Civil Action No. 05-cv-00164-EWN-CBS.
March 10, 2006
MEMORANDUM ORDER REGARDING DEFENDANTS' MOTION TO STRIKE PLAINTIFFS' UNTIMELY EXPERT REPORT
THIS MATTER comes before the court on Defendants Roxane White, the Denver Department of Human Services and the City and County of Denver's Motion to Strike Plaintiffs' Untimely Expert Report (Document # 56), filed on December 21, 2005. Defendants' motion seeks to strike portions of a revised expert report served on December 2, 2005 by Dr. Robert Bardwell, an expert retained by Plaintiffs Freeman, McClenton, James and Dennis. The aforementioned Plaintiffs filed their Response to Defendants' Frivolous Motion to Strike on January 10, 2006. Defendants filed a Reply in Support of their Motion to Strike on January 13, 2006, and a recitation of Supplemental Authorities in Support of Their Motion to Strike on January 24, 2006. Plaintiffs filed a Supplement to their Response on Frivolous Motion to Strike on February 10, 2006. Pursuant to an Order of Reference dated February 1, 2005, this matter was referred to the Magistrate Judge to, inter alia, "issue orders necessary for compliance with the scheduling order, including amendments or modifications of the scheduling order upon a showing of good cause," and "hear and determine any pretrial matter, including discovery matters and non-dispositive motions." This court held a hearing on the pending motion on February 13, 2006, and has carefully considered the parties' briefs, the arguments of counsel, the entire case file and the applicable law. For the following reasons, Defendants' motion is granted in part and denied in part.
Pro se Plaintiffs Lewis and Smith apparently did not join in retaining Dr. Bardwell and they are not parties to the pending discovery dispute.
Plaintiffs initiated the instant action on January 31, 2005, alleging that they were the victims of discrimination on the basis of race. Plaintiffs, who were employees of the City of Denver's Department of Human Services, were terminated as part of a budgetary layoff. The Complaint alleges that Defendants manipulated the layoff selection process to eliminate a disproportionate number of African American employees, including Plaintiffs. According to Plaintiffs, "in implementing the layoffs, the Defendants used a variety of methods including the job code system, the elimination of certain positions, and the demotion process, to eliminate African American supervisors and employees from the department." See Scheduling Order, at 2.
Defendants deny that Plaintiffs were the victims of racial discrimination. To the contrary, Defendants insist that the layoffs in question were based upon layoff units approved and determined by the Career Service Board after public hearings. According to Defendants, the reduction in force affected all races and resulted in the layoff of a significantly higher number of Caucasian and Hispanic employees. Id. at 3.
The court held a Fed.R.Civ.P. 16 scheduling conference on April 25, 2005, which was attended by the attorney who originally represented all of the named plaintiffs. At that scheduling conference, I set a deadline of July 22, 2005 for designating affirmative experts pursuant to Fed.R.Civ.P. 26(a)(2)(B), and a deadline of August 22, 2005 for designating rebuttal expert witnesses. The court also set a deadline of September 19, 2005 for completing all discovery. The Scheduling Order submitted by the parties acknowledged that "Plaintiff (sic.) anticipates using an economic expert and possibility a statistics expert and that defendants would anticipate calling rebuttal experts." Id. at 8.
On June 17, 2005, Plaintiffs' original counsel moved to withdraw "due to irreconcilable differences" with his clients. The court held a hearing on July 5, 2005, at which time I granted counsel's motion to withdraw. I also set a status conference for July 14, 2005 and stayed all discovery deadlines pending that status conference. However, Plaintiffs were instructed to respond to Defendants' outstanding interrogatories on or before July 12, 2005. During the status conference on July 14, 2005, defense counsel objected to the sufficiency of some of the pro se Plaintiffs' discovery responses. Plaintiffs were directed to provide full and complete responses to interrogatories and requests for production by July 28, 2005.
On July 22, 2005, new counsel entered his appearance on behalf of Plaintiffs Freeman, McClenton, James and Dennis. On July 29, 2005, that attorney moved to amend the Rule 16 Scheduling Order, seeking to extend existing pre-trial deadlines by 60 days. In pertinent part, Plaintiffs Freeman, McClenton, James and Dennis indicated that "[a]s of July 22, 2005, Plaintiffs' prior attorney had correctly identified the need for an economics and statistical expert, but it appears that no such experts have either been employed or contacted." See Plaintiffs' Motion to Amend Scheduling Order, at 2. Plaintiffs Freeman, McClenton, James and Dennis argued that without the requested extension of time, "the referenced Plaintiffs will not be able to obtain expert witnesses necessary for the presentation of their claims and the proof of their damages, essentially ending their case." Id. at 4.
On August 22, 2005, this court extended various pretrial deadlines, including the deadlines for designating expert witnesses. Affirmative experts were to be designated on or before September 16, 2005 and rebuttal experts were to be designated by October 17, 2005. The deadline for completing fact discovery was September 30, 2005 and the deadline for completing expert discovery was October 31, 2005.
On or about September 19, 2005, Plaintiffs Freeman, McClenton, James and Dennis designated Dr. Robert Bardwell as an expert in the field of statistics and tendered Dr. Bardwell's 17-page "Preliminary Report on the Impact of Age (sic.) Race on Transfers and Terminations at the Denver Department of Human Services." See Exhibit A-1 to Defendants' Motion to Strike. Dr. Bardwell indicated that he was providing
The substance of Dr. Bardwell's Preliminary Report was set forth in nine of those 17 pages.
a preliminary report issued without the detailed data that would be required to study the pattern of position changes prior to the layoff. . . . This report may be amended and additional analyses performed if defendants produce the employee Transaction History Database, or other data which would detail the transfers that could have resulted in the patterns of statistically significant race bias presented in this report.Id. at 2. Apparently, Dr. Bardwell was under the impression "that defendants have refused to produce" the Transaction History Database. However, it does not appear that Freeman, McClenton, James and Dennis ever moved to compel production of this, or any other, data by Defendants. Dr. Bardwell's Preliminary Report specifically states that his analysis "is based on the Rank Order List by Jobcode, [EEO.xls], and the list of employees who were finally affected by the layoffs." See Exhibit A-1, at 4. The court has provided a "Summary of Findings" presented in Dr. Bardwell's "Preliminary Report" in Appendix I attached to this Memorandum Order.
Defendants tendered a Rebuttal Report to Plaintiffs' Preliminary Report on October 17, 2005. See Exhibit A-2 to Defendants' Motion to Strike. This report, prepared by Dr. Ronald Farina, relied on two Data Population Frame data sets: (1) "EEO.xls" based on the status of the Department of Human Services layoff eligibility list dated approximately January, 2004; and (2) "Rank Order List by Jobcode." Id. at 4. Both of these data sets were cited in Dr. Bardwell's Preliminary Report. It would appear that Dr. Farina and Dr. Bardwell relied on the same data sets. Indeed, during his deposition on October 31, 2005, Dr. Farina testified that he did not consider any data that was not available to Dr. Bardwell. According to Dr. Farina, "one of the things I was doing was looking at the analysis that Dr. Bardwell did, and I couldn't very well do that with a different data set." See Exhibit A-3, at 47-48, attached to Defendants' Motion to Strike.
On or about November 14, 2005, Dr. Bardwell completed his "Revised Report on the Impact of Race on Transfers and Terminations at the Denver Department of Human Services." In contrast to his earlier effort, Dr. Bardwell's Revised Report consisted of 34 pages of text and five pages of attachments. A "Summary of Findings" presented in Dr. Bardwell's "Revised Report" is included in Appendix II attached to this Memorandum Order. Dr. Bardwell stated that his Revised Report was submitted "in response to the report and additional data submitted by the defendant's expert, Dr. Farina." See Exhibit A-5 attached to Defendants' Motion to Strike, at 3. Dr. Bardwell concedes that the "analyses in this [revised] report are more detailed than those in my preliminary report, and eliminate any issues about the data used, since Dr. Farina's data file is used in all analyses." Id. While Dr. Bardwell claims to have prepared his Revised Report, in part, on "additional data submitted by the defendant's expert," that additional data is not specifically identified or described in the Revised Report. To the contrary, a careful review of Dr. Bardwell's Revised Report does not disclose any new data that was not available to Dr. Bardwell at the time he prepared his Preliminary Report. The court is left to conclude that Dr. Bardwell's Revised Report relies on the exact same data and documentation enumerated in his Preliminary Report.
This explanation seems at odds with Plaintiffs' Response to Defendants' Motion to Strike, which claimed that Dr. Bardwell's Revised Report "was made, and could only be made, once he had the opportunity to review Defendants' expert's deposition testimony." See Plaintiffs' Response, at 4.
Plaintiffs' Supplement, filed on February 10, 2006, contends that "Defendants provided their expert with information that had been requested by Plaintiffs but not provided to them." In the same Supplement, Plaintiffs accuse Defendants of providing "their expert witness with materials withheld from Plaintiffs' expert and only disclosed and described at or after the time for initial discovery." See Supplement to Plaintiffs' Response on Frivolous Motion to Strike, at 2. Notwithstanding these serious accusations, Plaintiffs never moved to compel further discovery responses and never moved for relief under Fed.R.Civ.P. 37.
On October 31, 2005, Defendants moved for summary judgment, arguing in part that Plaintiffs' claim of disparate impact is not actionable under 42 U.S.C. §§ 1981 and 1983. Defendants' brief took the position that
Because disparate impact claims are not actionable in §§ 1983 and 1981 claims, and because Plaintiffs could not, in any event, isolate and identify a specific employment practice that had a statistically significant impact for a prima facie case, Defendants do not address at this time whether Plaintiffs' expert's distorted statistical analysis utilizes the "appropriate comparables," . . . or whether it crosses a "threshold of reliability" for use in a prima facie case.See Defendants' Memorandum Brief in Support of Their Motion for Summary Judgment, at 32.
Plaintiffs' Response to the pending Motion for Summary Judgment contends that Dr. Bardwell's Revised Report "presents extremely strong evidence of race bias in the (2004) layoffs at DDHS." See Plaintiffs' Response to Defendants' Motion for Summary Judgment, at 19. According to Plaintiffs, Dr. Bardwell's "analysis demonstrates that the way in which Defendants restructured their consolidation codes right before the layoff; the way they moved people about, the way they changed the bumping rules between the 2003 layoff and the 2004 layoff, all conjoined to create a layoff strategy that reeks of racism." Id. at 51.
Defendants have moved to strike the Revised Report of Dr. Bardwell pursuant to Fed.R.Civ.P. 37(c), arguing that Plaintiffs Freeman, McClenton, James and Dennis without substantial justification failed to comply with Fed.R.Civ.P. 26(a)(2)(B) and Fed.R.Civ.P. 26(e)(1). The pending motion contends that Plaintiffs' substitution of a new statistical analysis in place of Dr. Bardwell's Preliminary Report after the close of discovery "is prejudicial and harmful to Defendants, and makes a mockery of the Court's scheduling order deadlines and the entire discovery process." See Defendants' Motion to Strike, at 4.
While conceding that Dr. Bardwell's Revised Report was tendered after the October 31, 2005 deadline for expert discovery, Plaintiffs characterize the pending motion to strike as "frivolous" in the absence of any showing that the Revised Report was made in bad faith. To the contrary, Plaintiffs suggest that Dr. Bardwell had an affirmative duty to provide a supplemental report that "simply clarified" his earlier analysis. See Plaintiffs' Response, at 4. In support of their position, Plaintiffs' Response and Supplement rely exclusively on this court's prior rulings in Sender v. Mann, 225 F.R.D. 645 (D. Colo. 2004) and Harvey v. United States, 2005 WL 3164236 (D. Colo. 2005). Notably, Plaintiffs' February 10, 2006 Supplement does not address any of the precedents cited in Defendants' Reply, filed on January 13, 2006, or Defendants' Supplement, filed on January 24, 2006. In short, Plaintiffs summarily dismiss Defendants' Motion to Strike as "frivolous" and demand the imposition of sanctions against Defendants.
As a threshold matter, the court must acknowledge the important role a scheduling order plays in the management of its docket. Cf. Washington v. Arapahoe County Department of Social Services, 197 F.R.D. 439, 441 (D. Colo. 2000) (noting that a "scheduling order is an important tool necessary for the orderly preparation of a case for trial"). See also Rent-a-Center, Inc. v. 47 Mamaroneck Ave. Corp., 215 F.R.D. 100, 101 (S.D.N.Y. 2003) ("scheduling orders are designed to offer a degree of certainty in pretrial proceedings, ensuring that at some point both the parties and the pleadings will be fixed and the case will proceed"). Where a party or party's attorney fails to comply with deadlines established in a scheduling order, Rule 16 of the Federal Rules of Civil Procedure provides the court with considerable discretion to impose sanctions and to "make such orders with regard thereto as are just." See Fed.R.Civ.P. 16(f). Some courts have held that untimely expert disclosures should be addressed pursuant to Rule 16(f), rather than Fed.R.Civ.P. 37(c). See, e.g., Akeva LLC v. Mizuno Corp., 212 F.R.D. 306, 309 (M.D.N.C. 2002). The sanctions available under either Rule are quite similar.
Both Rule 16(f) and Rule 37(c) permit the court to impose the sanctions contemplated by Rule 37a(b)(2)(B) and (C), and authorize the court to award reasonable expenses, including attorneys fees.
Rule 26(a) disclosures are designed to accelerate the exchange of basic information, to "help focus the discovery that is needed, and facilitate preparation for trial or settlement." See Advisory Committee Notes to 1993 Amendments to Fed.R.Civ.P. 26(a). The disclosures required by Rule 26(a) should provide the parties "with information essential to the proper litigation of all relevant facts, to eliminat[e] surprise, and to promot[e] settlement." Windom v. FM Industries, Inc., 2003 WL 21939033, *2 (D. Neb. 2003) (quoting Rolscreen Co. v. Pella Prods. of St. Louis, Inc., 145 F.R.D. 92, 94 (S.D. Iowa 1992)). See also City and County of San Francisco v. Tutor-Saliba Corp., 218 F.R.D. 219, 221 (N.D. Cal. 2003) (noting that Rule 26(a) seeks to "`accelerate the exchange of basic information' that is 'needed in most cases to prepare for trial or make an informed decision about settlement'").
In short, the Rule 26(a) disclosure requirements should "be applied with common sense in light of the principles of Rule 1, keeping in mind the salutary purposes that the rule is intended to accomplish. The litigants should not indulge in gamesmanship with respect to the disclosure obligations." Cf. Advisory Committee Notes to 1993 Amendments to Fed.R.Civ.P. 26(a). Counsel who make the mistake of treating Rule 26(a) disclosures as a mere technical formality do their clients no service and necessarily risk the imposition of sanctions. See Santiago v. Furniture Chauffeurs, Piano Movers, Packers and Handlers Local 705, 2001 WL 11058 *7 (N.D. Ill. 2001) ("civil litigation is not a game of hide-the-ball").
Specific disclosure requirements are imposed upon expert witnesses. Rule 26(a)(2)(B) "requires that persons retained or specially employed to provide expert testimony . . . must prepare a detailed and complete written report, stating the testimony the witness is expected to present during direct examination, together with the reasons therefor." See Advisory Committee Notes to 1993 Amendments to Fed.R.Civ.P. 26(a) (emphasis added). See also Reed v. Binder, 165 F.R.D. 424, 428 n. 6 (D.N.J. 1996) (an expert report should indicate "`what' the expert saw, heard, considered, read, thought about or relied upon in reaching the conclusions and opinions to be expressed"). The drafters of the 1993 Amendments anticipated that Rule 26(a)(2)(B) disclosures would eliminate unfair surprise to an opposing party and promote the efficient use of resources.
The information disclosed under the former rule in answering interrogatories about the "substance" of expert testimony was frequently so sketchy and vague that it rarely dispensed with the need to depose the expert and often was even of little help in preparing for a deposition of the witness. Revised Rule 37(c)(1) provides an incentive for full disclosure; namely, that a party will not ordinarily be permitted to use on direct examination any expert testimony not so disclosed.See Advisory Committee Notes to 1993 Amendments to Fed.R.Civ.P. 26(a). Another court has cogently observed that Rule 26
contemplates that each expert will work diligently to amass the factual data necessary for his expert analysis and, having done that, will prepare and submit a timely and comprehensive report complying fully with the requirements of Rule 26(a)(2)(B). With that report in hand, and on the basis of the justifiable assumption that the report as submitted may be relied upon as a definitive disclosure of the testimony (including the opinions and bases therefor) of the expert, the opposing party will be in a position . . . to determine whether to "arrange for expert testimony from other witnesses. . . ."Dixie Steel Erectors, Inv. v. Grove U.S., LLC, 2005 WL 3558663, *9 (W.D. Ok. 2005). See also Salgado v. General Motors Corp., 150 F.3d 735, 741 n. 6 (7th Cir. 1998) ("expert reports are to be a detailed and complete statement of the testimony of the expert on direct examination"); Sharpe v. United States of America, 230 F.R.D. 452, 458 (E.D. Va. 2005) (noting that an "expert's report must be `detailed and complete' in order to `avoid the disclosure of sketchy and vague expert information'").
Plaintiffs and Dr. Bardwell cannot circumvent the expert disclosure requirements by characterizing his September 19, 2005 product as a Preliminary Report. Rule 26(a)(2)(B) does not contemplate the submission of a "preliminary" expert report. Cf. In re TMI Litigation Cases Consolidated II, 922 F. Supp. 997, 1005 n. 9 (M.D. Pa. 1996) (suggesting that a "preliminary report" exposes the opposing party to unfair surprise and the risk of ambush), rev'd in part on other grounds, 193 F.3d 613 (3d Cir. 1999). To the contrary, an opposing party is entitled to rely on a written report "as a definitive statement of the expert's direct testimony and of the basis for that testimony." Dixie Steel Erectors, Inc. v. Grove U.S. LLC, 2005 WL 3558663, *6 (W.D. Ok. 2005) (noting that expert disclosures under Rule 26 should not be an iterative process). See also Kern River Gas Transmission Co. v. 6.17 Acres of Land, 156 Fed.Appx. 96, 102 (10th Cir. 2005) (noting that a "preliminary" expert report does not comply with Rule 26(a)(2), which requires a "complete statement of all opinions to be expressed and the basis and reasons therefor").
Indeed, Fed.R.Civ.P. 37(a)(3) recognizes that an evasive or incomplete disclosure should "be treated as a failure to disclose."
Defendants never objected to the "preliminary" nature of Dr. Bardwell's September 19, 2005 report, notwithstanding the literal requirements of Rule 26(a)(2)(B) and the prevailing case law. They have, however, challenged the November 14, 2005 Revised Report as untimely and improper under Rule 26(e)(1). Thus, resolution of the pending motion turns on whether the Revised Report qualifies as an untimely new report or merely as a supplement under Rule 26(e)(1), for which there was no established deadline under the parties' Scheduling Order.
Certainly, "an expert's initial Rule 26 report cannot always anticipate every possible challenge to the report." Miller v. Pfizer, Inc., 356 F.3d 1326, 1332 (10th Cir. 2004), cert. denied, 543 U.S. 917 (2004). Rule 26 requires a party to supplement its expert disclosures "at appropriate intervals" if that party "learns that in some material respect the information disclosed is incomplete or incorrect and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing." See Fed.R.Civ.P. 26(e)(1). Unfortunately, Rule 26 does not provide a bright line for distinguishing between a supplemental expert report and a new report that essentially replaces an original.
A plain reading of Rule 26(e)(1) suggests that a supplemental report should be based upon additional or corrective information that was not available at the time of the expert's original report. See Minebea Co., Ltd. v. Papst, 231 F.R.D. 3, 6 (D.D.C. 2005) (Rule 26(e)(1) permits supplemental reports only for the narrow purpose for correcting inaccuracies or adding information that was not available at the time of the initial report). Any other interpretation would defeat the requirements of Rule 26(a)(2)(B), which mandates a "detailed and complete written report." Cf. In re Safeguard Scientifics, 2004 WL 2644393 (E.D. Pa. 2004) (noting that an expert is required to supplement a prior disclosure "[w]here additional relevant information because available such [that] the initial expert report is rendered `incomplete or incorrect'").
Rule 26(e) envisions supplementation when a party's discovery disclosures happen to be defective in some way so that the disclosure was incorrect or incomplete and, therefore, misleading. It does not cover failures of omission because the expert did an inadequate or incomplete preparation. To construe supplementation to apply whenever a party wants to bolster or submit additional expert opinions would reek havoc in docket control and amount to unlimited expert opinion preparation. Akeva, LLC v. Mizuno Corp., 212 F.R.D. at 310. The Rule 26(e)(1) duty to supplement cannot be transformed into a tool for obfuscation. Cf. Tenbarge v. Ames Taping Tool Systems, Inc., 190 F.3d 862, 865 (8th Cir. 1999) (noting that "discovery of expert opinion must not be allowed to degenerate into a game of evasion"). "It is disingenuous to argue that the duty to supplement under Rule 26(e)(1) can be used as a vehicle to disclose entirely new expert opinions after the deadline established by the court." Carter v. Finely Hospital, 2003 WL 22232844, *2 (N.D. Ill. 2003). See also Coles v. Perry, 217 F.R.D. 1, 4 (D.D.C. 2003) (noting that the objectives underlying Rule 26(a)(2)(B) would be nullified if an expert was permitted to "supplement" an earlier report by addressing a new matter after discovery has closed); Baker v. Indian Prairie Community Unit, School District No. 204, 1999 WL 988799, *3 (N.D. Ill. 1999) (refusing to allow plaintiffs to "ambush" defendants with new expert opinions after the expert opinion disclosure deadline and after they moved for summary judgment).
The Tenth Circuit has acknowledged that "on occasion it may be appropriate to permit the party using the expert to submit supplements to the report in response to assertions by opposing experts that there are gaps in the expert's chain of reasoning." Miller v. Pfizer, Inc., 356 F.3d at 1332 (observing that a court's failure to permit supplementation of an expert report could constitute an abuse of discretion by "unreasonably limiting the evidence" at trial). In Miller, however, the Tenth Circuit found that the trial court had not abused his discretion by refusing to allow the plaintiffs "to buttress the theory of their case by producing a new analysis by the retained expert of long-available data." Id. at 1334. In Wilson v. Sundstrand Corp., 2003 WL 22012673, *7 (N.D. Ill. 2003), the district court held that supplementation of an expert's report under Rule 26(e)(1) was not necessarily improper simply because the supplementation was prompted by a line of inquiry pursued by opposing counsel during the expert's deposition. However, in that case, plaintiffs' counsel produced the supplemental report prior to completion of the expert's deposition and prior to the deadline for serving defendants' rebuttal expert reports. Compare Confederated Tribes of Siletz Indians of Oregon v. Weyerhaeuser Co., 2003 WL 23715981, *2 (D. Or. 2003) (in denying a motion to strike supplemental expert opinions that were filed in response to defendant's motion for summary judgment, the court conceded that plaintiffs should have anticipated the need for this expert testimony and provided the expert report prior to the motion for summary judgment; however, declined to preclude plaintiffs from presenting those supplemental opinions once the deficiency was brought to their attention).
Weighing the particular circumstances in this case, it is difficult to bring Dr. Bardwell's Revised Report within the safe harbor of Rule 26(e)(1). A careful review reveals that Dr. Bardwell's Revised Report relies on the very same data cited in his Preliminary Report and in Dr. Farina's rebuttal. There is no indication that Dr. Bardwell's revised opinions drew on new data not available to the parties prior to September 19, 2005. During his deposition on October 26, 2005, Dr. Bardwell acknowledged that he was familiar with the criticisms expressed in Dr. Farina's rebuttal report, but testified that he had not corrected any of the "errors" identified by Dr. Farina, had not run any other statistical analysis using any other database, and had not changed any of the opinions or conclusions expressed in his Preliminary Report. Moreover, Dr. Bardwell testified on October 26, 2005, that his Preliminary Report accurately reflected all of the opinions and conclusions he had reached as a result of his statistical analysis in this case. See Exhibit A-4, at 9, attached to Defendants' Motion to Strike.
I find no indication that Dr. Bardwell's Revised Report was prompted by "additional or corrective information" that had not otherwise "been made known to the . . . parties during the discovery process." See Fed.R.Civ.P. 26(e)(1). Indeed, the stated rationale for Dr. Bardwell's Revised Report is open to question. Plaintiffs' counsel states that the Revised Report "was made, and could only be made, once [Dr. Bardwell] had the opportunity to review Defendants' expert's deposition testimony." Yet, Dr. Bardwell conceded during his October 26th deposition that he was aware of Dr. Farina's opinions but perceived no reason to change his opinions or conclusions, and had no reason to correct "errors" identified by Defendants' expert.
I conclude that Dr. Bardwell's Revised Report was not the result of new information, but rather represented a belated attempt to bolster and enhance his earlier report. Cf. Wechsler v. Hunt Health Systems, Ltd., 381 F. Supp.2d 135, 156 (S.D.N.Y. 2003) (finding that a supplemental expert report exceeded the boundaries of Rule 26(e) where portions of the supplemental report altered "a formerly complete analysis" and did not rely on new information acquired after submission of the original report); Beller ex rel. Bellerv. United States, 221 F.R.D. 689, 694-95 (D.N.M. 2003) (holding that Rule 26(e)(1) cannot be exploited by a party to "strengthen, broaden and deepen" an expert's earlier opinions, particularly where no new information was discovered subsequent to the preparation of the original report). A simple comparison illustrates this point. Dr. Bardwell's Preliminary Report contained approximately eight pages of text, two statistical tables and four specifically enumerated findings. The Revised Report consists of approximately 33 pages of text, ten statistical tables and 17 separate findings. The court concludes that Dr. Bardwell's Revised Report was not intended to "correct" his previous opinions, but rather represents an attempt to substantially rework his earlier report and to counter Defendants' Motion for Summary Judgment. Cf. Sharpe v. United States, 230 F.R.D. at 462-63. Based upon the foregoing analysis, I find that Dr. Bardwell's Revised Report was untimely under Rule 26(a)(2)(B) and does not qualify as a supplemental report under Rule 26(e)(1).
Having found a violation of Rule 26, the court must determine an appropriate sanction. Rule 37(c)(1) of the Federal Rules of Civil Procedure states that where a party fails to make a disclosure required by Rule 26(a) or Rule 26(e)(1), that party may not use at trial or on a motion any witness or information not so disclosed, unless the court determines that the failure to disclose was substantially justified or harmless. See Fed.R.Civ.P. 37(c)(1). The non-moving party has the burden of showing that they were substantially justified in failing to comply with Rule 26(a) or Rule 26(e). Nguyen v. IBP, Inc., 162 F.R.D. 675, 680 (D. Kan. 1995). While Rule 37(c)(1) is written in mandatory terms, it also vests the court with discretion to impose "other appropriate sanctions" in addition to or in lieu of an order striking witnesses or evidence not properly disclosed. See Woodworker's Supply, Inc. v. Principal Mutual Life Insurance Co., 170 F.3d 985, 993 (10th Cir. 1999) (recognizing that Rule 37(c) vests broad discretion with the trial court). See also Yeti by Molly, Ltd. v. Deckers Outdoor Corp., 259 F.3d 1101, 1106 (9th Cir. 2001) (noting that the district court's discretion should be given particularly wide latitude in imposing sanctions under Rule 37(c)(1)); Ortiz-Lopez v. Sociedad Espanola de Auxilio Mutuo Y Beneficiencia de Puerto Rico, 248 F.3d 29, 34 (1st Cir. 2001) (holding that district courts have broad discretion in meting out Rule 37(c) sanctions for Rule 26 violations).
For purposes of Rule 37(c)(1), a party's failure to disclose is substantially justified where the non-moving party has a reasonable basis in law and fact, and where there exists a genuine dispute concerning compliance. Nguyen v. IBP, Inc., 162 F.R.D. at 680. "Failure to comply with the mandate of the Rule is harmless when there is no prejudice to the party entitled to the disclosure." Id.
The determination of whether a Rule 26(a) violation is justified or harmless is entrusted to the broad discretion of the district court. A district court need not make explicit findings concerning the existence of a substantial justification or the harmlessness of a failure to disclose. Nevertheless, the following factors should guide its discretion: (1) the prejudice or surprise to the party against whom the testimony is offered; (2) the ability of the party to cure the prejudice; (3) the extent to which introducing such testimony would disrupt the trial; and (4) the moving party's bad faith or willfulness.Woodworker's Supply, Inc. v. Principal Mutual Life Insurance Co., 170 F.3d at 993
Applying the Woodworker's factors to the record before the court, the court cannot conclude that Plaintiffs' non-compliance was harmless. Cf. Lintz v. American General Finance, Inc., 1999 WL 619045, * 6 (D. Kan. 1999) (noting that non-compliance is harmless only when there is no prejudice to the opposing party). By requiring an expert to provide a "detailed and complete written report," Rule 26(a)(2)(B) sought to reduce the length of expert depositions or eliminate entirely the need for those depositions. See Indemnity Insurance Company of North America v. American Eurocopter, LLC., 227 F.R.D. 421, 425 (M.D.N.C. 2005) ("merely identifying a witness as an expert without requiring the identification of the opinions may render the depositions more difficult"). The cost considerations associated with expert depositions are particularly significant given Rule 26(b)(4)(B), which requires a party seeking discovery to pay the expert a reasonable fee for time spent responding to discovery. Dr. Bardwell's Revised Report substantially diminishes the value of his October 26th deposition and, almost certainly, will necessitate a second deposition if the Revised Report is not struck. Jones v. Thompson, 996 F.2d 261, 264 (10th Cir. 1993) (recognizing that delay and mounting costs can equate to prejudice).
On October 31, 2005, Defendants moved for summary judgment, arguing, in part, that Plaintiffs' disparate impact claims are not actionable under §§ 1983 and 1981 and because Plaintiffs could not isolate and identify a specific employment practice that had a statistically significant impact for a prima facie case. Given that line of attack, Defendants elected not to address Dr. Bardwell's statistical analysis. Although Dr. Bardwell apparently completed his Revised Report on November 14, 2005, it was not provided to Defendants until December 2, 2005. The same Revised Report was tendered to the District Court on December 9, 2005 as an attachment to Plaintiffs' Response to Defendants' Motion for Summary Judgment. In that Response, Plaintiffs argue that
Dr. Bardwell's analysis demonstrates that the way in which Defendants restructured their consolidation codes right before the layoff; the way they moved people about; the way they changed the bumping rules between the 2003 and the 1004 layoff, all conjoined to create a layoff strategy that reeks of racism. . . . But the importance and meaningfulness of statistical analysis is to be able to see through such surface clouds to penetrate to the actual final effect of how the system plays out. . . . In this case, the statistical significance of some of Dr. Bardwell's modeling arrive at numbers a quantum leap past mere evidence of discrimination. . . . Acknowledging, as one must, the good faith of Defendants' counsel, it appears they have been snookered by a failure to understand the power of properly modeled chi-square and multiple regression analysis. . . .See Plaintiffs' Response to Defendants' Motion for Summary Judgment, at 51 (emphasis added). It would be more accurate to say that Defendants were "snookered" into believing that Dr. Bardwell's initial report fully complied with Rule 26(a)(2)(B) and that Plaintiffs' expert would not be bolstering his analysis with additional opinions formulated after his deposition on October 26, 2005. Cf. Perkasie Industries Corp. v. Advance Transformer, Inc., 143 F.R.D. 73, 76-77 (E.D. Pa. 1992) (holding that a party is not permitted to postpone identifying the substance if its witnesses' testimony until a critical point in the proceedings at which it will become extremely burdensome for his opponent to prepare effectively to meet them).
The 2001 edition of Webster's Encyclopedic Unabridged Dictionary defines the verb form of snooker" as "to deceive, cheat or dupe."
As for the second and third Woodworker's factors, i.e., the ability to cure any prejudice and the potential disruption to a future trial, the court recognizes that a trial date has not been set in this case. That fact affords a greater ability to cure any prejudice to Defendants and minimizes the possibility of trial disruption. To suggest, however, that sanctions are not appropriate simply because the trial court can provide a further extension of time or delay the trial would effectively absolve Plaintiffs of any responsibility to comply with their discovery obligations. Rule 26(a) and (e) are designed to facilitate case management and the discovery process. Certainly, Plaintiffs' failure to comply with Rule 26(a) (20(B) and their untimely supplementation of Dr. Bardwell's analysis have disrupted those objectives.
Finally, with respect to Woodworker's fourth factor, the court cannot unequivocally conclude that Plaintiffs' expert disclosure violations were the product of bad faith or willfulness. The unexplained delay of 18 days in providing Defendants with Dr. Bardwell's Revised Report is troubling. However, Plaintiffs' lack of bad faith alone would not be enough to overcome the other three factors. Cf. Jacobsen v. Deseret Book Co., 287 F.3d 936, 955 (10th Cir. 2002). Plaintiffs should not be permitted to ignore their disclosure obligations and then avoid sanctions simply by claiming their violation was not willful.
Admittedly, Rule 37(c)(1) is written in mandatory terms, and indicates that a party who fails to comply with Rule 26(a) or Rule 26(e)(1) "is not permitted" to use at trial or on a motion any information not so disclosed. However, the court is also mindful that exclusion of evidence is an drastic sanction. The Tenth Circuit has suggested that such a sanction should be reserved for those cases where an expert's name and written report were withheld completely. Summers v. Missouri Pacific Railroad System, 132 F.3d 599, 604 n. 7 (10th Cir. 1997). Compare Moss v. Feldmeyer, 979 F.2d 1454, 1459 (10th Cir. 1992) (permitted defendant to amend the final pretrial order to add one expert witness who had been designated more than two weeks before trial and to expand the scope of another expert's testimony on the morning of trial; noted that plaintiff had received a summary of their expert reports prior to their trial testimony and that both experts were available for depositions before trial).
Consistent with the primary goal of Rule 37 sanctions, which is to deter misconduct, "the sanction imposed should be the least severe of those available, which appears adequate to deter and punish the wrongdoer." Hite v. The PQ Corp., 1998 WL 895893, *2 (D. Kan. 1998). Defendants' motion requests an order striking Dr. Bardwell's Revised Report and the opinions expressed therein from Plaintiffs' Response to Defendants' Motion for Summary Judgment, as well as an award of the reasonable expenses incurred in bringing the Motion to Strike. After carefully considering the Woodworker's factors and the objectives underlying Rule 37(c), I decline to strike Dr. Bardwell's Revised Report. See Scott v. IBM Corp., 196 F.R.D. 233, 247 n. 9 (D.N.J. 2000) (noting that exclusion of evidence under Rule 37(c) is not appropriate unless the resulting prejudice cannot be cured); Comuso v. National Railroad Passenger Corp., 1998 WL 800342 *2 (E.D. Pa. 1998) (holding that the importance of the excluded testimony is also an important final consideration in considering Rule 37(c) sanctions).
While I can appreciate that Defendants were surprised upon receiving Dr. Bardwell's Revised Report, their reply brief in support of summary judgment addresses Dr. Bardwell's statistical analyses at some length. See Defendants' Reply in Support of Their Motion for Summary Judgment, at 52-56. After attacking Dr. Bardwell's methodology and statistical results, Defendants conclude that "no matter which of Dr. Bardwell's reports Plaintiffs' try to rely upon, Bardwell's failure to show disparate treatment between similarly situated employees . . . fails to eliminate nondiscriminatory explanations for any of his alleged statistical disparities." Id. at 56. Cf. Summers v. Missouri Pacific Railroad System, 132 F.3d at 605 (in allowing plaintiffs' belated expert disclosure, suggested that defendant "could hardly have been surprised" as evidenced by the fact that defendant was already prepared to respond to the newly disclosed opinions). "The purpose of the federal rules is to provide the maximum opportunity for each claim to be decided on the merits rather than on the basis of procedural factors." State Distributors, Inc. v. Glenmore Distilleries Co., 738 F.2d 405, 411 (10th Cir. 1984). Any substantive flaws in the analyses provided by Dr. Bardwell or Dr. Farina should be resolved by the District Court or the trier of fact. Under the particular circumstances of this case, I do not find that the interests of justice would be served by striking Dr. Bardwell's Revised Report.
It also bears noting that Defendants' Motion for Summary Judgment chose to attack Plaintiffs' disparate impact claim on legal grounds, and essentially dismissed Dr. Bardwell's statistical analysis as irrelevant. The court presumes that Defendants would have made the same legal arguments even if they had received proper notice of Dr. Bardwell's revised opinions.
Accordingly, for the foregoing reasons, I will deny in part, and grant in part, Defendants' Motion to Strike Plaintiffs' Untimely Expert Report (Document # 56), filed on December 21, 2005. While the court will not strike Dr. Bardwell's Revised Report, I cannot condone Plaintiffs' failure to comply with Rule 26 and established pretrial deadlines. Defendants were entitled to presume that they were deposing Dr. Bardwell on all the opinions he intended to offer at trial. Obviously, that is not the testimony they obtained on October 26th. Therefore, the court will impose the following monetary sanctions. Plaintiffs Freeman, McClenton, James and Dennis must reimburse Defendants for any fees they paid to Dr. Bardwell in connection with his October 26th deposition. The aforementioned Plaintiffs must also reimburse Defendants for any court report fees and transcript costs incurred by Defendants in connection with that deposition. I will also require Plaintiffs to reimburse Defendants for the reasonable attorneys fees incurred in connection with Dr. Bardwell's deposition, up to a maximum of $750. Cf. United States v. Big D Enterprises, Inc., 184 F.3d 924, 936 (8th Cir. 1999) (holding that a government lawyer may recover attorneys fees imposed as a sanction in connection with a discovery dispute). Finally, the court will require Plaintiffs Freeman, McClenton, James and Dennis to pay the reasonable fees and costs incurred by Defendants in connection with the pending Motion to Strike, including the reasonable fees incurred by Defendants' counsel to appear at the hearing on February 13, 2006. I will permit Defendants to re-depose Dr. Bardwell on the opinions and analysis presented in his Revised Opinion, however, that deposition will be conducted at Defendants' expenses consistent with the Federal Rules of Civil Procedure. The court concludes that no other sanction is appropriate or necessary to achieve the objectives underlying Rules 16, 26 and 37.
Defendants should file with the court an appropriate bill of costs within two weeks of the date of this Memorandum Order. If Plaintiffs Freeman, McClenton, James and Dennis file an objection to this Memorandum Order pursuant to Fed.R.Civ.P. 72(a), the court will suspend the payments of sanctions until that objection is ruled upon by the District Court.
APPENDIX I "Summary of Findings" Presented in the Bardwell "Preliminary Report"1. African American employees were laid off at more than two and one-half the rate of other employees.
2. The race disparity in layoffs for African American employees is statistically significant at the 0.032 level, exceeding the common threshold of prima facie evidence.
3. After controlling for the other factors, the likelihood of being laid off for African American employee was 377 percent the likelihood for other comparable employees. In other words, African American employees were almost four times as likely to be laid off as similarly situated employees who were not African American.
4. The probability that we would see so many African Americans laid off in an race-neutral situation is 0.0083, or 83 in 10,000. If race were not a significant factor in layoffs, we would expect that controlling for other factors would reduce the race effect to statistical insignificance. In contrast, controlling for the effects of other factors leaves a highly-significant 377 percent impact for African American employees.
APPENDIX II "Summary of Findings" in the Bardwell "Revised Report"1 Revised chi-square analysis of the entire DHHS workforce using Dr. Farina's database confirms a statistically significant disparity in layoffs at DHHS adversely affecting African American employees. African American employees were laid off at two and one-half the rate of other employees. This disparity is statistically significant at the 0.03 level. exceeding the common threshold for prima facie evidence.
2. The chi-square analysis for all employees at DHHS except those who were classified as bilingual confirms a similar race disparity in layoffs.
3 Of the 76 Job Codes in the data provided by the defendant, only 13 of the Job Codes experienced any layoffs. In other words, 100 percent of the employees were retained in the other 63, or 82 percent of the Job Codes. This extreme variation in layoff rates in Job Codes between 0 and 100 percent is of statistical and practical significance.
4. Chi-square analysis of layoffs among employees in the 13 Job Codes where some employees were laid off demonstrates a more extreme level of race disparity in layoffs than in the workforce as a whole. The results of the chi-square analysis for this subset of employees is statistically significant at the 0.02 level.
5. Tabulation of layoffs within specific Job Codes shows that 100 percent of the African American employees in five Job Codes were laid off, while no other employees were laid off in four of those Job Codes, and a significantly lower percentage of other employees were laid off in the remaining Job Code.
6. Chi-square analysis of the 320 employees in the Job Codes which are informative about race disparity in layoffs, reveals a much greater disparity practically and statistically in layoffs between African American and other employees. In these Job Codes the impact on African American employees was 469 percent as great as the impact on non-African American employees. This disparity is statistically significant at the 0.002 level, far exceeding that required for prima facie evidence.
7. Dr. Fariña claims that in this case defendants made no relevant distinction when they coded some employees with the word "Layoff" while other employees were coded "Retirement" or "Resignation". In fact, in a recent case in which Dr. Fariña and I both served as experts, Dr. Fariña testified for the plaintiffs, that involuntary separations should be distinguished from other separations based on defendant's coding, and he based all his analyses on involuntary separations.
8. Federal regulations and procedures which specify the way a chi-square analysis is to be applied in employment discrimination cases in the Title VII framework, clearly state that only a 2×2 chi-square analysis should be employed.
9. Dr. Fariña's 2×4 chi-square analysis will frequently result in false positive and false negative findings of discrimination. In other words, many situations where employees have been discriminated against and the level of discrimination would result in a statistically significant finding using the accepted 2×2 analysis, the 2×4 analysis would find no discrimination. Conversely, a situation where the accepted 2×2 analysis would not result in a finding of statistically significant race disparities, the 2×4 analysis insisted on by Dr. Fariña, could result in a statistically significant finding of evidence of discrimination.
10. In a recent case in which Dr. Fariña and I both served as experts, Dr. Farina testified for the plaintiffs, and in that case used 2×4 chi-square analysis that could generate false positives. In my report in that case, I provide examples of false findings of discrimination using Dr. Fariña's 2×4 method, when the employment decision being analyzed was perfectly neutral.
11. Dr. Fariña repeatedly touts his flawed 2×4 methodology as being "race neutral," and by contrast, attempts to impugn the accepted 2×2 analysis as being not "race neutral." This misuse of language is an attempt to associate a phrase with a positive connotation with his unsupportable methodology. In fact, it is nonsensical to assume that an analysis of discrimination against a particular racial group should be "race neutral."
12. Dr. Fariña claims that my variable representing length of service is inappropriately coded, and then withdraws this criticism in his deposition.
13. Dr. Fariña make an erroneous and irrelevant claim that logistic regression models must contain a continuous variable, and then significantly retracts his claim in his deposition.
14. Dr. Fariña repeatedly complains in his report that I did not document how I encoded the job category variable, and then retracts this criticism in his deposition.
15. Dr. Fariña complains about the complexity of the 76 Job Codes and uses this as an excuse for not including a Job Code variable in his model. Excluding this critical variable invalidates his results. Citations and authorities are provided which state that exclusion of a variable known to be strongly related to the dependent variable severely biases the results and renders them invalid.
16. Dr. Fariña's variable selection method is based on sophistic model building methodology which statistical authorities specifically caution against.
17. Dr. Fariña's arguably most incomprehensible claim is that he is going to analyze the affect of race on layoffs without including race as a variable in his logistic regression analysis. Dr. Fariña's approach — which not only fails to include one of the most crucial variables, Job Code, but also unceremoniously excludes the variable of interest, race — is a bizarre fabrication that is irrelevant to the issue at hand.