Under Daubert, district courts serve as gatekeepers to ensure that expert testimony is reliable before it is presented to a jury. But “reliable” with regard to what? Methodology? Data? This was the central question the 7th Circuit recently answered in reviewing a district court’s decision to exclude the plaintiff’s damages expert based on his data selection.

Business downfall: A building collapse in 2006 damaged the plaintiff’s Paris office. In addition to the property damage, the plaintiff was unable to conduct business, so it lost income. Its claims for insurance coverage triggered a dispute with the defendant insurer, which the parties litigated in federal district court. The plaintiff’s damages expert calculated a business interruption loss of over €5.1 million. His determination followed the general procedure set down in the insurance policy—computing lost revenues minus noncontinuous expenses—and the method for estimating the various inputs. A policy provision required that in calculating net profit “due consideration shall be given to the experience of the business before the date of damage or destruction and to the probable experience thereafter had no loss occurred.” By comparing the total revenues from the five-month period that preceded the collapse to total revenues generated in the same five-month period in 2005, the expert estimated a growth rate of 7.76%. Even though he had reviewed data going back to 2003—and the historical data indicated a downward trend—he decided to use a shorter period from which to extrapolate the rate based on company information that new management and policies had turned the business around by the end of 2005.

Ruling on the defendants’ Daubert motion, the district court found his calculation unreliable. Although the expert had used an “appropriate and recognized method” to compute projected revenues, his “analysis [broke] down” when it came to his estimated growth rate, “one of the most important parts of the business-interruption calculation.” He was not “an expert on business management,” and his conversations with the company’s managers were not a reliable basis for the revenue forecast. Had he selected a period that was not so short, the court “might have found his analysis reliable.” The plaintiff asked the district court for reconsideration based on additional information on how the expert had developed the growth rate, but to no avail. In excluding the testimony, the court effectively nixed the plaintiff’s business interruption claim.

The plaintiff was able to salvage its claim by appealing to the 7th Circuit. At the start of its analysis, the Court of Appeals noted that Daubert gives district courts considerable—but not unlimited—leeway to determine whether expert testimony is reliable. “Reliability, however, is primarily a question of the validity of the methodology employed by an expert, not the quality of the data used in applying the methodology or the conclusions produced.” In other words, the jury, as the trier of fact, not the court, gets to decide whether the data underlying the expert’s analysis is sound and whether the expert’s conclusions are correct.

Here, after the district court found the expert’s methodology was reliable, it should have stopped its analysis. Instead, it “drilled to a third level” to evaluate the quality of the data inputs the expert chose for his growth rate extrapolations. In so doing, it “unduly scrutinized” the quality of the data the expert used for his model, the appellate court said. Therefore, it reversed and remanded for a fuller evaluation of the testimony.

Find an extended discussion of Manpower, Inc. v. Ins. Co. of Pa., LLC, 2013 U.S. App. LEXIS 20959 (Oct. 16, 2013), in the February issue of Business Valuation Update; the court opinion will be available soon at BVLaw. Digests of the earlier district court rulings and the opinions are also available there.