In an emerging theory of liability, plan fiduciaries’ treatment of participants’ personal data is coming under scrutiny. Over the last five years, we have seen how the collection of many individuals’ personal data can become a valuable asset in the right hands — whether it’s used to influence an election, design a marketing plan that targets individuals based on their specific preferences and needs, or just to compile large troves of information to analyze trends. See “The World’s Most Valuable Resource Is No Longer Oil, But Data,” The Economist (May 6, 2017). Participants’ awareness of, and concerns regarding, the collection, use, sale, and transfer of personal data is evolving. In today’s world, where data is considered as valuable as other commodities, it is not surprising that the way plan fiduciaries look at protecting participants’ personal data is changing. What is surprising is that Employee Retirement Income Security Act of 1974 (ERISA) litigation is one new avenue being used to try to force plan fiduciaries to protect participants’ data.
As part of a broader wave of “excessive fee” lawsuits involving 401(k) and 403(b) plans, three lawsuits were filed against prominent universities involving, among other aspects, claims concerning the use of participant data: New York University (“NYU 2”)1; Northwestern University2; and Vanderbilt University. In another three 403(b) excessive fee cases — against John Hopkins University, MIT, and Emory University — participant data cross-selling restrictions were included in the settlements, even though claims regarding cross-selling were not raised in the complaints.3 At first, participant data claims seemed limited to 403(b) plans, but in 2020, that litigation expanded to include two 401(k) plans sponsored by ADP TotalSource Group4 and Shell Oil Company. These are referred to below collectively as the “Participant Data Cases.”
In the Participant Data Cases, participant plaintiffs allege that third-party administrators and recordkeepers are using participants’ personal data to cross-sell profitable non-plan products to plan participants. Such personal data includes:
- Identifying Information: i.e., participant’s name, contact information, social security number, date of birth, marital status, phone numbers, and work and personal email addresses;
- Financial Information: i.e., income levels, contribution history, account balance, and expected retirement age; and
- investment Preferences: i.e., investment histories and investment holdings.
The plaintiffs allege that recordkeepers are soliciting participants to purchase expensive and lower rate of return non-plan products such as Individual Retirement Accounts (IRAs) and Individual Retirement Annuities, high-interest credit cards, life insurance, banking products, advisory accounts, individual brokerage accounts, and options trading accounts.
The plaintiffs in these Participant Data Cases allege that plan fiduciaries (1) breached their fiduciary duty and (2) allowed prohibited transactions to occur when they did not prevent recordkeepers from using participants’ personal data to cross-sell non-plan products. Under ERISA’s strict fiduciary standards, selecting and monitoring plan service providers, such as recordkeepers for a plan, is a fiduciary function. Fiduciaries must act for the exclusive purpose of providing benefits to participants and their beneficiaries, and defraying the reasonable expenses of administering the plan. In an action against plan fiduciaries alleging a breach of fiduciary duty concerning plan assets, fiduciaries can be personally liable, and any recovery may be for the benefit of the entire plan.
Participant Data Case plaintiffs are seeking restitution on behalf of their plans for allegedly unjust profits that recordkeepers earned using participant data; or alternatively, they seek a surcharge against the fiduciaries for the value of the participant data that recordkeepers used. Although plaintiffs are seeking relief on behalf of the plan generally, these are defined contribution plans — so any recovery would likely be allocated into each participant’s individual account. The plaintiffs are also seeking injunctive relief to prevent future use of participant data for cross-selling purposes.
This article addresses: the general participant data claims that have been raised in defined contribution plan excessive fee complaints; the Seventh Circuit’s decision in Divane v. Northwestern University; the more sophisticated arguments being raised after Northwestern University, particularly in the Shell Oil Company case; and the trend in some Participant Data Case settlements to include cross-selling restrictions.
I. Participant Data Claims Generally
In the Participant Data Cases, the plaintiffs consistently allege that participant data is a plan asset and that defendants allowed recordkeepers to use highly confidential personal information of participants and retirees to sell the recordkeepers’ investment and wealth management products. The plaintiffs argue that defendants breached their fiduciary duty and caused the plan to engage in prohibited transactions with the recordkeepers by (1) enabling recordkeepers to profit from their role as plan service providers (outside of the fees negotiated in the service agreements) and (2) failing to protect valuable plan assets (the participants’ data). To support the second argument, participants hinge their case on one core underlying premise: that participant data is a plan asset that plan fiduciaries have an obligation to protect under ERISA.
Participant data claims first arose in 2017, after a non-ERISA whistleblower complaint filed with the SEC raised concerns about recordkeepers’ use of participant data in ERISA plans. The whistleblower alleged that a recordkeeper for many 403(b) plans, TIAA-CREF, used participants’ data to engage in allegedly abusive practices to solicit participants‘ purchase of its own more expensive non-plan products. According to the whistleblower, recordkeeper- affiliated financial planners would use scare tactics during educational opportunities to try to sell more expensive non-plan products to participants of TIAA-CREF’s existing retirement plan clients. In response to the whistleblower complaint, in 2019, TIAA-CREF did an internal review, updated all of their training materials and settled with the SEC regarding the allegations. Without acknowledging fault, TIAA-CREF agreed to (1) correct necessary disclosures, (2) evaluate whether clients should be moved to lower-cost share classes, and (3) review their policies and procedures regarding disclosures for their mutual fund class selection. In the Participant Data Cases filed under ERISA, we have seen plaintiffs cite to articles referencing this whistleblower complaint, and the alleged predatory practices, to support their participant data claims.
So far, no courts have accepted the legal theory that participant data is a plan asset under ERISA — however, several pending cases may determine the ultimate outcome of this new theory of liability.
II. Divane v. Northwestern University
In Divane v. Northwestern University, 2018 WL 2388118 (N.D. Ill. May 25, 2018), aff’d, 953 F.3d 980 (7th Cir. 2020), petition for cert. filed, (U.S. Jun. 19, 2020) (No. 18-2569), the district court became the first court to rule on the participant data theory of liability. The district court dismissed the plaintiffs’ first amended complaint and denied their request for leave to file a second amended complaint which included new participant data allegations. The parties briefed the issue, and the district court found that the plan fiduciaries did not breach their fiduciary duty by allowing recordkeepers to have access to participants’ confidential information, which is required to perform necessary recordkeeping functions. The district court noted that the plaintiffs failed to “cite a single case in which a court has held that releasing confidential information or allowing someone to use confidential information constitutes a breach of fiduciary duty under ERISA.” Northwestern Univ., 2018 WL 2388118, at *12. Nor do plaintiffs provide any support that participant data is a plan asset in a prohibited transaction. While participant data does have some value, it is not a plan asset under “ordinary notions of property rights.” Id. Finding the plaintiffs’ arguments with respect to participant data posed too abstract an injury for standing purposes, the district court denied the request for leave to file the proposed second amended complaint.
The Seventh Circuit affirmed the district court’s order dismissing the action. Without specifically addressing the participant data claims, the Seventh Circuit determined leave to amend was futile as all the new claims in the second amended complaint — including the participant data claims — were essentially the same claims in different counts, and therefore improperly pled. The plaintiffs filed a petition for certiorari to the U.S. Supreme Court challenging the dismissal of the first amended complaint and the decision to deny leave to amend, which is pending.5
III. Harmon v. Shell Oil Company
A year to the month after the Seventh Circuit decided the Northwestern University 403(b) plan case, participant data allegations were front and center in a 401(k) plan excessive fee case, Harmon v. Shell Oil Company, et al., No. 3:20-cv-00021, 2021 WL 1232694 (S.D. Tx. Mar. 30, 2021), where another district court rejected the legal theory that participant data is a plan asset. In Shell Oil Company, unlike the other Participant Data Cases, the plaintiffs included allegations against the plan recordkeeper, Fidelity, as a co-defendant. All allegations against the recordkeeper were premised on the legal theory that participant data is a plan asset. Ordinarily, recordkeepers do not exercise discretion over plan assets and are not plan fiduciaries. In Shell Oil Company, the plaintiffs alleged that the recordkeeper was a plan fiduciary because it had control over the participants’ data, which was alleged to be a plan asset. When the recordkeeper brought a motion to dismiss, the entire focus was on the participant data claims. With the issue of whether participant data is a plan asset now determinative of the outcome for the recordkeeper, it raised novel arguments against this legal theory that had not been addressed in the prior Participant Data Cases.
First, the recordkeeper argued that the plaintiffs failed to establish Constitutional standing under Article III because they lacked an injury in fact. The recordkeeper asserted that the plaintiffs failed to allege that plan participants actually transferred assets out of the plan to their detriment on the basis of the recordkeeper’s cross-selling solicitations, as opposed to other reasons. The recordkeeper also argued that simply soliciting participants a few times was insufficient to establish an injury. The plaintiffs countered that at least one named plaintiff rolled money out of the plan based on the recordkeeper’s solicitation and was injured because the IRA into which he rolled money charged higher administration costs than the amounts charged to his account in the plan. The recordkeeper responded to that argument by pointing out that the amended complaint is devoid of actual comparisons showing that the named plaintiff’s plan funds were rolled into more expensive IRA funds than the plan’s investments.
Second, the recordkeeper asserted that ERISA’s statutory framework requires that plan assets be held in trust for the exclusive benefit of plan participants and beneficiaries, which it argues is not practical when applied to phone numbers, e-mail addresses, and investment history. Plan participants have access to and may disseminate this information outside of the plan, including when they go to a competitor for financial products. The recordkeeper also releases this information in the aggregate to data collection agencies that collect information to help advise plan fiduciaries.
The plaintiffs countered that it is the compilation of each participant’s data into a comprehensive financial picture, including the participant’s personal data, call notes, information on major life events, investment history, and goal retirement dates, that they are referring to as a plan asset. In this way, the plaintiffs argued, participants can still use their personal information, and it does not affect the fiduciaries’ exclusive control of a plan asset (the compilation of each participant’s data that is held only by the recordkeepers).
Third, the recordkeeper argued that the plaintiffs’ theory of participant data as a plan asset is unworkable under ERISA. There are two regulations defining plan assets under ERISA, 29 C.F.R. Section 2510.3-101 (defining plan assets in the context of plan investments) and 29 C.F.R. Section 2510.3–102 (defining plan assets in the context of participant contributions). No regulatory body has ever found that participant data is a plan asset. Also, the DOL allows plans to file a Form 5500-SF if the plans’ assets can be readily valued. The recordkeeper argued that if participant data is a plan asset, it would be difficult for any plans to place a value on such an asset and, as a result, no plan would be able to file a Form 5500-SF, making this form superfluous. Further, a finding that participant data is a plan asset would affect all plans, not just ERISA defined contribution plans. The recordkeeper went on to argue that if the case progresses past the motion to dismiss phase, it will have to use participant data in its defense — which would be a breach of fiduciary duty in and of itself if participant data is a plan asset.
Finally, the recordkeeper argued that the weight of legal precedent shows that participant data cannot be a plan asset, relying on Northwestern University (7th Cir. 2020) (discussed above), and two other ERISA cases decided in another context: Patient Advocates, LLC v. Prysunka, 316 F. Supp. 2d 46 (D. Me. 2004) (an ERISA preemption case involving a state statute requiring disclosure of health plan participants’ data), and Walsh v. Principal Life Insurance, 266 F.R.D. 232 (S.D. Iowa 2010) (finding recordkeepers could access and use participant data to send letters soliciting retail products, but the case did not address whether participant data is a plan asset).
The plaintiffs countered these arguments by asserting that they were inconsistent with the recordkeeper’s position in other litigation against former employees of the recordkeeper and in internal memos of the recordkeeper that referred to customer information as being the recordkeeper’s proprietary information that was as valuable as “the formula of Coke to Coca-Cola.” Armed with the recordkeeper’s analogy of the value of participant data, the plaintiffs attempted to rebut the legal precedent cited by the recordkeeper by referencing decades of SEC and insurance brokerage cases that treat customer data as an asset. The plaintiffs alleged the recordkeeper went to great lengths in internal memos, policies, and litigation to prevent competitors from using participants’ data collected by the recordkeeper to sell competing products. The plaintiffs further alleged that recordkeepers who cross-sell non-plan products consider participants’ data as their own proprietary information, even though they only have access to this information by virtue of their position as a recordkeeper to the plan.
The district court in Shell Oil Company found that participant data is not a plan asset. In a succinct and well-reasoned opinion, the court focused on two main questions: (1) Have any other courts found that participant data is a plan asset? and (2) Is participant data an asset ERISA was designed to protect? In oral arguments, the plaintiffs conceded that no courts have found participant data to be a plan asset. The Shell Oil Company district court, like the district court in Northwestern University, declined to be the first court to make such a finding — noting the three prior ERISA cases that expressly contradict plaintiffs’ arguments that participant data is a plan asset.4 As to the second question, the Shell Company court focused on ERISA’s statutory language which states “plan assets [are] defined by such regulations as the Secretary [of Labor] may prescribe.” (citing 29 U.S.C. § 1002(42)). The court noted that there are no regulations that describe participant data as a plan asset, and the only two regulations that define plan assets do so in the context of investments and contributions. Finding that participant data is not a plan asset, the court held that the recordkeeper was not a fiduciary and had not engaged in prohibited transactions by using participant data for profit, and dismissed all claims against the recordkeeper. The next day, the district court also dismissed the participant data claims against Shell Oil Company and the plan’s trustees for the same reasons.
IV. Settlement Agreements Restricting Recordkeepers’ Use of Participant Data
While the issue of using participant data to cross-sell non-plan products is still developing in the courts, settlements in four of the 403(b) excessive fee cases, Emory, John Hopkins, MIT, and Vanderbilt, included provisions limiting recordkeepers’ use of participant data for cross-selling purposes. In all four settlements, the recordkeeper is permitted to use participant data in situations where the participant initiates a conversation about the recordkeeper’s other products.
For example, in the case involving Emory University’s 403(b) plan, the settlement requires that Emory University prohibit recordkeepers from:
Us[ing] information received as a result of providing services to the Plans and/or the Plans’ participants to solicit the Plans’ current participants for the purpose of cross-selling non-Plan products and services, including, but not limited to, Individual Retirement Accounts (‘IRAs‘), non-Plan managed account services, life or disability insurance, investment products, and wealth management services, unless in response to a request by a Plan participant.
Similar restrictions are mirrored in the other settlements referencing participant data. These settlement agreements are limited to the individual cases that settled and don’t necessarily lend support for plaintiffs’ participant data allegations generally; however, it is advisable for plan fiduciaries to be aware of provisions that protect participant data in other plans’ recordkeeping agreements so that they can decide whether to include such provisions in their own recordkeeping contracts.
It remains to be seen how the courts may further rule on the issue of whether participant data is a plan asset under ERISA. So far, one district court in the Seventh Circuit and one district court in the Fifth Circuit have found that participant data is not a plan asset. The Seventh Circuit affirmed the Northwestern University decision without directly addressing the participant data claims, and the plaintiffs are likely to appeal to the Fifth Circuit the Shell Oil Company decision granting Fidelity’s motion to dismiss. There are at least two cases currently pending (one 403(b) case and one 401(k) case) that allege participant data is a plan asset. Only time will tell whether similar allegations will be made in future excessive fee litigation. With the current legal landscape, plaintiffs will likely face an uphill battle with the participant data claims. While this issue plays out further in the courts, we recommend that plan fiduciaries review their own recordkeeping agreements and consider adding cross-selling restrictions if they want to preclude their recordkeepers from using participant data for cross-selling purposes in the future. Even if participant data isn’t a plan asset under ERISA, this litigation raises interesting questions about recordkeepers cross-selling non-plan products to plan participants.
1 As of April 12, 2021, the case, Sacerdote v. Cammack LaRhette Advisors, LLC, No. 17 cv 8834 (S.D.N.Y.) (NYU 2) is pending, but is stayed until the Second Circuit decides Sacerdote v. New York University (NYU 1) No. 18-2707 (2d Cir.) (a related case that does not involve participant data claims).
2 Participant data allegations were only in the proposed second amended complaint, which the district court denied leave to file.
3 Also, although not raised as a claim in the complaint, plaintiffs in the Yale 403(b) excessive fee case made
participant data and cross-selling an issue when their proposed expert included revenue earned from cross-selling in his calculations. Vellali v. Yale University, No. 3:16-cv-01345 (D. Conn. Dec. 4, 2020) (Dkt. 272).
4 As of April 12, 2021, there are motions to dismiss pending in this case, Berkelhammer v. ADP TotalSource Group, Inc., No. 2:20-cv-05696 (D.N.J.) (oral arguments scheduled June 2, 2021).
5 As of April 15, 2021. See also, Hughes v. Northwestern Univ., 141 S.Ct. 231 (2020) (inviting Acting Solicitor General to file a brief expressing the views of the United States).
6 Northwestern University, 2018 WL 23p88118; Walsh, 266 F.R.D. 232; Patient Advocates, LLC, 316 F. Supp. 2d 46.