The ACC Docket series "Privileged Conversations” is a collection of interviews and discussions on in-house privilege and ACC’s efforts to defend it worldwide.
In January, the ACC Docket sat down with Thomas Bean, a partner at Verrill in Boston, to discuss ACC’s amicus curiae filing in Attorney General v. Facebook, Inc. The Facebook case stems from the Cambridge Analytica investigation and Facebook’s internal investigation, which were co-led by in-house and outside counsel. The case called into question the extent to which the attorney-client privilege and work product doctrines protect documents created as part of that internal investigation. On March 24, the Massachusetts Supreme Judicial Court issued its decision.
In this edition of Privileged Conversations, the ACC Docket and Danette Wineberg, who teaches in-house practice of law at the University of New Hampshire Franklin Pierce School of Law and serves as chair of the Advocacy Committee of the ACC Northeast Chapter, ask Tom Bean about the implications of the decision on internal investigations led by in-house counsel in the United States.
ACC Docket: Danette, you were first to bring Attorney General v. Facebook, Inc. case to the attention of the ACC Advocacy Committee as an opportunity to fight against the frittering of in-house privilege in the United States. Tell us what happened.
Danette Wineberg: I learned about this case from a Northeast Chapter Board member. Facebook had produced documents in response to the first two of three civil investigative demands, but the AG’s third demand contained a group of requests seeking information about the apps that Facebook decided to review during its investigation.
A separate request demanded Facebook’s internal communications correspondence concerning certain categories of apps and/or apps for which Facebook conducted specific reviews or investigations.
Facebook contested these requests. It claimed that its investigation had been "designed, managed, and overseen" by its inside and outside counsel, and that these attorneys "devised and tailored the [investigation’s] methods, protocols, and strategies to address the specific risks posed by these legal challenges." It refused to turn over the documents responsive to the contested requests claiming that the documents were protected from disclosure by the attorney-client privilege and the work product doctrine.
Wineberg: What do you see as the key lessons from the court’s decision for in-house counsel?
Thomas Bean: Three lessons may be gleaned from the Facebook decision concerning applicability of the attorney-client privilege and the work product doctrine to internal investigations:
- The attorney-client privilege protects communications between an attorney and client, and not the underlying facts, even when those facts are contained in such communications. Even if information collected in the ordinary course of business is relayed to counsel through confidential communications, the facts themselves are generally not shielded from disclosure. However, factual information communicated to counsel for the purpose of obtaining legal advice, or in response to counsel’s request for information sought to assist counsel in providing legal advice to management, will likely be protected by the attorney-client privilege if no previously existing documents created in the ordinary course contain this factual information.
- The decision distinguished between retrospective investigations conducted with the “prospect of litigation in mind,” and investigations conducted in the ordinary course to improve ongoing or future operations. Although the prospect of litigation need not be the sole or even the primary purpose of an investigation, it is essential that it be at least one of the purposes of the investigation. Thus, simply funneling an organization's investigation through counsel does not bring with it the protection of the work product doctrine if the organization would have conducted the activities irrespective of anticipated litigation.
- While organizations often want to issue public statements to assure the public and/or stakeholders that the organization is investigating an allegation or an acknowledged problem, even carefully drafted, general statements about an internal investigation may jeopardize protection of attorney opinion work product that would otherwise be protected. The court reasoned that “publicized general statements regarding [an investigation] ... reduc[es] the risk that disclosure of [the information] will reveal confidential opinion work product.” It concluded, “a corporate defendant cannot publicize its ‘internal investigation to assert the propriety of its actions to third parties and simultaneously ... block third parties from testing whether its representations about the internal investigation are accurate.’"
Wineberg: While the Facebook decision is based on Massachusetts law, given its importance and the number of non-Massachusetts decisions on which it relies, in-house counsel in many states will surely want to review it carefully.
Wineberg: How does this decision affirm attorney-client privilege protection for investigations like the one Facebook undertook?
Bean: In its ruling, the court reiterated that "[a] construction of the attorney-client privilege that would leave internal investigations wide open to third-party invasion would effectively penalize an institution for attempting to conform its operations to legal requirements by seeking the advice of knowledgeable and informed counsel." It highlighted the distinction between attorney-client communications, which are protected by the privilege, and the underlying facts, which are not if they are available from another source.
The court noted that the first group of the AG’s contested requests for information asked only for the factual information that Facebook collected, which did not require disclosing communications between Facebook and its attorneys.
These requests therefore did not implicate the attorney-client privilege. The court found persuasive that the AG had agreed that Facebook could produce the information requested “in a spreadsheet format” in lieu of producing responsive documents themselves.
The court viewed the AG’s request which asked for Facebook’s “internal communications” about the apps differently. The court said that any confidential communications between or among Facebook, its outside counsel, and other members of the investigatory team during the investigation would almost certainly be privileged, and directed Facebook to prepare a detailed privilege log. The court also reiterated the bedrock principle that Facebook could not be compelled to answer the question, “What did you say or write to the attorney?”
Wineberg: US law distinguishes between the attorney-client privilege and the work product doctrine. Privilege generally keeps communications between attorneys and clients secret. Under the work product doctrine, an opposing party may not obtain materials prepared by or for an attorney in anticipation of litigation.
Read Privileged Conversations: Netherlands Court Affirms In-house Privilege to learn how the ruling in Royal Dutch Shell has helped more clearly establish the parameters of in-house privilege in the Netherlands.
Wineberg: What was the court’s position on the work product doctrine?
Bean: Although the work product doctrine shields documents prepared “in anticipation of litigation” from discovery, the court made clear that: (1) there does not need to be any litigation pending when the document was created so long as it was prepared with “the prospect of litigation in mind” and (2) the threat of litigation does not need to be the only, or even primary, motivation for creating the document for it to be protected — although it must be a necessary reason.
Applying these principles, the court found that the app information Facebook compiled during its internal investigation was clearly work product. Importantly for in-house counsel, the court carefully distinguished between Facebook’s internal app investigation, which was focused on past violations, and Facebook’s business-as-usual compliance program aimed at improving its ongoing operations. The court highlighted that Facebook’s compliance team could not and would not have conducted the large-scale investigation into app-developer misconduct in the ordinary course of its business.
The more difficult question was whether the app information was “fact” work product or “opinion” work product, a line the court recognized is sometimes not clear. The distinction is important because opinion work product is rarely discoverable, whereas fact work product may be discoverable if the party seeking the information demonstrates a “substantial need” for the information and “undue hardship” in obtaining it through other methods.
Wineberg: Is there something unique about the work product analysis as applied to this case that is particularly relevant to in-house counsel?
Bean: The specific facts here are important and informative. In response to negative publicity about the scandal, Facebook had made multiple public statements about the existence and progress of its internal investigation. And, in response to earlier demands from the AG, Facebook had told the AG the criteria that it used to identify apps during each phase of its investigation. In fact, the AG’s contested requests closely tracked the language that Facebook had used during this explanation.
With this background in mind, the court focused its opinion-or-fact work product analysis on whether the AG’s requests would reveal any confidential information about Facebook’s investigation that was not previously disclosed.
Noting that the first group of the AG’s contested requests allowed Facebook to remove any attorney impressions, and reflected information touched upon in Facebook’s public statements, the court was skeptical about whether producing the app information requested would reveal opinion work product. However, the court acknowledged the possibility that at least two of these contested requests might reveal counsel’s previously undisclosed thought processes and remanded the case to the trial court for a “document-by-document work product review.”
Any documents later identified by the trial court as “fact” work product would need to be produced because the AG demonstrated a substantial need and undue hardship. It observed that the AG is statutorily obligated to protect Massachusetts consumers; there is a strong public interest in disclosure when the AG is investigating potentially unfair or deceptive trade practices involving Massachusetts consumers that is absent in a typical civil dispute. Furthermore, the AG would be unable to otherwise obtain this relevant factual information (or its substantial equivalent) absent Facebook’s disclosure. Importantly, the court observed that “[t]his is not a case where the internal investigation involved simply interviewing key employees and other witnesses or reviewing a manageable number of documents.”
Wineberg: The ACC brief emphasized how the decision of the lower court would constrain certain regulated organizations, such as hospitals, to conduct internal investigations. What impact do you believe these arguments may have had on the court’s decision?
Bean: While we cannot know the impact the brief had, we were pleased to give voice to the views of the in-house bar, including important regulated nonprofit industries such as hospitals. We were pleased that the decision affirmed many of the interests of in-house counsel as a whole and provided guidance to improve in-house practice.
ACC Docket: Danette, thank you for helping us bring the salient points of the Facebook decision to our readers. Tell us more about your in-house background and what you enjoy most about teaching.
Wineberg: I had not heard much about in-house practice when I went to law school. I began my career in a law firm but had an opportunity to move in-house at a company close to where I lived. Ultimately that led me to serve several companies, the most recent as vice president, general counsel, and secretary to The Timberland Company, prior to its sale. I have been a member of the global board of ACC and of the ACC Northwest Chapter. Currently, I chair the Advocacy Committee for the chapter, and I also teach an in-house counsel course at the University of New Hampshire School of Law, where I also work in the Academic Success Program. I love working with law students and having the opportunity to teach about the unique aspects of in-house practice because I feel it is an exciting, interesting, and challenging career, and I feel so fortunate to have been part of the in-house bar.