DOJ Antitrust Division Issues New Model Second Request, with New Predictive Coding Instructions
On November 28, 2016, the Department of Justice (DOJ) Antitrust Division issued an updated Model Second Request, aimed at revising and streamlining the model to conform to “current division practice.” The updated Model will be used for all Second Requests issued on or after December 12, 2016. The new model contains significant changes to merging parties’ obligations during a Second Request, as well as a substantial formatting overhaul.
Regarding the use of ediscovery technology during a Second Request from the DOJ, the predictive coding instructions were meaningfully modified.
First, the new model appears to signal an increased acceptance of use of predictive coding during a second request. Specifically, the searching and predictive coding instruction begins with the following new language, “Before using software or technology…” seemingly indicating that the Antitrust Division recognizes that it is not a matter of “if” parties are leveraging technology but “when” and “how” that technology will be used.
Second, the new model requires merging parties and their counsel to be more astute than ever before when it comes to ediscovery technology. For example, if search terms are used, merging parties must now submit a list of stop words and operators for the platform being used. Also, if predictive coding technology is used to identify or eliminate documents, merging parties must provide more than just a description of the methods being used. Under this new model, the Antitrust Division also is requiring information about the use of subject matter experts to review seed sets and training documents, effectiveness metrics (such as recall, precision and confidence-intervals) and validation protocols, including sampling protocols used to categorize non-responsive documents.
The new predictive coding and searching instruction is provided in full below:
November 2016 Version – DOJ Model Second Request
- Before using software or technology (including search terms, predictive coding, de-duplication, or similar technologies) to identify or eliminate documents, data, or information potentially responsive to this Request, the Company must submit a written description of the method(s) used to conduct any part of its search. In addition, for any process that relies on search terms to identify or eliminate documents, the Company must submit: (a) a list of proposed terms; (b) a tally of all the terms that appear in the collection and the frequency of each term; (c) a list of stop words and operators for the platform being used; and (d) a glossary of industry and company terminology. For any process that instead relies on predictive coding to identify or eliminate documents, you must include (a) confirmation that subject-matter experts will be reviewing the seed set and training rounds; (b) recall, precision, and confidence-level statistics (or an equivalent); and (c) a validation process that allows for Department review of statistically-significant samples of documents categorized as non-responsive documents by the algorithm.
As these new instructions reinforce, Second Requests are synonymous with sheer complexity. At Kroll Ontrack, we have leading technology backed by human experts who know how to successfully navigate a Second Request. Kroll Ontrack is uniquely equipped to help manage your document productions to the FTC, DOJ and other global competition bureaus.