Ninth Circuit Reverses Excessive Fee Cases for Salesforce and Trader Joe’s

Nearly every court that has decided a motion to dismiss in an excessive fee case following the January 26 Northwestern Supreme Court opinion has allowed the complaint to proceed to discovery.  A North Carolina court in the Shoe Show case went so far as to rule that it is an issue of fact as to whether a $40 million plan has sufficient bargaining leverage to reduce recordkeeping fees.  And now the Ninth Circuit piles on by reversing both the Salesforce and Trader Joe’s dismissals of excessive fees cases.

The pleading standard continues to be diluted, as courts are subjecting quality plan fiduciaries to unwarranted liability risk and litigation expense.  Legitimate and illegitimate cases are being allowed to proceed to discovery without any way to distinguish low-cost quality plans from higher-cost plans.  Under this judicial regime, all fiduciaries are guilty until proven innocent, and have to incur significant cost and litigation risk to prove their fiduciary process was prudent.  In the Trader Joe’s and Salesforce cases, fiduciaries have significant potential liability if they use revenue sharing to pay recordkeeping fees, as the appellate court stripped any ability to defend the practice at the pleadings stage – even though Trader Joe’s had demonstrated on the record that all revenue sharing not used for plan expenses was returned to participants.

The Trader Joe’s Excessive Fee Cases

Trader Joe’s Company has been harassed by plaintiff lawyers regarding its defined contribution plan for well over two years in two separate lawsuits.  The first case was filed on December 30, 2019 by an inexperienced law firm in excessive fee cases.  The plan was alleged to have $1.75B in assets, covering 35,474 employees at the end of 2018.  The two plaintiffs were former employees with limited tenures – one had less than one year with the company, and the other with less than two years.  Both plaintiffs likely had minimal amounts in the plan.  They likely paid less than $100 in plan expenses in total, but that perspective is usually lost in these cases.  The crux of the complaint was that the plan’s recordkeeper Capital Research charged millions of dollars in excessive plan administration fees that totaled “roughly $140 per participant” when a “reasonable recordkeeping fee would be $40 per plan participant.”  There was no evidence for this claim, and no justification for the $40 purported benchmark.  Like many excessive fee complaints, the facts are often inaccurate or misrepresentations.  In its motion to dismiss, lawyers for the plans asked for judicial notice of the Capital recordkeeping contract, which charged a base fee of $11,650 plus $48 per participant.  This showed that the actual recordkeeping fees were $48-49 per participant – nowhere close to the fabricated $140 claim.  The district court dismissed the lawsuit as implausible and granted leave to amend, but plaintiffs appear to have abandoned the case.

The more experienced Capozzi Adler law firm quickly jumped in with a new excessive fee complaint and new former employees as plaintiffs.  The First Amended Complaint more fairly discloses the actual recordkeeping schedule, which charges exactly what the defense had informed the court in the prior case:  $11,650 + $48 per participant when the plan has at least 1,000 participants in the plan.  But plaintiffs still argued that the $48 amount was unreasonable because it “boxed” in the plan to pay the same amount per participant even though the number of participants continued to grow from 25,000 to over 35,000.  Plaintiffs went a step further by challenging the revenue sharing agreement to fund the recordkeeping contract.  Specifically, plaintiffs calculated the revenue sharing amount that represented the differential between the retail share fee compared to the institutional share class fee for the same funds and alleged $30.5m in revenue sharing.  For example, $1.1B of the plan was invested in the American Funds Balanced Fund with the R4 revenue share class at 0.63% compared to the R6 Institutional Share Class at 0.28%.  The following two paragraphs are copied directly from the complaint:

111. To the  extent  Defendants  chose  the  higher  cost  shares  to  pay  for recordkeeping, which for purposes of this First Amended Complaint is assumed to be $48 per participant, the math doesn’t add up:

112. That means the Plan’s fiduciaries collected $30 million from Plan participants to pay for $7.5 million worth of recordkeeping That is a whopping discrepancy of $23 million. (emphasis added by Euclid).

The purported “whopping discrepancy” is only hyperbole designed to create an issue of fact to survive a motion to dismiss, because the sophisticated Capozzi lawyers know that the plan rebates excess revenue sharing to participants.  Indeed, the next section of the complaint summarizes the revenue sharing recapture account in which $2.2m to $3.2m was transferred annually.  Plaintiffs do not acknowledge the math, because it does not serve their interests, but this is the amount of the revenue share above the $48 expense per participant.  Plaintiffs criticize that the plan did not completely empty the revenue recapture account each year, keeping $1.2 to $1.8m in the account at any one time, and thus they allege that participants suffered lost opportunity costs.  But this is far from a “whopping discrepancy” or math “that doesn’t add up,” and fully explains the differential.  They may have had an argument that additional plan expenses were paid out the revenue recapture account, but they instead alleged the more disingenuous $23 million “discrepancy.”

To the extent that there is any doubt as to whether there is a $23 million “discrepancy,” however, the defense asserted in the motion to dismiss by attorney affidavit that:

“As the Plan’s recordkeeping agreement makes plain, the Fund Revenue deposited in the recapture account is used to pay Plan expenses, including Capital Research’s negotiated fees, and the remainder is returned to Plan participants.” Vergara Decl.  Ex. 1 at 7 (emphasis added by Euclid).

There are thus no losses or excessive fees from using the retail share classes, because all revenue sharing that is not used for legitimate plan expenses is returned to plan participants.  The complaint is intentionally designed to confuse the issue and create a false of issue of fact.

The lower court, however, does not appear to have been confused by the complaint.  The district court dismissed the amended complaint on all counts, including the claim that the plan paid excessive fees by using the retail share class, because the alleged additional fees were explained by the revenue sharing recapture arrangement described above.  But on appeal, the Ninth Circuit Court of Appeals reversed.  The appellate court held that the complaint plausibly alleged that the plan fiduciaries failed to monitor and control the offering of a number of mutual funds in the form of “retail” share classes that carried higher fees than those charged by otherwise identical “institutional” share classes of the same investments.  The court accepted as true that “the choice resulted in more than $30,464,538 in extra fees.”  The court further held that the plan’s “explanation for the more expensive choice is unavailing at the pleading stage” because “[t]hough the parties signed a revenue sharing agreement that might provide some explanation for this choice, the agreement shows only what could occur in theory—not what occurred in fact.”  (emphasis added by Euclid).

If you follow our summary of the complaint and motion to dismiss, the appellate court does not appear to have understood the revenue sharing contract or the factual allegations.  The other explanation is that plaintiffs have successfully confused the court with the misleading claims.  There is no hypothetical revenue rebate to participants “that might provide some explanation” – it is the only truthful explanation.  To recount, the complaint itself identified that $7.5m of the $30.5m in revenue sharing went to pay the contractual $48 per participant recordkeeping charge.  Then the question is what happens to the remaining $23 million – the excess revenue sharing than was necessary to meet the contractual $48 per participant recordkeeping fee.  This question was answered in the lead lawyer’s affidavit filed under penalty of perjury, and her answer was that “the remainder is returned to plan participants” after paying plan expenses.  So the court is plain wrong when it says that the disputed amount is $30 million, and is also wrong that the revenue sharing offset is just a theoretical argument and not what happened in actual fact.  The appellate court has created a dispute of material fact when there was none.

A comparison of the Salesforce and Trader Joe’s excessive fee cases demonstrate that the cases are being treated the same, even if the investment and fee structures are different.  From the decisions, you would assume that both cases had the same revenue sharing agreement – that both plans had retail share classes with revenue sharing that increased plan expenses as the plans grew in asset size.  The Ninth Circuit in both cases held that claims of excess fees from not using lower-cost institutional share classes constitute plausible claims of fiduciary imprudence, and cannot be explained or justified by revenue sharing at the pleading stage.  But based on our reading of the pleadings in both cases, the revenue sharing agreements were not the same in both plans.  To the contrary, the Trader Joe’s revenue sharing agreement captured all revenue sharing fees above a set amount to the plan, which was then rebated back to participants.  The Salesforce motion to dismiss argued that the revenue sharing explained the difference, but did not give the same specificity or revenue recapture evidence as the Trader Joe’s motion to dismiss.  While there might be a factual dispute in the Salesforce case, there is no factual issue to debate or justify reversal to the district court in the Trader Joe’s case.

The Euclid Perspective

We do not pretend to like revenue sharing or indirect recordkeeping compensation.  But it is a long-standing and legitimate way to pay plan administration costs.  The Department of Labor has never declared that it is imprudent to use revenue sharing.  More importantly, revenue sharing does not harm plan participants if the excess investment fees are returned to participants who paid the investment fees, which is what happened in the Trader Joe’s plan.  It was not apparent from the pleadings that the revenue sharing in the Salesforce plan was capped or returned to participants.  So, it is a more plausible complaint to allege that participants potentially paid higher investment fees than necessary when the plan utilized a higher share class for the JP Morgan target-date funds.  But the undisputed record in the Trader Joe’s case was that the (1) recordkeeping fee was capped at $48 per participant; and (2) the fee differential between the retail and institutional share classes for the Trader Joe’s plan was returned to participants after paying plan expenses.  Nevertheless, the Trader Joe’s fiduciaries have been held to a standard in which they are automatically at fiduciary liability risk for using revenue sharing, even if the contract guaranteed that plan participants were protected.  They will have to bear the continued expense of discovery and the burden of litigation risk and potential liability.  If the Ninth Circuit’s rulings are followed by other courts, any plan using revenue sharing will have to defend an expensive fiduciary malpractice case before a judiciary ill-equipped to understand the complex arrangement.  And given that very few cases go to trial or final disposition, both companies will be pressured to settle in order to avoid liability – all because their motions to dismiss were reversed.

As long as the Department of Labor allows the second-guessing of plan fiduciaries to continue, fiduciary malpractice is being evaluated by judges with no requisite experience and they are defaulting to a low pleading standard.  The problem is that plans with low fees and good processes are treated the same as plans with high fees.  In ninety percent of the cases, there is no way for quality plan fiduciaries to dismiss a meritless case.  We continue to fault defense firms for defending every case – whether against a high-fee or low-fee plan – the very same way, as it is contributing to the confusion in the federal courts.  But courts have abdicated their responsibility to vet implausible cases and protect good fiduciaries.  The Department of Labor could restore some justice, but has taken no steps to articulate a fair standard.  The only guidance they have provided is commentary that the high-fee Northwestern plan may have imprudent investments and recordkeeping fees.  But they have given no guidance to protect quality plans with low fees.  That leaves only Congress to amend ERISA.  Consequently, it just does not pay to serve as a plan fiduciary in the current unfair environment.