AstraZeneca AB v. United Food & Commercial Workers Unions
777 F.3d 9
| 1st Cir. | 2015Background
- AstraZeneca marketed Nexium and settled separate Hatch-Waxman paragraph IV suits with Ranbaxy, Teva, and DRL by paying them and obtaining agreements delaying generic entry until main patents expired in 2014 (so-called reverse payments).
- Named plaintiffs are third-party payors (TPPs) and consumers who allege the settlements unlawfully foreclosed competition, caused supra-competitive Nexium prices from April 14, 2008 to May 27, 2014, and seek a Rule 23(b)(3) damages class.
- The district court certified a nationwide class of purchasers of Nexium (with defined exceptions) despite acknowledging that some class members might be uninjured; defendants appealed under Rule 23(f).
- Plaintiffs’ liability theory relies on Actavis and a rule-of-reason antitrust challenge to reverse-payment settlements; plaintiffs’ damages expert estimated but-for prices using a yardstick (Prevacid) to model generic pricing.
- Defendants argued certification failed because the class includes a (they contend substantial) number of uninjured members (brand‑loyalists, coupon users, certain TPP-plan structures, etc.), which would defeat predominance and defeat administrability and Article III standing for unnamed class members.
- The First Circuit affirmed, holding a class may include a de minimis number of uninjured members so long as (1) the class is definite/ascertainable, (2) aggregate recovery can be limited to actual injury, and (3) a feasible, protective mechanism exists to exclude uninjured members prior to judgment.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Whether a Rule 23(b)(3) class may include some uninjured members | A class may include a de minimis number of uninjured members; individualized culling can occur later | Any uninjured members (even de minimis) defeat predominance because common proof cannot establish fact of injury for all | A class may include a de minimis number of potentially uninjured members; presence of such members does not automatically defeat certification |
| Whether plaintiffs met predominance with common proof of antitrust impact and damages | Rosenthal's yardstick model shows widespread overcharge and common proof of impact for vast majority | Defendants’ expert identifies five groups likely uninjured, undermining common proof and predominance | District court did not abuse discretion; plaintiffs met burden and defendants failed to show more-than-de minimis uninjured group |
| Whether an administratively feasible mechanism exists to exclude uninjured members before judgment | Injured vs. uninjured can be sorted later by presumptions or affidavits/testimony; Rule 23 does not demand final sorting at certification | No specific workable method has been proposed; class-wide affidavits would be unworkable and may violate defendants’ rights | Court held it was sufficient that a feasible mechanism could be developed later and defendants’ speculation that none could be was insufficient to deny certification |
| Article III standing for the class | Named plaintiffs suffered injury and have standing; unnamed members need not each show standing at certification so long as injured representatives exist | Because some class members lack injury, class-wide standing is deficient | Named plaintiffs’ standing suffices for certification; only injured class members will recover and standing for individual recoverants will be ensured before payment |
Key Cases Cited
- FTC v. Actavis, 133 S. Ct. 2223 (2013) (reverse-payment settlements evaluated under rule-of-reason antitrust analysis)
- Comcast Corp. v. Behrend, 133 S. Ct. 1426 (2013) (damages model must measure damages tied to the asserted theory of liability to satisfy Rule 23(b)(3))
- Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011) (Rule 23(a) commonality requirement and limits on merits inquiry at certification)
- Amgen Inc. v. Connecticut Retirement Plans, 133 S. Ct. 1184 (2013) (class certification does not require proving each element on a classwide basis; rigorous analysis limits merits probing)
- Halliburton Co. v. Erica P. John Fund, 134 S. Ct. 2398 (2014) (presumptions may be used at certification and individualized rebuttal does not necessarily defeat predominance)
- Illinois Brick Co. v. Illinois, 431 U.S. 720 (1977) (federal rule barring indirect-purchaser antitrust damages suits; background for state-law indirect purchaser claims)
- Messner v. Northshore Univ. HealthSystem, 669 F.3d 802 (7th Cir. 2012) (small percentage reductions in class size from removing uninjured members may not defeat certification)
- In re New Motor Vehicles Canadian Export Antitrust Litig., 522 F.3d 6 (1st Cir. 2008) (plaintiffs must show at certification that their model can preliminarily identify which consumers were impacted)