DoubleLine Capital LP v. Odebrecht Finance, Ltd

Background

DoubleLine Capital LP, DoubleLine Income Solutions Fund, and DoubleLine Funds Trust sued Odebrecht, S.A. (OSA), Construtora Norberto Odebrecht, S.A. (CNO), and Odebrecht Engenharia e Construção S.A. (OEC). The plaintiffs allege that misstatements and omissions about Odebrecht’s international bribery scheme induced them to buy Odebrecht bonds at inflated prices. Their claims include federal securities-fraud claims under Section 10(b) of the Securities Exchange Act and Securities and Exchange Commission Rule 10b-5, a control-person claim under Section 20(a), and state-law claims for fraud and negligent misrepresentation.

The opinion states that earlier proceedings established, for purposes of the action, that defendants made the alleged misstatements and omissions, that they were material, and that defendants acted with the required intent. The remaining issues include loss causation—whether the alleged fraud caused the plaintiffs’ losses—and damages.

The Expert Report

DoubleLine offered the expert report of Sanjay Unni, Ph.D., concerning loss causation and damages. Unni conducted an event study, a statistical analysis designed to separate the effect of particular company-related news from broader market movements. He analyzed the daily returns of two Odebrecht bonds held by DoubleLine between June 19, 2015, when Marcelo Odebrecht was arrested, and May 6, 2016, when he was formally indicted.

Unni compared the bonds’ performance with two J.P. Morgan emerging-market corporate-bond indexes matched primarily by credit rating and time to maturity. He identified days with statistically significant abnormal returns and then reviewed historical news to identify days involving disclosures or developments concerning Odebrecht’s bribery scheme. His report identified 19 such days and estimated damages of $17,307,000.

Defendants’ Challenges

The defendants moved to exclude Unni’s report and testimony under Federal Rule of Evidence 702 and the standards established in Daubert v. Merrell Dow Pharmaceuticals. They did not challenge Unni’s qualifications, the relevance of his testimony, or the need for expert assistance. Instead, they argued that his methods were unreliable.

The defendants challenged Unni’s choice of benchmark indexes because the indexes were not specifically matched to Odebrecht’s industry or geographic markets. They also argued that the indexes had weak correlations with the subject bonds. In addition, they contended that Unni failed to account adequately for other economic news, including developments in Brazil, and criticized his treatment of days when the two bonds moved in different directions or had offsetting returns on consecutive days.

The defendants raised three further methodological objections: Unni constructed a portfolio based on DoubleLine’s holdings rather than all Odebrecht bonds; he used a t-test that allegedly assumed abnormal returns followed a normal distribution; and he did not adjust for the problem of testing many separate dates for statistical significance.

Court’s Analysis

Judge Moses concluded that the report satisfied Rule 702. The court explained that expert evidence must be based on sufficient information, reliable methods, and a reliable application of those methods. The court’s role is to assess reliability and relevance, not decide whether the expert’s conclusions are ultimately correct. Challenges that affect the weight or credibility of admissible evidence generally may be addressed through cross-examination and contrary evidence.

Regarding the benchmark indexes, the court found that academic research and the limited case law concerning corporate-bond event studies support matching bonds by credit rating and time to maturity. The defendants did not identify authority requiring industry- or region-specific indexes, nor did they identify a better alternative benchmark. The court also found that the reported positive correlations, although characterized by defendants as weak, did not make the study inadmissible. These objections went to the weight of the testimony rather than its admissibility.

Regarding confounding factors, the court found that Unni had taken reasonable steps to separate fraud-related news from broader economic developments. He considered statistically significant abnormal returns, reviewed news specifically related to Odebrecht’s alleged misconduct, and excluded days involving news about the Lava Jato investigation that did not specifically mention Odebrecht or broader Brazilian economic developments. The court acknowledged that defendants identified legitimate questions about whether Unni consistently applied his criteria, but concluded that reasonable experts could disagree about those judgments.

The court likewise rejected the remaining objections as grounds for exclusion. Using DoubleLine’s holdings to construct the subject-bond portfolio was not an apparent error because the damages analysis concerned the securities DoubleLine actually held. The use of a traditional t-test was recognized as a standard event-study method. Finally, although multiple testing could potentially lead to false findings of statistical significance, defendants did not show how an adjustment would change the results or identify a flaw serious enough to require exclusion.

Disposition

The court concluded that DoubleLine had shown it was more likely than not that Unni’s analysis was reliable at each step. Judge Moses therefore denied defendants’ motion to exclude Unni’s testimony. The ruling allowed the report and testimony to be presented; it did not decide the ultimate liability, loss-causation, or damages issues.