Allstate Insurance v. Countrywide Financial Corp.
984 F. Supp. 2d 1021
C.D. Cal.2013Background
- This is an order granting in part and denying in part defendants’ motions to exclude seven expert reports by Cowan and Green on RMBS loan sampling.
- Six consolidated Countrywide RMBS cases involve proposed random sampling of loans from SLGs to test underwriting/valuation representations.
- Two categories of expert reports: six Case-Specific Reports (sampling 100 loans per certificate class) and one Broad Report (1,040 loans across 447 securitizations).
- Court applied Fed. R. Evid. 702 Daubert standards to assess qualification, reliability, and helpfulness of the sampling methodologies.
- Court held Cowan and Green qualified as experts; Case-Specific sampling methodology is admissible; Broad Report sampling methodology is inadmissible for unreliability and potential systematic bias.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Whether Cowan and Green are qualified as experts. | Plaintiffs say both are qualified in statistics/economics to design sampling. | Defendants contend Green lacks statistical expertise; Cowan is qualified. | Both Cowan and Green qualified as experts. |
| Whether Case-Specific sampling methods satisfy Rule 702 reliability. | Case-Specific sampling uses standard methods, stratification by FICO; sufficient data for 95% conf., ±10% MOE. | Sample size and possible data issues render reliability questionable. | Case-Specific sampling method is reliable under Rule 702. |
| Whether Broad Report sampling method is admissible. | Broad sampling estimates defect rates for all Countrywide RMBS and supports systemic abandonment. | Broad sampling suffers from nonrandom data, selection bias, and multi-stage cluster flaws. | Broad Report excluded as scientifically unreliable under Rule 702. |
| Whether extrapolation methodology must be proposed upfront. | Extrapolation method may be chosen after data collection; timing litigations decisions aside from scientific validity. | Extrapolation methodology should be set forth in advance. | Admissibility reserved; extrapolation plan to be evaluated with future expert submissions. |
Key Cases Cited
- Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993) (gatekeeping; reliability of expert testimony)
- Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999) (broader gatekeeping standard for experts across disciplines)
- In re Silicone Gel Breast Implants Prods. Liab. Litig., 318 F.Supp.2d 879 (C.D. Cal. 2004) (three-part Rule 702 analysis (qualification, reliability, usefulness))
- Assured Guar. Mun. Corp. v. Flagstar Bank, FSB, 920 F.Supp.2d 475 (S.D.N.Y. 2013) (sampling reliability; representativeness and data concerns)