UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK
JONATHAN NNEBE, et al.,
Plaintiffs, No. 06-cv-4991 (RJS) -v- ORDER MATTHEW DAUS, et al.,
Defendants.
RICHARD J. SULLIVAN, Circuit Judge: Before the Court is Defendants’ motion in limine to preclude Plaintiffs’ experts – Michael Rempel and Melissa S. Ader – from testifying pursuant to Federal Rule of Evidence 702. (Doc. No. 610 (“Defs.’ MIL”)). For the reasons stated below, Defendants’ motion is GRANTED. To offer expert testimony under Rule 702, a witness must be “qualified as an expert by knowledge, skill, experience, training, or education.” Fed. R. Evid. 702. Even then, the expert’s opinion may be admitted only if it (1) “help[s] the trier of fact to understand the evidence or to determine a fact in issue,” (2) “is based on sufficient facts or data,” (3) “is the product of reliable principles and methods,” and (4) reflects a reliable application of “the principles and methods to the facts of the case.” Id. “Thus, in assessing admissibility, the trial court must determine whether the proffered expert testimony is relevant, i.e., whether it ‘has any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence,’ and whether the proffered testimony has a sufficiently ‘reliable foundation’ to permit it to be considered.” Campbell ex rel. Campbell v. Metro. Prop. & Cas. Ins. Co., 239 F.3d 179, 184 (2d Cir. 2001) (alterations omitted) (quoting Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 587 (1993)). Plaintiffs’ first expert, Michael Rempel, is a criminologist and university professor who has researched and developed risk-assessment tools – typically used in connection with bail and parole proceedings – that “aid judges and other decision-makers in understanding the likelihood that someone charged with a crime will re-offend.” (Doc. No. 606-1 (“Rempel Report”) at 3–5.)
Rempel seeks to opine that taxi drivers as a group pose an “overwhelmingly low risk” of reoffending after they are arrested. (Rempel Report at 2.) According to Rempel, this conclusion is supported by studies in New York and elsewhere, which show that (1) the “nature and severity” of an individual’s charge has only a “weak relationship” to recidivism, (2) individuals with multiple recent convictions have the highest likelihood of reoffending, (3) individuals under twenty-five years of age are far more likely to reoffend than individuals older than twenty-five, and (4) persons who are unemployed are more likely to reoffend than those who have jobs. (Rempel Report at 5–6, 8.) But Rempel’s opinion about the likelihood of recidivism among taxi drivers as a group is not relevant and will not “help the trier of fact to understand the evidence or to determine a fact in
issue.” Fed. R. Evid. 702(a). In this trial, the jury will determine (among other things) (1) whether each of the twenty individual drivers would have requested a hearing before the New York Office of Administrative Trials and Hearings (“OATH”), and whether each of those drivers would have persuaded an OATH ALJ that he was not a danger, thereby securing reinstatement of his license. Given those facts in issue, Rempel’s opinion about the risk posed by drivers generally can be relevant only if OATH ALJs would have based their individual reinstatement decisions on factors such as those identified in Rempel’s report. Indeed, if OATH ALJs would not have relied on empirical data like Rempel’s in their decisions, then Rempel’s opinion could not help the jury determine whether an OATH ALJ would have reinstated any of the twenty drivers’ licenses. And on that front, nothing suggests that OATH ALJs would have employed anything like Rempel’s empirical analysis. For starters, neither Rempel nor Plaintiffs have shown that OATH ALJs routinely use risk-assessment tools or “empirically based factors” like the ones relied on by Rempel to make licensing decisions. (See Rempel Report at 1.) And while Plaintiffs contend that
“nothing in OATH’s rules” would “preclude[]” a driver from “presenting statistical or generalized evidence” of this sort (Doc. No. 615 (“Pls.’ Opp. to MIL”) at 4), that is not the same as showing that ALJs actually do – or at least would – factor such evidence into their decisions.1 The closest Plaintiffs come to bridging that gap is Rempel’s unsubstantiated conclusion that, because drivers are now prevailing before OATH at higher rates than under the old regime, it must mean that “empirical data and present [OATH] decision-making tendencies are converging.” (Rempel Report at 3.) But Rempel offers no evidence to suggest that this “converg[ence]” is because OATH ALJs are now looking to empirical data. Tellingly, while Rempel says that he reviewed the last twenty OATH decisions, he does not point to a single decision in which the ALJ relied on those statistics. Moreover, Rempel fails to explain how such a small and unrepresentative
sample size – consisting of only the twenty most recent decisions rather than a random sampling of decisions since OATH adopted the new protocols in 2019 – could constitute “sufficient facts or data” on which an expert can form an opinion under Rule 702. See Chen-Oster v. Goldman, Sachs & Co., 114 F. Supp. 3d 110, 124 (S.D.N.Y. 2015) (excluding expert opinion as unreliable because the “sample size” was “too small” and potentially “subject to bias”). Rempel’s opinion is even less relevant because he fails to tailor his analysis to the twenty individual drivers who are the subject of the upcoming trial. Generally, when a party seeks to
1 For the same reason, Rempel’s report and testimony would not “assist the jury [in] understand[ing] the evidence,” (Pls.’ Opp. to MIL at 9), since the jury does not need to know the empirical statistics on an individual’s likelihood of recidivism where there is no indication that OATH ALJs rely on (or would have relied on) such data. prove his individual case through evidence about what happened to other persons, that evidence will be relevant only if those other persons are “similarly situated” to the proponent. When seeking to prove selective prosecution, for instance, a defendant must offer statistics about “similarly situated defendants” charged with similar crimes, rather than “raw statistics regarding overall
charges.” United States v. Bass, 536 U.S. 862, 864 (2002). Likewise, plaintiffs can show employment discrimination by showing how “similarly situated” employees were treated. E.E.O.C. v. Bloomberg L.P., No. 07-cv-8383, 2010 WL 3466370, at *12 (S.D.N.Y. Aug. 31, 2010) (excluding expert’s statistical report “because he did not compare class members to other similarly situated employees”); see also Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442, 458 (2016) (citing Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011)) (explaining that plaintiffs could not rely on evidence about how “other employees were discriminated against” unless those employees were “similarly situated” to plaintiffs).
Free access — add to your briefcase to read the full text and ask questions with AI
UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK
JONATHAN NNEBE, et al.,
Plaintiffs, No. 06-cv-4991 (RJS) -v- ORDER MATTHEW DAUS, et al.,
Defendants.
RICHARD J. SULLIVAN, Circuit Judge: Before the Court is Defendants’ motion in limine to preclude Plaintiffs’ experts – Michael Rempel and Melissa S. Ader – from testifying pursuant to Federal Rule of Evidence 702. (Doc. No. 610 (“Defs.’ MIL”)). For the reasons stated below, Defendants’ motion is GRANTED. To offer expert testimony under Rule 702, a witness must be “qualified as an expert by knowledge, skill, experience, training, or education.” Fed. R. Evid. 702. Even then, the expert’s opinion may be admitted only if it (1) “help[s] the trier of fact to understand the evidence or to determine a fact in issue,” (2) “is based on sufficient facts or data,” (3) “is the product of reliable principles and methods,” and (4) reflects a reliable application of “the principles and methods to the facts of the case.” Id. “Thus, in assessing admissibility, the trial court must determine whether the proffered expert testimony is relevant, i.e., whether it ‘has any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence,’ and whether the proffered testimony has a sufficiently ‘reliable foundation’ to permit it to be considered.” Campbell ex rel. Campbell v. Metro. Prop. & Cas. Ins. Co., 239 F.3d 179, 184 (2d Cir. 2001) (alterations omitted) (quoting Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 587 (1993)). Plaintiffs’ first expert, Michael Rempel, is a criminologist and university professor who has researched and developed risk-assessment tools – typically used in connection with bail and parole proceedings – that “aid judges and other decision-makers in understanding the likelihood that someone charged with a crime will re-offend.” (Doc. No. 606-1 (“Rempel Report”) at 3–5.)
Rempel seeks to opine that taxi drivers as a group pose an “overwhelmingly low risk” of reoffending after they are arrested. (Rempel Report at 2.) According to Rempel, this conclusion is supported by studies in New York and elsewhere, which show that (1) the “nature and severity” of an individual’s charge has only a “weak relationship” to recidivism, (2) individuals with multiple recent convictions have the highest likelihood of reoffending, (3) individuals under twenty-five years of age are far more likely to reoffend than individuals older than twenty-five, and (4) persons who are unemployed are more likely to reoffend than those who have jobs. (Rempel Report at 5–6, 8.) But Rempel’s opinion about the likelihood of recidivism among taxi drivers as a group is not relevant and will not “help the trier of fact to understand the evidence or to determine a fact in
issue.” Fed. R. Evid. 702(a). In this trial, the jury will determine (among other things) (1) whether each of the twenty individual drivers would have requested a hearing before the New York Office of Administrative Trials and Hearings (“OATH”), and whether each of those drivers would have persuaded an OATH ALJ that he was not a danger, thereby securing reinstatement of his license. Given those facts in issue, Rempel’s opinion about the risk posed by drivers generally can be relevant only if OATH ALJs would have based their individual reinstatement decisions on factors such as those identified in Rempel’s report. Indeed, if OATH ALJs would not have relied on empirical data like Rempel’s in their decisions, then Rempel’s opinion could not help the jury determine whether an OATH ALJ would have reinstated any of the twenty drivers’ licenses. And on that front, nothing suggests that OATH ALJs would have employed anything like Rempel’s empirical analysis. For starters, neither Rempel nor Plaintiffs have shown that OATH ALJs routinely use risk-assessment tools or “empirically based factors” like the ones relied on by Rempel to make licensing decisions. (See Rempel Report at 1.) And while Plaintiffs contend that
“nothing in OATH’s rules” would “preclude[]” a driver from “presenting statistical or generalized evidence” of this sort (Doc. No. 615 (“Pls.’ Opp. to MIL”) at 4), that is not the same as showing that ALJs actually do – or at least would – factor such evidence into their decisions.1 The closest Plaintiffs come to bridging that gap is Rempel’s unsubstantiated conclusion that, because drivers are now prevailing before OATH at higher rates than under the old regime, it must mean that “empirical data and present [OATH] decision-making tendencies are converging.” (Rempel Report at 3.) But Rempel offers no evidence to suggest that this “converg[ence]” is because OATH ALJs are now looking to empirical data. Tellingly, while Rempel says that he reviewed the last twenty OATH decisions, he does not point to a single decision in which the ALJ relied on those statistics. Moreover, Rempel fails to explain how such a small and unrepresentative
sample size – consisting of only the twenty most recent decisions rather than a random sampling of decisions since OATH adopted the new protocols in 2019 – could constitute “sufficient facts or data” on which an expert can form an opinion under Rule 702. See Chen-Oster v. Goldman, Sachs & Co., 114 F. Supp. 3d 110, 124 (S.D.N.Y. 2015) (excluding expert opinion as unreliable because the “sample size” was “too small” and potentially “subject to bias”). Rempel’s opinion is even less relevant because he fails to tailor his analysis to the twenty individual drivers who are the subject of the upcoming trial. Generally, when a party seeks to
1 For the same reason, Rempel’s report and testimony would not “assist the jury [in] understand[ing] the evidence,” (Pls.’ Opp. to MIL at 9), since the jury does not need to know the empirical statistics on an individual’s likelihood of recidivism where there is no indication that OATH ALJs rely on (or would have relied on) such data. prove his individual case through evidence about what happened to other persons, that evidence will be relevant only if those other persons are “similarly situated” to the proponent. When seeking to prove selective prosecution, for instance, a defendant must offer statistics about “similarly situated defendants” charged with similar crimes, rather than “raw statistics regarding overall
charges.” United States v. Bass, 536 U.S. 862, 864 (2002). Likewise, plaintiffs can show employment discrimination by showing how “similarly situated” employees were treated. E.E.O.C. v. Bloomberg L.P., No. 07-cv-8383, 2010 WL 3466370, at *12 (S.D.N.Y. Aug. 31, 2010) (excluding expert’s statistical report “because he did not compare class members to other similarly situated employees”); see also Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442, 458 (2016) (citing Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011)) (explaining that plaintiffs could not rely on evidence about how “other employees were discriminated against” unless those employees were “similarly situated” to plaintiffs). Here, Rempel offers only “raw statistics regarding [taxi drivers] overall,” Bass, 536 U.S. at 864, and makes no attempt to show how “similarly situated” drivers – e.g., those facing similar charges and/or with comparable criminal histories – have fared at their
hearings under the new protocols that OATH adopted in 2019. Without such refinement, Rempel’s overbroad opinions about taxi drivers as a group “cannot assist the trier of fact” and are inadmissible under Rule 702.2 The report authored by Plaintiffs’ second expert, Melissa Ader, likewise fails to meet the requirements of Rule 702. Ader is an attorney at an organization that has represented eighteen
2 Even if these statistics were relevant, the Court would exclude them under Rule 403 as likely to “mislead[] the jury.” Fed. R. Evid. 403; see Adams v. Austal, U.S.A., L.L.C., 754 F.3d 1240, 1258 (11th Cir. 2014) (“Even when [evidence showing the treatment of similarly situated persons] is relevant under Rule 401, the district court retains the discretion to exclude that evidence, under Rule 403, if it is unduly prejudicial, confusing, misleading, or cumulative.”). Here, Rempel’s report would likely invite the jury to draw the unwarranted conclusion that a given individual driver is unlikely to reoffend simply because drivers as a group are statistically unlikely to do so. drivers in OATH reinstatement proceedings; of those, she has personally represented seven drivers and has supervised other attorneys in representing six more. (Doc. No. 607-1 (“Ader Report”) at 2.)3 She offers a three-page report that advances several opinions regarding OATH proceedings, including (1) that the “demographics and socioeconomic background” of taxi drivers made their
suspensions uniquely harmful, (2) that OATH ALJs do not treat the fact of an arrest as “dispositive” and often grant reinstatement even to drivers charged with assault and weapons offenses, (3) that the 2020 to 2023 time period is not a reliable benchmark for assessing how likely drivers were to request an OATH hearing, and (4) that “criminal defense lawyers rarely advise drivers facing misdemeanor charges to [forgo] OATH hearings for fear of self-incrimination.” (Ader Report at 3–4.) To begin with, Ader may have legal experience in OATH proceedings, but she has no discernible expertise in “demographics and socioeconomic backgrounds.” (Id. at 3.) Ader is not a sociologist by training or trade and is not qualified under Rule 702 to give opinions on such topics. Fed. R. Evid. 702.
Ader’s expert opinions on OATH proceedings fare no better. Though Ader has worked with drivers in a handful of OATH proceedings, she does not base her opinions on “sufficient facts or data” nor does she form them using “reliable principles and methods.” See Fed. R. Evid. 702(b), (c). For starters, Ader has personally appeared in, or supervised others in, just thirteen OATH proceedings. (Ader Report at 1–2.) That sample size is a tiny fraction of the hundreds of OATH proceedings that have occurred since the Taxi and Limousine Commission altered its suspension protocols in 2019 and is not a sufficient basis for an expert opinion as to how OATH
3 Citations to the Ader Report refer to the ECF page numbers because that submission does not have independent page numbering. ALJs render reinstatement decisions. See Chen-Oster, 114 F. Supp. 3d at 124. Moreover, Ader provides no indication that these thirteen cases are in any way representative of the twenty drivers in the upcoming trial. As a result, the Court is not persuaded that Ader’s opinion is based on a relevant or reliable cross-section of OATH hearings. See id. (excluding an expert’s analysis
because his selection of the underlying sample was “subject to bias”). Nor does Ader even attempt to explain what “reliable principles and methods” she used to reach her opinions on OATH hearings. Fed. R. Evid. 702(c). “An expert opinion requires some explanation as to how the expert came to [her] conclusion and what methodologies or evidence substantiate that conclusion.” Riegel v. Medtronic, Inc., 451 F.3d 104, 127 (2d Cir. 2006). “[U]nsupported assertions . . . made without explanation or elaboration that would allow a fact finder to follow [her] reasoning and come to the same conclusion” are inadmissible expert opinions. 523 IP LLC v. CureMD.Com, 48 F. Supp. 3d 600, 650 (S.D.N.Y. 2014). Indeed, Ader offers only conclusions – such as that OATH ALJs do not treat arrests for even violent crimes as “dispositive” – without detailing how and why she came to that determination.4 Nowhere does
she indicate how many OATH hearings she has reviewed or whether (and how) she has developed a topology for which sorts of arrests are likely to lead to reinstatement decisions under the current OATH protocols. Her proffered opinion on the unreliability of the hearings conducted in the 2020 to 2023 timeframe as a basis for assessing the likelihood of drivers to request an OATH hearing is equally conclusory and devoid of analysis, as is her opinion that “criminal defense lawyers rarely advise drivers facing misdemeanor charges to [forgo] OATH hearings for fear of self- incrimination.” (Ader Report at 3–4.) It bears noting that Ader herself does not even claim to be
4 Defendants do not seem to dispute that the nature of the charge is not dispositive in ALJ determinations, making Ader’s opinion on this point of questionable relevance in any event. a criminal defense lawyer. In sum, Ader’s “report” strays far beyond her actual qualifications, relies on an insufficient data set, and offers no explanation of her methodology. She cannot testify as an expert. For these reasons, Defendants’ motion in limine to preclude Rempel and Ader as expert
witnesses is GRANTED.
SO ORDERED. Dated: October 18, 2023 New York, New York _______________________________ RICHARD J. SULLIVAN UNITED STATES CIRCUIT JUDGE Sitting by Designation