Opinion
Civil Action 4:22-cv-62-SA-JMV
04-14-2023
DYAMONE WHITE, ET AL. PLAINTIFFS v. STATE BOARD OF ELECTION COMMISSIONERS, ET AL. DEFENDANTS
ORDER DENYING MOTION TO STRIKE ON SATISFACTION OF CONDITIONS
JANE M. VIRDEN, UNITED STATES MAGISTRATE JUDGE.
This matter is before the court on Defendants' motion [119] to strike Plaintiffs' rebuttal report of Traci Burch, Ph.D. (“Dr. Burch”) and a portion of Plaintiffs' rebuttal report and corrected rebuttal report of Byron D'Andra Orey, Ph.D. (“Dr. Orey”) because these disclosures exceed the scope of permissible expert rebuttal or supplementation pursuant to FRCP 26, and the governing factors weigh in favor of striking them. For the reasons discussed below, but only on the conditions specified hereafter, the court will deny the motion to strike. In the event the conditions specified hereafter are not satisfied within the time set forth below, the motion will be deemed granted.
This Section 2 Voting Rights Act (“VRA”) action, filed on April 25, 2022, challenges MISS. CODE ANN. § 9-3-1, the 1987 statute that defines the three election districts from which Mississippi elects its Supreme Court Justices, Public Service Commissioners, and Transportation Commissioners. A Case Management Order [47], was entered on July 19, 2022, setting forth, in relevant part, a trial date and an approximate 9-month discovery period set to expire on April 19, 2023. This discovery period, far from being “expedited” as Plaintiffs now contend, is actually 3 months longer than the typical 6-month discovery period assigned to cases in this district.
The CMO also set the deadlines for designation of expert witnesses and, consistent therewith, on October 3, 2022, Plaintiffs served their expert witness disclosures, including the initial reports of political science professors, Dr. Burch and Dr. Orey. In a nutshell, Dr. Burch's initial report was that Black Mississippians vote at a lower rate of turnout than White Mississippians, and that this lower turnout rate could be explained by the disparity in educational attainment. For her opinion, she relied on the data obtained from the Current Population Survey Voting and Registration Supplement (“CPS data”) supplied by the U.S. Census Bureau. In Dr. Orey's initial designation he opined, in relevant part, but offered nothing in support thereof, that the outcome of a formally non-partisan race, like that for the Supreme Court, could not be driven by partisanship.
Pursuant to an Agreed Order [73] extending Defendants' expert designation deadline to January 6, 2023, Defendants on that date served their expert witness disclosures, which consisted of, among others, the reports of Dr. Swanson (expert in demography) and Dr. Bonneau (expert in political science). Dr. Swanson's report, as discussed below, stated that Dr. Burch's expert report contained a flawed analysis of the CPS data on which she relied. Dr. Swanson opined that that data, when properly analyzed, actually supported the finding that Black Mississippians do not have a lower rate of voter turnout than White Mississippians. In Dr. Bonneau's report, in relevant part, he opines, based on data and analyses, that party rather than race explains racially polarized voting in Mississippi. He offers this information in opposition to Dr. Orey's unsupported opinion to the contrary.
On February 6, 2023, Plaintiffs served, of relevance here, what were styled “rebuttal expert reports,” from Dr. Burch and Dr. Orey [111].
The rebuttal report of Traci Burch, Ph.D.
Dr. Burch's initial report relied upon CPS data to support her opinions addressing voter turnout by race and education level. In defense expert Dr. Swanson's report responding to Dr. Burch's initial opinions, he also presented opinions on voter turnout that relied upon his own analysis of CPS data and opined that the voter turnout opinions that Dr. Burch presented in her initial report were predicated on a flawed analysis of CPS data.
In her “rebuttal” report [119-8] served February 6, 2023, Dr. Burch acknowledges multiple errors in her analysis of the CPS data. Specifically, she states that “Dr. Swanson is correct that the estimates in my initial report reflect a calculation error” as it relates to the inclusion of children aged 15-17 in an educational attainment variable. Dr. Burch further states that in preparing her initial report, she erroneously “thought that the educational attainment variable that I was using excluded children.” She also states that she “calculated total turnout for both racial groups incorrectly.” In an apparent effort to rectify these issues with her initial report, Dr. Burch states in her rebuttal report that she now thinks that the CPS data - which she relied upon to support the voter turnout opinions presented in her initial report - “is not reliable as a benchmark for voter turnout” or “for voter turnout by race.” Dr. Burch further asserts in her rebuttal report that “Dr. Swanson's analysis is flawed” as it relates to voter turnout by race because he, too, used the CPS data in formulating his opinions. Further, in her rebuttal report, Dr. Burch states that she has “conducted additional analyses which employed alternative methods of looking at voter turnout by race.” These analyses, which include the “logit regression analysis” are new and are purportedly predicated on data obtained from the “2020 Cooperative Election Study (CES)” (“CES data”). The CES is a set of data that is nowhere used or identified in Dr. Burch's initial report. The logit regression analysis, as newly employed by Dr. Burch to analyze the newly identified CES data, resulted in new regression tables and probability statistics related to voter turnout by race, and Dr. Burch uses this new analysis in her rebuttal report to bolster her opinion “that Black voter turnout is lower than white turnout.” Additionally, Dr. Burch states in her rebuttal report that “[t]o further bolster [her] CES analysis,” she is presenting “a second method of estimating the racial gap in turnout” - namely, “ecological inference (EI)” (“EI”). She describes EI as “us[ing] Bayesian statistical methods to estimate voting behavior.” No such EI analysis appears in Dr. Burch's initial report, but in her rebuttal report she nevertheless asserts that her initial opinions are supported by her newly performed EI analysis using newly identified CES data. Dr. Burch's “rebuttal” report also purports to bolster her opinion regarding voter wait times, which she asserts affect voter turnout. And again, to do so she relies exclusively on her newly performed analysis of newly identified CES data.
The rebuttal report of Byron D'Andra Orey, Ph.D.,
At Paragraphs 6-8 and Table 1 of his rebuttal report [119-9] served February 6, 2023, and his “corrected” rebuttal report [119-1] served February 24, 2023, Dr. Orey seeks to offer new EI and empirical analyses to bolster his initial opinion that the outcome of a formally non-partisan race, like that for the Supreme Court, could not be driven by partisanship. In particular, in Paragraph 6 of his rebuttal report and corrected rebuttal report, Dr. Orey states that he has “conducted additional EI analyses on two other endogenous/quasi-endogenous contests.” In Paragraph 7 of both rebuttal reports, Dr. Orey states that he “ha[s] also conducted an empirical analysis to provide evidence that blacks and whites prefer different candidates.” Finally, in Paragraph 8 of both rebuttal reports, Dr. Orey states that “[c]onsistent with my previous report submitted on October 3, 2022, I conduct an EI analysis of the 2011 primary election,” which he further describes in Table 1. Table 1 describes Dr. Orey's analyses of the three new elections referenced in Paragraphs 6-8 of his rebuttal report. None of these elections was among the 17 elections between 2011 and 2020 analyzed in Dr. Orey's initial report to prove that the current Mississippi Supreme Court districts afford Black voters less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. According to Dr. Orey, these 3 newly analyzed elections are offered to rebut the opinion of Dr. Bonneau that party rather than race explains racially polarized voting in Mississippi.
The Law
L.U.Civ.R. 26(a)(2) “requires ‘full and complete disclosure' of expert materials no later than the time specified in the Case Management Order.” Kee v. Howard L. Nations, P.C., Civil Action No. 4:20-cv-00127-SA-JMV, 2021 WL 5370322, at *1 (N.D. Miss. Nov. 16, 2021). “Absent a finding of just cause, failure to make full expert disclosures by the expert designation deadline is grounds for prohibiting introduction of that evidence at trial.” Id. (quoting L.U.Civ.R. 26(a)(2)) (internal quotation marks omitted).
On the other hand, FRCP 26(a)(2)(D)(ii) “allows for a rebuttal report by an expert so long as the report is intended solely to contradict or rebut evidence on the same subject matter identified” in an expert report served by another party. La. Health Care Self Ins. Fund v. United States, Civil Action No. 12-766-JJB-RLB, 2014 WL 3720526, at *1 (M.D. La. July 25, 2014). FRCP 26(e)(2) also provides for supplemental expert reports to present “additions or changes,” FED. R. CIV. P. 26(e)(2), to expert information that do not constitute “material additions to the initial report.” McReynolds v. Matthews, Civil No. 1:16-CV-318-HSO-MTP, 2017 WL 5573194, at *5 (S.D.Miss. Nov. 20, 2017) (quoting Harmon v. Ga. Gulf Lake Charles, L.L.C., 476 Fed.Appx. 31, 38 (5th Cir. 2012)). “The purpose of rebuttal and supplementary disclosures is not to provide an extension of the deadline by which a party must deliver the lion's share of its expert information.” DAK Americas Miss., Inc. v. Jedson Eng'g, Inc., Civil No. 1:18cv31-HSO-JCG, 2019 WL 8375811, at *3 (S.D.Miss. Sept. 27, 2019). Rather, FRCP 26(a) requires a party's initial expert disclosures to be “complete and detailed.” See id. (citing Sierra Club, Lone Star Chapter v. Cedar Point Oil Co., 73 F.3d 546, 571 (5th Cir. 1996)). “Consistent with the above, it is appropriate to strike or exclude from consideration expert affidavits which are filed after the expert disclosure deadline and which amount to new opinions.” Cooper v. Meritor, Inc., No. 4:16-CV-52-DMB-JMV, 2019 WL 545187, at *6 (N.D. Miss. Feb. 11, 2019). Nevertheless, where the testimony at issue is important and a continuance would cure much of the prejudice, if any, to the opposing party, exclusion is “disproportionately harsh.” In re Complaint of C.F. Bean L.L.C., 841 F.3d 365, 374 (5th Cir. 2016) (“The Fifth Circuit has ‘repeatedly emphasized that a continuance is the preferred means of dealing with a party's attempt to designate a witness out of time.'”) (quoting Betzel v. State Farm Lloyds, 480 F.3d 704, 708 (5th Cir. 2007)).
Further, the decision of whether to strike expert disclosures as untimely or otherwise improper is within the Court's discretion and involves the consideration of four factors: (1) the explanation for the untimely/improper disclosure; (2) the importance of the testimony; (3) the potential prejudice in allowing the testimony; and (4) the availability of a continuance to cure such prejudice. See Kee, 2021 WL 5370322 at *3; see also Harvey v. Caesars Entm't Operating Co., Civil Action No. 2:11CV194-B-A, 2014 WL 12653851, at *3 (N.D. Miss. May 6, 2014).
The rebuttal report of Traci Burch, Ph.D., exceeds the scope of permissible expert rebuttal or supplementation
While as Plaintiffs point out, the line between what is proper or improper rebuttal or supplementation can sometimes be blurry, the undersigned finds that with respect to the so-called rebuttal opinion of Dr. Burch, it could not be clearer. This is not a proper rebuttal opinion or a mere supplement to her original opinion. While the initial opinion of Dr. Burch that White voter turnout exceeds Black voter turnout did not change, the rebuttal report is based on an entirely new set of data and analyses of that data that was occasioned by Dr. Burch realizing, after reading Dr. Swanson's expert report, that when properly analyzed, the CPS data actually supported the opposite of her opinion. Moreover, in addition to an entirely new data set and new means of analysis, the so-called rebuttal offers up entirely new opinions: that CPS data is unreliable and that CES data is reliable. It is plain that this is not proper rebuttal or supplementation.
Portions of the rebuttal report of Byron D'Andra Orey, Ph.D., likewise exceed the scope of permissible expert rebuttal or supplementation
Dr. Orey's initial report analyzed 17 elections between 2011 and 2020 in an attempt to prove that the district lines used in Mississippi Supreme Court elections grant Black voters less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. His “rebuttal” report adds three new elections between 2011 and 2016, the analysis of which he asserts is offered to rebut the opinion of the defendants' expert, Dr. Bonneau, “that party rather than race explains racially polarized voting in Mississippi.” Were that the whole story, the undersigned likely would find the rebuttal report proper, but the rest of the story dictates otherwise. In this case, Dr. Orey's initial report opined that party politics has no role in judicial elections, but he elected not to offer any data or analysis in support thereof. Having done so and thereby having usurped Defendants' expert's opportunity to address that data or analysis, Plaintiff should not be free to offer it under the guise that it is proper rebuttal or supplementation.
Analysis of the governing factors
1. Have Plaintiffs offered a reasonable explanation for the improper rebuttal disclosures of Dr. Burch and Dr. Orey?
Plaintiffs' explanation for not timely making the disclosure of Dr. Burch's new opinions and analyses is essentially that she did not earlier realize the CPS data when properly analyzed contradicted her opinion on voter turnout among Black and White Mississippians. Once she realized this, she determined that the CPS data was unreliable (due to asserted overreporting of voter participation) even though it is the data she elected to rely on in her initial report and is the data that she has relied on as an expert witness in similar cases.
Moreover, as Defendants point out in their reply, to the extent Dr. Burch has concerns about CPS data due to overreporting of voter participation, she would have had those concerns -though apparently did not disclose them - when she elected to rely on that data as an expert. She only concluded the data was unreliable due to overreporting when she determined that once a mathematical error in her initial report was corrected for, the data actually undermined her opinion. I do not find this explanation affords a reasonable basis to excuse the untimely rebuttal. This factor weighs in favor of striking Dr. Burch's rebuttal expert report.
Plaintiffs' explanation for the untimeliness of the contested portions of the rebuttal report of Dr. Orey is that it was not necessary that he offer any support for his initial opinion that race, not partisanship, drives polarized voting in Supreme Court elections until Defendants' expert provided an analysis to demonstrate otherwise. This explanation is wholly unpersuasive and contrary to law. L.U.Civ.R. 26(a)(2) “requires ‘full and complete disclosure' of expert materials no later than the time specified in the Case Management Order.” Kee, 2021 WL 5370322, at *1. This factor weighs in favor of striking the contested portions of Dr. Orey's report.
2. The importance of the challenged testimony
Defendants argue, first, that because the motion to strike the rebuttals will not preclude Drs. Burch and Orey from testifying altogether, this factor is neutral. I disagree. The mere fact that an expert can testify on some subjects, but not others, does not establish that the contested testimony is not important. Secondly, Defendants argue that if these matters were important to Plaintiffs' case, they would have addressed them in their initial reports. I find that the fact that the plaintiffs should have addressed these matters in the initial reports does not necessarily mean they are unimportant. On the contrary, for example, while the defendants make this argument, at the same time they appear to acknowledge that the plaintiff is obligated to prove “whether partisanship or race drives observed polarized voting.” See [135] at 8. And similarly, Plaintiffs now contend the Fifth Circuit has identified the question of whether partisanship or race drives observed polarized voting is an important (and potentially dispositive) one. See League of United Latin Am. Citizens, Council No. 4434 v. Clements, 999 F.2d 831, 850 (5th Cir. 1993) (en banc). With regard to Dr. Burch's rebuttal report, Plaintiffs similarly argue that the question of relative voter turnout rates by race has been found to be significant by the Fifth Circuit in analyzing the Senate Factors in voting rights litigation, and they argue striking Dr. Burch's disclosure will leave Plaintiffs with no expert evidence on this topic. NAACP v. Fordice, 252 F.3d 361, 368 (5th Cir. 2001). I find based on the foregoing that this factor weighs in favor of denying the motion to strike as concerns both Dr. Burch and Dr. Orey.
3. Will Defendants be unfairly prejudiced if the untimely rebuttals are permitted
Defendants argue that the demographic analysis that underpins the expert opinions at issue in this VRA Section 2 case is complex and highly specialized, and accordingly they have expended considerable time and financial resources to enable their applied demography expert, Dr. Swanson, to analyze and respond to the voter turnout opinion that Dr. Burch presented in her initial report. Namely that “Black people in Mississippi have had less access to quality education and therefore have lower educational attainment for the reasons discussed in this section; this lower educational attainment leads to lower voter turnout.” Defendants point out that the only data supporting her opinion, was her calculation, based on CPS data, that “56.1% of white Mississippi citizens voted in the 2020 general election, compared with 53.0% of Black Mississippi citizens.” But, as Dr. Swanson explained in his expert report, and Plaintiffs appear to concede, Dr. Burch inaccurately included individuals under 18 years of age as eligible voters. By using the wrong number of eligible voters, she necessarily determined the wrong percentage of participation by actually eligible voters. As correctly calculated, Dr. Swanson's analysis of the CPS data revealed that in every year since 2012, Black voter turnout exceeded White voter turnout in Mississippi. And, in a further effort to test Dr. Burch's opinion that an “overall gap in turnout between Black and white Mississippians exists,” Dr. Swanson examined an additional set of data from The Social Science Research Center at Mississippi State University that has conducted annual statewide surveys of registration and voting frequency from 2015 to 2021. Dr. Swanson analyzed this data and determined that it, too, indicated that Black turnout generally exceeds White turnout in Mississippi.
As noted, in her rebuttal report, Dr. Burch admits that she miscalculated turnout in the 2020 general election but rather than merely correct her mathematical error, she throws out the data set (the CPS) on which she relied in her initial report and adopts an entirely new data set (the CES) and new analyses to conclude “that 60% of White respondents voted in the 2020 General Election, compared with 46% of Black Mississippi respondents.”
In order to examine the veracity of Dr. Burch's new opinion, Dr. Swanson has explained by way of Declaration that he will need to examine the new data source-viz., the CES Data-on which she relies for the first time, as well as the new models she has built from new data sources. See [119-14]. Acquisition and analysis of an entirely new source of data, together with checking Dr. Burch's use of that data, will require, according to Dr. Swanson, a substantial amount of time and effort. As set forth in detail in his declaration, Dr. Swanson estimates that it would take between 164 to 180 man-hours of additional work to critically examine and respond in writing to the results presented in Dr. Burch's rebuttal report. This is in addition to the eight hours he reports already having spent reading and assessing Dr. Burch's rebuttal report for purposes of preparing his declaration.
Similarly, Dr. Bonneau will be required to perform additional research and analysis to investigate the veracity of the EI and empirical analyses performed by Dr. Orey in connection with three newly analyzed elections.
Moreover, Plaintiffs' suggestion that there is no real prejudice because there is plenty of time to do the additional work necessitated by the untimely disclosures of Dr. Burch and Dr. Orey is simply unrealistic given that discovery is scheduled to end on April 19, 2023, the dispositive and Daubert motions deadline is June 1, 2023, and the trial date is not far behind. In fact, with a trial setting of December 4, 2023, even if a continuance of the discovery deadline were sought, the most that could be allowed would amount to roughly 50 business days from the current deadline. Given Dr. Swanson's estimation of the time he would need to adequately address Dr. Burch's new data, opinions, and analyses, alone, such an extension would not cure the problem occasioned by the untimely new opinions, data and analyses.
The four-month rule requires that at least four months remain between the motions deadline and the trial date, in a given case. As such, the latest motions deadline, without a trial continuance, would be August 4, 2023. Accordingly, the discovery deadline would be on or about July 5, 2023 (July 4, 2023, being an official holiday), which is roughly 50 business days from the current discovery deadline.
The court notes, based on Dr. Swanson's estimation of 164 to 180 man-hours, this would mean roughly 3.5 hours of a typical 8-hour workday for the entire 50 business day extension would need to be devoted to this endeavor. Even then, no time would then be afforded to take discovery of Dr. Swanson on those opinions, and no time would be left for Dr. Swanson to even begin to address Dr. Orey's new data and analysis.
Where a defendant is forced to incur additional expense due to a plaintiff's improper expert disclosures, the prejudice factor weighs in favor of the defendant and exclusion of the challenged disclosures. Cf. Raymond James Trust, N.A., Trustee of E.C. Care Trust v. Natchez Hosp. Co., Civil Action No. 5:19-CV-103-DCB-MTP, 2021 WL 2556593, at *4 (S.D.Miss. June 22, 2021). Because it is clear that substantial expense and time will be required to address the new analyses and data set forth in the “rebuttal” reports of Drs. Burch and Orey, I find that this factor weighs in favor of granting the motion to strike.
4. Would a continuance cure the prejudice
Because the prejudice to the defendants in this case stems from both a lack of time to complete discovery and substantial expense, it is plain that a trial continuance - and accompanying extension of the discovery and dispositve/Daubert motions' deadlines - alone will not cure the prejudice occasioned. However, I find that if Plaintiffs move for a trial continuance (in order to accommodate an extension of the related deadlines) and also stipulate that, if a trial continuance is granted, they will pay the reasonable expert fees and costs actually incurred by Defendants in having their experts respond to the untimely rebuttal opinions, such prejudice can be substantially eliminated.
FRCP 37 allows for the court to order payment of reasonable expenses occasioned by a party's failure to timely comply with FRCP 26(a). Fed.R.Civ.P. 37(c)(1)(A).
Conclusion
Accordingly, IT IS THEREFORE ORDERED that Plaintiffs shall have seven (7) business days in which to both move for a trial continuance and stipulate that, should it be granted, they will be responsible for reasonable expert fees and costs actually incurred by Defendants in having their experts respond to the untimely rebuttal opinions of Drs. Burch and Orey. Should Plaintiffs fail to do either, or both, the motion to strike the “rebuttal” reports will be granted effective on the 8th business day from the date of this Order.
SO ORDERED.