Opinion
No. 99 Civ. 9568 (LAK).
January 31, 2006
Philip Raible James McKenney MINTZ GOLD LLP, Attorneys for Plaintiff Reuben Malave
Jason L. Solataroff Catherine E. Anderson GISKAN SOLATAROFF Attorneys for Plaintiff Manuel Gomez
Joshua H. Epstein Shari B. Finkelstein DREIER, LLP, Attorneys for Plaintiff Wilfred Maldonado
Barry Lax THE LAX LAW FIRM Attorney for Plaintiff Parnell Peterson
James M. Lemonedes William S.J. Fraenkel Assistant Corporation Counsel MICHAEL A. CARDOZO CORPORATION COUNSEL OF THE CITY OF NEW YORK Attorneys for Defendants
MEMORANDUM OPINION
This action began in 1999 as a putative class action on behalf of between 14,000 and 17,000 black and Hispanic uniformed members of the New York City Police Department ("NYPD"). The core of the action, if not its exclusive focus, was a claim that the disciplinary processes of the NYPD treated black and Hispanic officers differently than others by, among other things, charging them with infractions more often than white officers and meting out harsher punishments to them than to whites for comparable infractions.
The class action proceeded under the "pattern or practice" theory set out in International Brotherhood of Teamsters v. United States. After certification of a class, and on the eve of what promised to be a very lengthy trial, the case settled. There now remain individual claims by four plaintiffs who opted out of the class settlement. The matter is before the Court on defendants' motion in limine to exclude statistical, expert and anecdotal evidence that plaintiffs propose to offer at trial.
431 U.S. 324, 342, 360-61 (1977).
Latino Officers Ass'n City of New York, Inc. v. City of New York, No. 99-9568, 2003 WL 22300158 (S.D.N.Y. Oct. 8, 2003).
The remaining plaintiffs are Messrs. Gomez, Malave, Maldonado, Sanchez and Peterson. The Court has been advised, however, that Mr. Sanchez is in the process of withdrawing his individual claim in order to submit it for adjudication by the special master set up under the settlement. This order therefore addresses only the claims of the other four.
The nub of the present dispute concerns the parties differing perspectives concerning the impact of the settlement of the class action. Plaintiffs seek to offer all of the statistical and expert evidence and a substantial number of anecdotal witnesses in support of the four remaining claims that would have been offered in the class action pattern or practice trial. Defendants argue that the expert and statistical evidence should be excluded as irrelevant to what remains to be tried, and unreliable under Rule 403. They seek also to limit the number of anecdotal witnesses.
I
In a pattern or practice class action, the initial focus is in proving an agency-wide intentional discriminatory practice or policy, not individual incidents of discrimination. In consequence, "statistical evidence constitutes the core of plaintiffs' prima facie case." Once plaintiffs in such a case succeed in proving the alleged discriminatory practice or policy, the class obtains injunctive relief, and individual class members who seek individualized relief enjoy a presumption that employment actions of which they complain were affected, thus shifting the burden of proof to the defense.
Robinson v. Metro North Commuter R.R. Co., 267 F.3d 147, 158 n. 5 (2d Cir. 2001), cert. denied, 535 U.S. 951 (2002).
Latino Officers, 2003 WL 22300158, at *3.
These four plaintiffs specifically have disavowed any reliance on a pattern or practice theory. With the settlement of the class action, the role of statistical evidence changed. The issues in the cases of the four remaining plaintiffs concern what happened to them. While they are not precluded from introducing competent and probative statistical evidence, such evidence runs a risk of being collateral. It therefore is necessary to scrutinize both their claims and the proposed evidence.
Tr., Jan. 14, 2005, at 14.
See, e.g., Abrams v. Lightolier, Inc., 50 F.3d 1204, 1217 (3d Cir. 1995).
See, e.g., Gilty v. Oak Park, 919 F.2d 1247, 1252 (7th Cir. 1990).
A. Claims of the Remaining Plaintiffs
Plaintiff Gomez contends that he was compelled to resign from the police academy in 1998 based on inaccurate information, that he thereafter filed an EEOC complaint challenging his resignation, that he was rehired, and that he subsequently has been subjected to retaliatory investigations, interrogations and harassment at the 43rd precinct in the Bronx. He claims that the NYPD maintains a work environment hostile to Latino and African-American officers ("HWE"). He somewhat recently has been the subject of disciplinary charges and a departmental trial, although no decision yet has been reached.
Plaintiff Malave, formerly of the 45th precinct in the Bronx, was terminated by the NYPD in 1998 after he was found guilty of departmental charges involving his being drunk and soliciting an undercover officer posing as a prostitute. The Court previously held that he is precluded from relitigating the factual findings underlying the termination decision. He claims, however, that he was treated more harshly than white officers with similar or worse disciplinary histories and charges. He too makes an HWE claim.
Latino Officers Ass'n v. City of New York, 253 F. Supp.2d 771, 786 (S.D.N.Y. 2003).
Plaintiff Maldonado claims that he was terminated by the NYPD for improper fraternization while he was assigned to the Police Academy and for giving testimony in an unemployment insurance proceeding while serving in the 32nd precinct in Manhattan without obtaining the required permission of the Department. He too complains that the NYPD has an HWE and that he was treated more harshly than white officers in comparable circumstances.
Plaintiff Parnell Peterson claims that he was subjected to disparate disciplinary treatment when he was given departmental charges for, among other things, failing to conduct an investigation, to prepare required reports, and to make required notifications in connection with his response to a domestic disturbance call in 1995. He forfeited 30 vacation days, but claims that white officers were not disciplined for similar infractions. He claims also that he has been retaliated against and experienced an HWE at the 47th precinct in the Bronx.
B. The Proposed Statistical Evidence
Plaintiffs propose to call two statistical experts. Their proposed testimony, in summary, would be as follows:
1. Richard Faust
Mr. Faust proposes to testify to the following conclusions:
(a) Analysis of NYPD's Computer Analysis and Tracking System ("CATS") data indicates that:
(1) Latino and African-American officers were respectively 50 percent and 70 percent more likely than white officers to receive formal charges and specifications during the period 1995-2000.
(2) Latino and African-American officers were substantially more likely to have been found guilty on departmental charges than white officers.
(3) Latino and African-American officers found guilty of departmental charges were more likely to have been terminated or placed on probation than white officers.
(4) In general, disparities in the treatment of Latino and African-American officers as compared to white officers at each stage of the disciplinary process support the inference that race and ethnicity were motivating factors in the decisions.
(b) A review of the NYPD 1998 Report of the Disciplinary Task Force data shows substantial racial and ethnic disparities in the bringing of charges, their dispositions, and the penalties imposed. Mr. Faust disagrees with the authors' conclusion that no inference of discrimination was warranted by the data. He would have recommended more comprehensive statistical and in-depth case studies to see if white and minority NYPD members were treated differently in the disciplinary system.
(c) Black officers in the 26th precinct were subjected to command discipline substantially more often than white officers in that precinct during the years 1995-1999.
(d) Defendants' expert, Dr. Ali Saad, reduced racial disparities by the use of use of incomplete data and several data analysis strategies.2. Mark Killingsworth
(a) Higher proportions of Latino and African-American officers were subjects of disciplinary charges than of white officers, contrary to the conclusion of defendants' expert, Dr. Saad.
(b) Dr. Saad's hypothesis that any disparities in the frequency with which Latinos and African-Americans were charged with misconduct, as compared to whites, may be explained by differences in behavior is unsupported and his analysis flawed.
(c) Professor Killingsworth reaches similar conclusions with respect to racial and ethnic differences in the disposition of cases and charges and with respect to penalties imposed.
(d) Professor Killingsworth reiterates these conclusions and adds further criticism of Dr. Saad's work in a rebuttal report.C. Analysis
Revised first report of plaintiff's expert Richard Faust ( "Faust I") ¶ 10.
Id. ¶¶ 15-16.
Id. ¶¶ 19-20.
Id. ¶ 21.
Second report and affidavit of plaintiff expert Richard Faust ( "Faust II") ¶¶ 10, 13.
Third report and affidavit of plaintiff expert Richard Faust ( "Faust III") ¶ 15.
Fourth report of plaintiff's expert Richard Faust ( "Faust IV").
Report and declaration of plaintiffs' rebuttal expert ( "Killingsworth I") ¶¶ 9-23.
Id. ¶¶ 24-31.
Id. ¶¶ 32-63.
Report and declaration of plaintiffs' rebuttal expert Dr. Mark R. Killingsworth in response to report of defendants' expert Dr. Ali Saad ( "Killingsworth II").
The essence of defendants' position is that plaintiffs' statistical evidence was prepared for and relevant to a pattern-and-practice case that no longer is to be tried. It is at best collateral to the specific issues in the four particular cases remaining to be tried. Further, they contend that its unfair prejudicial effect would outweigh any probative value and that the proposed testimony is not reliable.
1. Relevance
The fact that this no longer is a pattern-and-practice case is not dispositive of the issue as to statistical evidence, as plaintiffs are permitted to introduce competent and probative statistical evidence even in individual disparate treatment cases. While there may be other valid objections to some or all of this evidence, the fact that this no longer is a pattern or practice case is not controlling. Each of the individual plaintiffs claims that race or ethnicity played a part in the fact that he was charged with misconduct and in the outcome of his disciplinary proceeding. The Court cannot say that statistical evidence tending to show that there were racial and ethnic disparities in those areas in the relevant time periods is not relevant to their cases. Indeed, in principle, appropriate statistical evidence could be quite probative of whether an adverse employment action occurred in circumstances giving rise to an inference of discrimination, which is an element of the prima facie case required by McDonnell Douglas. 2. Reliability
E.g., Lowery v. Circuit City Stores, Inc., 158 F.3d 742, 761 (4th Cir. 1998), vacated on other grounds, 527 U.S. 1031 (1999).
Defendants claim that the testimony of the plaintiffs' experts is unreliable because there is an insufficient "fit" between the department-wide statistics and the issues in each of the individual cases, in which the question is what happened to each plaintiff, and because the data upon which the experts' analyses rest are biased because they are based only upon records of formal disciplinary procedures and exclude other forms of discipline.
There arguably is some issue as to "fit," but the Court is persuaded that it is insufficient to warrant exclusion of the evidence. Likewise, defendants' attempt to show selection bias is unpersuasive. A pattern of disparate treatment in the charging, determinations of guilt, and penalties as between minority and white officers, if plaintiffs can establish that, would seem to be the appropriate comparison in the cases of individuals who were subjected to the formal disciplinary process.
One exception is Mr. Faust's proposed evidence about command discipline at the 26th precinct. As none of the remaining plaintiffs served there, this evidence will be excluded.
3. Rule 403
The most substantial of defendants' objections to the statistical evidence is that its unfair prejudicial effect would outweigh substantially its probative value and lead to confusion and waste of time.
Turning first to the question of alleged unfair prejudice, the ultimate question in each of the four remaining cases is whether race or ethnicity was a factor in whatever happened to the plaintiff, quite apart from anything statistics may show as to whether they appear to be factors in the department in general. By focusing the jury on department-wide data, defendants argue, a risk would be created that the jury would be led to decide these cases on the basis of their views about the department in general rather than on the basis of the facts of the individual cases.
It would be unfair to the defendants if the jury or juries gave undue weight to department-wide data to the detriment of proper attention to the facts of the individual cases. But the Court is capable of instructing the jury as to the proper role of the statistical data, and it must assume that the jury would follow such instructions. While a risk of unfair prejudice would remain, the Court cannot say that it would substantially outweigh the probative value of the evidence.
Nor would it be appropriate to exclude this evidence on the ground that it would cause confusion and waste time. To be sure, few jurors (or judges) look forward to hours or days of exceedingly technical examination concerning sampling, error rates, epistemological correlation, regression analysis, and other arcana of statistical methods. The information before the Court, however, does not now permit a conclusion that the jury could not keep such evidence straight or that the time devoted to it would be wasted.
Plaintiffs propose also to call sixteen identified witnesses, as well as additional NYPD witnesses they wish to name at a later time, in order to give anecdotal evidence in support of their claims. Defendants object that this is too much, especially given that only three would give testimony concerning the facts of the individual claims of any of the four remaining plaintiffs.
Too much remains to be done before this issue can be resolved in a confident manner. No determination yet has been made as to whether the cases of the four individuals will be tried together or separately and as to whether the Monell issues will be tried separately from the issue of discrimination vel non. These both are facts that likely would bear substantially on whether to permit all of the anecdotal witnesses to testify. Accordingly, this aspect of the motion will be denied without prejudice to renewal.
Defendants' motion in limine [docket item 216] is granted to the extent that Mr. Faust's testimony as to command discipline in the 26th precinct is excluded and is denied in all other respects, albeit without prejudice to renewal as to the anecdotal witness issue as indicated above. The parties are directed to submit, on or before February 21, 2006, either a joint proposal as to whether and to what extent they wish a joint or separate trials or, if they cannot agree, their respective proposals.
To the extent the proposed evidence is of a rebuttal nature, it of course would be admissible, if at all, only in response to appropriate evidence from the defendants.
SO ORDERED.