stated that there is “no diversity of citizenship” among the parties, but
the Court did not address that matter, instead focusing on the clear language of the class
definition, which refers only to “Missouri residents.”
Plaintiffs argued that, in light of the Court’s conclusion that the CAFA
jurisdictional elements had been established, the Court should have applied the local
controversy exception to the facts of the case. Under the local controversy exception, a
district court must decline to exercise jurisdiction over a class action (1) in which more
than two-thirds of the class members in the aggregate are citizens of the state in which the
action was originally filed, (2) at least one significant defendant is a citizen of the state in
which the class action was originally filed, (3) the principal injuries were incurred in the
state in which the action was filed, and (4) no other class action alleging similar facts was

3

filed in the three years prior to the commencement of the current class action. 28 U.S.C.
§ 1332(d)(4)(A); Westerfeld, 621 F.3d at 822.
As this Court recognized, there appears to be no doubt that the second, third, and
fourth factors for the local controversy exception are met here. The parties’ dispute
pertains to the first factor, whether more than two-thirds of the class members are
Missouri citizens. This Court allowed plaintiff jurisdictional discovery regarding state
citizenship. See Hood v. Gilster-Mary Lee Corp., 785 F.3d 263, 266 (8th Cir. 2015)
(describing discovery process but holding that “last-known address” evidence was not
persuasive ). The Eighth Circuit approves of the Seventh Circuit’s approach to this sort
of analysis, in which
plaintiffs may “take a random sample of potential class members, ascertain
the citizenship of each on the date the case was removed, and extrapolate to
the class as a whole. If the sample yields a lopsided result, the outcome is
clear without the need for more evidence. If the result is close to the
statutory two-thirds line, then do more sampling and hire a statistician to
ensure that the larger sample produces a reliable result.”
Id. (quoting Myrick v. WellPoint, Inc., 764 F.3d 662, 665 (7th Cir. 2014) (internal
changes adopted)).
Plaintiffs conducted discovery and retained an expert to submit an expert report
regarding the citizenship of the proposed class. (#127.) Defendants hired their own
expert, who submitted his own expert report and criticized the plaintiffs’ expert’s
methodology. For the following reasons, the Court finds that the local controversy
exception applies to this case to defeat the Court’s CAFA jurisdiction. The matter will be
remanded, and the defendants’ motion to dismiss will be denied as moot.

4

II.

Discussion
Plaintiffs, as the party invoking the local controversy exception, have the burden

to prove the exception by a preponderance of the evidence. Hood, 785 F.3d at 265.
Plaintiffs retained Dr. Charles Cowan as their expert in statistics and survey design and
implementation. Dr. Cowan has his Ph.D. in Mathematical Statistics. He has served the
federal government as Chief Statistician for both the FDIC and the National Center for
Education Statistics. He has further served as Chief of the Survey Design Branch for the
U.S. Census Bureau and has over 40 years of experience with surveys.
Dr. Cowan took a random sample of potential class members and ascertained their
citizenship --- or domicile ---that is, where the person (1) resides, and (2) intends to
remain. Myrick, 764 F.3d at 664. His survey asked putative class members whether they
made a relevant purchase from defendants during the class period, whether they resided
in Missouri when they made their purchase and continued to reside in Missouri through
the time the case was filed, and whether they intended to continue to live in Missouri.
He received 107 responses and determined 93 were Missouri citizens. Thus 86.7% of
putative class members had Missouri citizenship. Dr. Cowan also determined this was a
random and statistically significant sample.
Defendants do not take issue with Dr. Cowan’s credentials.2 They do, however,
submit an expert report from Dr. Stephen Cacciola that disputes much of Dr. Cowan’s
conclusion. Dr. Cowan later responded and refuted the defendants’ criticisms. The Rams
filed a sur-reply with Dr. Cacciola’s further opinions, and the plaintiffs filed a sur2

Plaintiffs state in their sur-response that the Rams do attack Dr. Cowan’s credentials in the
Rams’ sur-reply. The Court characterizes the Rams’ criticisms as instead directed at alleged
“defects” in Cowan’s methodology and conclusions. The Rams cite other cases in which Dr.
Cowan’s testimony has been rejected. The Court addresses those matters below.
5

response. The discovery and briefing process took more than one year to finally
complete.
The Court will not go into the details of the parties’ statistical debates. Although
defendants go to great lengths to discredit Dr. Cowan’s survey, the Court finds Dr.
Cowan’s analysis and opinion persuasive, even in the face of defendants’ arguments to
the contrary. Defendants’ criticisms fall into three general categories, discussed briefly
below:
First, defendants argue that Dr. Cowan did not obtain a representative sample of
proposed class members. Dr. Cowan responds that the nature of his sampling frame was
accounted for in his results to ensure accuracy based on his experience, education, and
training and in accordance with standards in his field. With respect to defendants’
argument that Dr. Cowan’s data did not include tickets sold at the ticket window on game
days, Cowan concluded that, as with other factors, there was no statistical basis to
differentiate the citizenship rate of Missouri residents who purchased online versus those
who purchased at the ticket window. Further, with respect to defendants’ argument
regarding single- versus multigame purchasers, defendants’ expert --- who is, notably, an
economist and not a statistician --- does not demonstrate that any distinction between
single-game and multi-game purchasers exists to refute Dr. Cowan’s analysis. As for
defendants’ contention that Dr. Cowan’s survey of merchandise purchasers is flawed
because he had merchandise data only from 2014 forward, Cowan opines that “there is no
reason to believe that whether a person purchased relevant merchandise before or after
May 7, 2014 is related to their state citizenship.” (#288-1 at 7.) Dr. Cowan concluded
that the Missourian citizenship of Rams merchandise purchasers was over 90%.

6

Second, defendants point out that Dr. Cowan’s survey asked about Missouri
citizenship as of the date of the filing of the case (January 13, 2016) rather than its date of
removal (February 11, 2016). Although Dr. Cowan opines that statistically, the 29-day
difference has no relevance or impact on his student, Dr. Cowan applied the February
date to his survey in a follow-up questionnaire and found that the rate of Missouri
citizenship actually went up to 89.9%.3
Defendants posit that Cowan’s follow-up questionnaire was plagued by a
nonresponse bias, but Cowan explains that his follow-up survey had a response rate of
83%, which is more than sufficient for confirming his conclusion that Missouri
citizenship was well in excess of two-thirds on both the January and February dates.
Defendants also suggest that the interviewers and interviewees would have been biased
with respect to the citizenship answer because they knew the purpose of the survey was
to determine citizenship. However, Dr. Cowan explains that neither interviewers nor
interviewees knew how citizenship would be used in this case, nor did they have any
incentive or understanding that would bias them toward one answer over another.
Third, defendants argue that Dr. Cowan cannot reliably extrapolate from the
survey results to the proposed class. Plaintiffs respond that Dr. Cowan used standard

3

Defendant’s expert suggests that citizenship rate was undermined by the observation of two
additional move-outs among the follow-up respondents. Dr. Cowan explains that a
statistician must understand the distinction between an observed value and an
average. The two numbers cannot be compared in the way that Defendants have
done in their attempt to dismiss my work. When I recalculate the percent using
the additional month and the extra 2 move-outs --- they should have for a proper
comparison --- I get the exact same percent that I had previously (0.2%). That is,
the extra two move-outs had no impact on this calculation.
(#350 at 6.)
7

procedures within his profession and that the correction of the few errors defendants
found resulted in a drop in the Missouri citizenship rate from only 86% to 85%.
Finally, defendants argue that this Court should reject Dr. Cowan’s analysis for the
same reasons that other courts have rejected his testimony in other cases. Defendants cite
to four cases in which Dr. Cowan’s expert testimony has been rejected by a court.
Notably, Dr. Cowan has testified in over 50 cases over the last four years.4 The four
cases that defendants cite are based upon entirely different factual circumstances, in
different areas of law, involve different types of information being sought, and different
types of opinions being offered compared to this case. They involved antitrust, property
valuations, and loans in residential mortgage-backed security (“RMBS”) cases in which
sampling problems arose. Here, the Eighth Circuit explicitly allows for sampling. In
RMBS cases, Dr. Cowan explains, defendants frequently argue that sampling is
prohibited as a matter of law. Dr. Cowan’s methodology in this case is reliable, and the
data he used was sufficient to employ that methodology and arrive at a reliable
conclusion.
Ultimately, even accepting defendants’ criticisms, Dr. Cowan’s opinion results in
finding a Missouri citizenship rate significantly larger than two-thirds at a 95%
confidence level. Again, defendants do not suggest that Dr. Cowan is not qualified as an
expert in these matters, nor is there any apparent basis for defendants to do so. Dr.
Cowan’s opinion easily supports that more than two-thirds of plaintiffs’ class comprises
Missouri citizens. Plaintiffs’ burden of proof is by a “preponderance” --- or more than
50% --- of the evidence. The Court finds that plaintiffs’ approach was disciplined and
reliable so as to meet this burden.
4

In contrast, the Court notes that the Rams’ expert has apparently never testified as an expert.
8

Accordingly,
IT IS HEREBY ORDERED that this matter shall be REMANDED to the Circuit
Court for the City of St. Louis, Missouri.
IT IS FURTHER ORDERED that the defendants’ motion to dismiss (#49) is
DENIED as moot.
Dated this 21st day of March, 2018.

STEPHEN N. LIMBAUGH, JR.
UNITED STATES DISTRICT JUDGE

9