378 F. Supp. 3d 589
E.D. Mich.2019Background
- Eastpointe, MI elects a mayor and four councilmembers at-large; no African American had been elected to City Council before 2017; black population rose rapidly 2000–2015 to roughly 41% by latest estimates.
- The United States sued under Section 2 of the Voting Rights Act, alleging the at-large system dilutes black voting power; defendants are the City and individual officials.
- Experts (Dr. Lisa Handley for plaintiff; Dr. John Alford for defendants) analyzed electoral cohesion and bloc voting using ecological regression and ecological inference; Handley also used Bayesian Improved Surname Geocoding (BISG) to estimate voters’ race from name/address data.
- The parties submitted cross-motions and Daubert challenges: defendants moved for summary judgment and to exclude BISG; plaintiff moved to strike defendants’ supplemental expert disclosures; defendants moved to strike a supplemental exhibit.
- The court treated many of the parties’ disputed statistics as accepted for summary-judgment purposes, reviewed Gingles preconditions, found genuine factual disputes on key elections, and addressed admissibility of BISG in the context of a bench trial.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| First Gingles precondition (size/compactness) | Black voters form a geographically compact majority in a single-member district; Handley’s illustrative map shows a >50% black VAP/CVAP district | At-large system already provides representation; illustrative plan efficacy irrelevant | Court: genuine issue of material fact resolved in plaintiff’s favor for summary-judgment purpose; first precondition met (summary judgment for defendants denied) |
| Second Gingles precondition (political cohesion) | Black voters vote cohesively for preferred candidates (conceded by defendants for summary judgment) | Conceded for summary-judgment purposes | Court: treated as conceded; second precondition established for summary-judgment analysis |
| Third Gingles precondition (majority bloc voting) | Recent endogenous and exogenous elections show white bloc voting usually defeats minority-preferred candidates (disputed by defendants) | Disputes specific election analyses and results (challenges Handley’s findings; favors their expert’s analyses) | Court: material factual disputes (including which candidates were black-preferred in key elections and presence of special circumstances) preclude summary judgment for defendants |
| BISG admissibility (Daubert challenge) | BISG is a reliable, peer-reviewed technique that refines race estimates for actual voters and helps detect racial bloc voting | BISG is untested/insufficient/unreliable for Section 2 analyses and should be excluded | Court: denied motion to exclude BISG without prejudice given bench-trial context; court will weigh BISG’s strengths/limits at trial (defendants may renew in limine) |
Key Cases Cited
- Thornburg v. Gingles, 478 U.S. 30 (explains three preconditions for §2 vote-dilution claims and the intensely local inquiry)
- Rogers v. Lodge, 458 U.S. 613 (discusses how at-large schemes can minimize minority voting strength)
- Cousin v. Sundquist, 145 F.3d 818 (6th Cir.) (directs that majority-bloc analysis considers minority-preferred candidates and relevant elections)
- City of Euclid v. [unidentified parties], 580 F. Supp. 2d 584 (N.D. Ohio) (endorses ecological regression/inference methods in vote-dilution context)
- Abbott v. Perez, 138 S. Ct. 2305 (addresses relevance of whether an alternative plan would enhance minority voters’ ability to elect preferred candidates)