Crowell & Moring

In a Valentine’s Day order, Judge Denise Cote of the Southern District of New York ruled that the defendants in a set of complex mortgage-backed securities cases could not use documents obtained through discovery in related litigation in the Central District of California. This even included documents that purportedly should have been produced in the New York cases. The court’s ruling rests primarily upon pragmatic notions of efficiency, fairness, and reasonableness—and its reasoning may prove useful beyond the four corners of these cases.

The New York cases are FHFA v. HSBC N. Am. Holdings, Inc., et al., Nos. 11 Civ. 6189, 6190, 6193, 6195, 6198, 6200, 6201, 6202, 6203, 6739, and 7010 (DLC) (S.D.N.Y.). (This litigation also has been referenced as FHFA v. UBS Americas Inc., No. 11 Civ. 5201, based on a component case that has since settled.) The California litigation includes FHFA v. Countrywide Financial Corp., No. 12-cv-1059-MRP (C.D. Cal.), among other cases.

Judge Cote’s order might be construed, as the defendants argued, as “subversion of the truth-seeking process” in favor of artificial rules of the game. But the context is important: document production had long ended in the New York cases, and the court concluded that introducing a large batch of new discovery material at that late stage could have called into question depositions already completed, thrown litigation strategies into disarray, and potentially delayed multiple trials.

The court sought to avoid undue burden and delay in what already has been lengthy and costly litigation. In so doing, the court championed the principles of efficiency, fairness, and reasonableness inherent in the Federal Rules of Civil Procedure over an impractical standard of perfection in discovery.

The litigation involves a series of claims by the Federal Housing Finance Agency (FHFA) against various financial institutions related to mortgage-backed securities sold to Fannie Mae and Freddie Mac. Judge Cote set the tone for the current dispute last summer, when she relied on Advisory Committee and Sedona Conference materials to hold that “the [Federal] Rules and sound policy thus require that a court not simply allow discovery of all potentially relevant material,” but rather limit the scope of even relevant discovery based on the burden it would impose.

In early 2014, document production had ended in the New York litigation and depositions were winding down, but in the California litigation document production had just begun. Some of the defendants sought to use documents they were beginning to receive from the California cases for a variety of purposes in the New York cases.

Judge Cote denied a request to use the California discovery material in the New York cases in a January 8 order, and doubled down in a February 14 order clarifying that the defendants cannot use the California documents for any purpose in the New York litigation, even to show that FHFA purportedly failed to comply with its discovery obligations in New York. She allowed only a narrow exception, permitting the defendants to submit California documents to show any perjured witness testimony.

Why not allow the defendants to use all relevant documents in their possession? According to Judge Cote, the defendants’ use of documents from the California litigation would “undermine the discovery process” and “encourage trial by ambush.”

Even though FHFA presumably was aware of what documents it produced in California, the court reasoned that allowing those documents to be used in the New York litigation would in effect extend document production through the deposition period and up to the eve of trial in New York. Because the parties could not have relied on the California documents for their depositions in the New York litigation (which had largely concluded) or for their trial preparation in New York (with the first trial scheduled for June 2014), Judge Cote foresaw significant hurdles with suddenly introducing the California discovery materials into the New York litigation record. She observed that either the parties would be forced to litigate—and prepare for trial—on a rapidly shifting playing field as new documents would continue to be produced in the California cases, or depositions might need to be retaken and trial delayed. Judge Cote viewed neither outcome as desirable or appropriate.

The defendants sought to use the California documents, in part, to show FHFA’s purported noncompliance with its discovery obligations in New York. Judge Cote rejected this argument as a “gotcha” tactic and pragmatically acknowledged that “no one could or should expect perfection” from the discovery process. The court ruled that although parties must conduct discovery diligently and in good faith, even if FHFA failed to produce certain relevant documents in the New York litigation, that alone would not demonstrate a failure to satisfy discovery obligations. Therefore, according to Judge Cote, it would serve little purpose to allow the defendants to point to California documents they contend should have been produced in the New York litigation.

To illustrate this important point, Judge Cote referenced her earlier decision to allow one of the defendants to produce documents through predictive coding, which is recognized as having some margin of error but still better accuracy than traditional manual review. Acknowledging that both predictive coding and traditional manual review “fell well short of identifying for production all of the documents the parties in litigation might wish to see,” the court emphasized that perfection in discovery is simply not attainable, nor is it required. Judge Cote found that good faith production (through traditional manual review or predictive coding) should be sufficient to create “a reliable factual record.”

Judge Cote’s ruling is a welcome reminder that reasonableness and good faith are all that should be required in discovery. Although parties and counsel must take their discovery obligations seriously, courts should be wary of litigants seeking to exploit the inevitable imperfections in discovery to gain leverage or some tactical advantage, increasingly through spoliation motions seeking disproportionate—or even unwarranted—sanctions. Parties should not have to fear that an eleventh hour revelation of some reasonable misstep, such as marginally relevant documents not captured through an entirely reasonable and good faith discovery protocol, will necessarily serve to reopen discovery, delay litigation, or lead to sanctions. Judge Cote’s ruling may prove to be a boon for parties using, and ultimately defending, predictive coding (or technology-assisted review) protocols—or other reasonable and effective solutions—for discovery.

That said, Judge Cote’s decision is heavily dependent on the specific context—namely, the facts that discovery had closed in New York and that those cases were headed to trial. If either of these facts were not present, perhaps the court would have reached a different ruling. The somewhat unique circumstances of the litigation call into question broader application of the court’s decision. Nevertheless, the ruling is at least a principled victory for those desiring greater pragmatism and understanding from the courts in assessing discovery.