Federal Rule of Civil Procedure 54(d)(1) – E-Discovery Costs
Federal Rule of Civil Procedure 54(d)(1) provides that costs incurred by the prevailing party may be taxed against the losing party. In turn, 28 U.S.C. § 1920 itemizes the allowable expenses that may be taxed, permitting recovery of “[f]ees for exemplification and the costs of making copies of any materials where the copies are necessarily obtained for use in the case.” 28 U.S.C. § 1920(4). Over the years, federal courts have grappled with the question of how to interpret “the costs of making copies” in the digital age.
Back in 2010-2011, there seemed to be a trend towards a broad interpretation of “making copies” to include many routine e-discovery costs. However, that trend reversed in 2012 when the Third Circuit addressed the issue in Race Tires America, Inc. v. Hoosier Racing Tire, 674 F.3d 158 (3d Cir. 2012). Applying § 1920 narrowly, the court in Race Tires held that e-discovery charges typically would not qualify as taxable costs. Only fees for “scanning and file format conversion” (e.g., the conversion of native files to TIFF) would be considered a recoverable cost of “making copies.” The court refused to tax costs for any of the steps leading up to the actual copying of materials, including ‘gathering, preserving, processing, searching, culling, and extracting’ information for discovery.
Although most courts since Race Tires have adopted its narrow interpretation of § 1920, not all courts have agreed with this approach. Most recently, the Sixth Circuit rejected the restrictive application of § 1920 in Colosi v. Jones Lang LaSalle, 781 F.3d 293 (6th Cir. 2015). In that case, the winning defendant submitted a bill of costs for $6,369.55, which included the cost of imaging the plaintiff’s personal computer. On appeal, plaintiff challenged the taxing of costs for the imaging.
The Third Circuit first addressed “whether imaging a hard drive, or other physical storage device, falls within the ordinary meaning of ‘making copies.’” Looking to various definitions of the word “copy” from ordinary dictionaries and legal publications, the court concluded that it did. According to the court, “a plain reading of the statute authorizes courts to tax the reasonable cost of imaging, provided the image file was necessarily obtained for use in the case.”
The court rejected the narrow reading of § 1920 from Race Tires (in which the court disallowed the costs of imaging drives), finding that construction to be “overly restrictive.” As the court in Colosi noted, “while the Third Circuit rightly worried over expanding the scope of § 1920 to include expensive electronic discovery procedures not contemplated by Congress, this concern more appropriately pertains to the context-dependent question of whether the prevailing party necessarily obtained its copies for use in the case.”
So what are the takeaways from Colosi? First, the very narrow interpretation of § 1920 in Race Tires is not universal—depending on the jurisdiction, prevailing parties may be able to recover broader categories of e-discovery costs. Second, winning parties should focus on whether each cost they seek to recover was incurred for copies “necessarily obtained for use in the case.” Parties might consider including in their ESI discovery agreements provisions specifying particular forms of production for different types of ESI that will be deemed “necessary”—it would be tough for a losing party to argue that an ESI “copying” cost was not necessarily obtained if the ESI agreement required it. Finally, Colosi highlights again the poor fit between § 1920—a statute written in the pre-digital age—and litigation today, where discovery is far more complex than simply “making copies.” Let’s hope that Congress sometime soon revises § 1920 to reflect the modern age.