606 F.Supp.3d 547
E.D. Mich.2022Background
- Plaintiffs are owners/lessees of 2014–2015 Chevrolet Cruze diesel vehicles who allege GM and Bosch installed software ("defeat devices") that caused real-world NOx emissions to exceed certified levels, producing an overpayment injury and supporting class claims.
- Extensive expert discovery: Plaintiffs proffered experts on PEMS/chassis-dyno testing (Smithers), ECU/software analysis (Levchenko), marketing (Shankar), and damages (Stockton); Defendants proffered rebuttal experts including Harrington (engineering/PEMS critique), Molden (PEMS methodology), Neuendorf and Bradlow (marketing), and Hitt (damages).
- Defendants moved under Daubert to exclude four Plaintiffs’ experts; Plaintiffs moved to exclude five of Defendants’ experts; Defendants also moved to strike Plaintiffs’ later rebuttal expert reports and related briefing.
- The court held multiple Daubert hearings on the papers and ruled: it denied defendants’ motions to exclude Plaintiffs’ experts (Smithers, Levchenko, Shankar, Stockton), denied most of Plaintiffs’ Daubert challenges to defendants’ experts, but granted Plaintiffs’ motion to exclude Ryan Harrington’s testimony on PEMS testing (Harrington may still testify about non-PEMS topics); the motion to strike Plaintiffs’ rebuttal reports was denied.
- Key legal determinations: (1) qualitative social‑science/methodologies (marketing/content analysis) and economic regression/overpayment models may be admissible even if not strictly ‘‘hard‑science’’; (2) qualification to criticize specialized testing (PEMS) requires specific experience with that methodology; (3) challenges to expert methodology or sampling generally go to weight, not admissibility, absent junk science or lack of reliable foundation.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Admissibility of Juston Smithers (PEMS & dynamometer testing) | Smithers’ PEMS/dyno testing shows defeat devices and elevated on-road NOx; methodology reliable though single‑vehicle testing | Smithers relied on one vehicle, ignored inconvenient facts, and extrapolated improperly | Denied: Smithers admissible; single‑vehicle PEMS testing meets industry practice and fits the issues (weight/credibility for cross‑examination) |
| Admissibility of Kirill Levchenko (ECU/software analysis) | Levchenko’s software calibration analysis corroborates Smithers and identifies temperature and dosing mechanisms (cycle‑beating) | Not an automotive engineer; used novel terminology and methodology | Denied: Levchenko qualified on software/calibration issues; methodology reliable and helpful; use of common‑meaning term "cycle‑beating" permitted |
| Admissibility of Venkatesh Shankar (marketing/content analysis) | Qualitative content analysis of GM ads, internal marketing and surveys shows themes (fuel economy, performance, environment) and materiality | Qualitative method not scientific; sample selection and coding unreliable | Denied: Qualitative content analysis and expert experience are admissible; sampling and coding criticisms go to weight, not exclusion |
| Admissibility of Edward Stockton (damages: DPP, Regression Pricing, Retail Replacement Cost) | Stockton’s overpayment and replacement‑cost models provide classwide per‑vehicle damages using standard economic data and methods | Models rest on assumptions, averaging, hypothetical inputs, and fleet uniformity; speculative for used/leased vehicles | Denied: Models use accepted economic techniques and industry data; challenges to inputs and assumptions go to weight and allocation is a post‑certification issue |
| Admissibility of Ryan Harrington (rebuttal on PEMS and other topics) | Harrington contends PEMS testing and Smithers’ methodology are unreliable | Harrington is a qualified automotive engineer and DOT Division Chief; his critique should be admitted | Granted in part: Harrington excluded from testifying about PEMS testing/data analysis (unqualified and unfamiliar with PEMS); may testify on other non‑PEMS topics he was retained for |
| Motion to strike Plaintiffs’ rebuttal expert reports and opposition briefs | — | Rebuttal reports were late, prejudicial, exceed scope, and required striking or rebriefing | Denied: Reports deemed declarations within scope; lateness was harmless given time since disclosure; policy favors deciding merits over striking evidence |
Key Cases Cited
- Frye v. United States, 293 F. 1013 (D.C. Cir. 1923) (articulating the historical "general acceptance" test for scientific evidence)
- Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993) (admissibility turns on reliability and relevance; gatekeeping role of trial judge)
- Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999) (Daubert gatekeeping applies to all expert testimony, not only "scientific")
- Gen. Elec. Co. v. Joiner, 522 U.S. 136 (1997) (court may exclude expert opinions where an analytical gap exists between data and conclusions)
- Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442 (2016) (statistical/representative evidence can be admissible in class contexts)
- Surles ex rel. Johnson v. Greyhound Lines, Inc., 474 F.3d 288 (6th Cir. 2007) (Daubert inquiry is context‑specific and flexible)