1 2 3 4 IN THE UNITED STATES DISTRICT COURT 5 FOR THE NORTHERN DISTRICT OF CALIFORNIA 6 7 GREGOR MIGUEL, et al., Case No. 20-cv-01753-MMC

8 Plaintiffs, ORDER DENYING DEFENDANTS' 9 v. MOTION FOR SUMMARY JUDGMENT

10 SALESFORCE.COM, INC., et al., 11 Defendants.

12 13 Before the Court is defendants’ Salesforce.com, Inc. (“Salesforce”), Board of 14 Directors of Salesforce (“Board”), Marc Benioff (“Benioff”), The Investment Advisory 15 Committee (“Committee”), Joseph Allanson (“Allanson”), Stan Dunlap (“Dunlap”), and 16 Joachim Wettermark’s (“Wettermark”) “Motion for Summary Judgment” (“Defs.’ Mot.”) 17 filed November 13, 2023. (See Doc. No. 112.) Plaintiffs have filed opposition, to which 18 defendants have replied. Having read and considered the papers filed in support of and 19 in opposition to the motion, the Court rules as follows.1 20 BACKGROUND 21 Plaintiffs are former Salesforce employees who participated in the Salesforce 22 401(k) Plan (“the Plan”). (See First Am. Compl. ¶¶ 20–23, Doc. No. 38.) In the operative 23 complaint, the First Amended Complaint (“FAC”), plaintiffs allege defendants breached 24 their fiduciary duties to the Plan and Plan participants in violation of the Employee 25 Retirement Income Security Act of 1974 (“ERISA”),

et seq. (See 26 FAC ¶ 10.) Plaintiffs allege the Committee, Allanson, Dunlap, and Wettermark 27 1 (collectively, “Committee Defendants”) breached their fiduciary duty of prudence by 2 selecting and retaining investment options with high costs relative to other, comparable 3 investments. (See

.) 4 Specifically, plaintiffs fault the Committee Defendants for: (1) offering only 5 JPMorgan’s “Institutional” share class of Target Date Retirement Funds (“TDFs”) rather 6 than the lower cost “R5” or “R6” share class of TDFs (see

); (2) retaining nine 7 actively managed JPMorgan SmartRetirement funds rather than passively managed 8 collective investment trusts (“Passive Blend CITs”) (see

); and (3) failing to 9 substitute the Fidelity mutual funds with Fidelity collective investment trusts (“CITs”) (see 10

). Plaintiffs also allege the Board, Salesforce, and Benioff (collectively, “Monitoring 11 Defendants”) breached their fiduciary monitoring duty by failing to adequately monitor the 12 Committee Defendants. (See

¶¶ 135–138.) 13 LEGAL STANDARD 14 Pursuant to Rule 56 of the Federal Rules of Civil Procedure, a “court shall grant 15 summary judgment if the movant shows that there is no genuine issue as to any material 16 fact and that the movant is entitled to judgment as a matter of law.” See Fed. R. Civ. P. 17 56(a). 18 The Supreme Court’s 1986 “trilogy” of Celotex Corp. v. Catrett,

(1986), Anderson v. Liberty Lobby, Inc.,

(1986), and Matsushita Elec. 20 Indus. Co. v. Zenith Radio Corp.,

(1986), requires that a party seeking 21 summary judgment show the absence of a genuine issue of material fact. The moving 22 party need not “produce evidence showing the absence of a genuine issue of material 23 fact,” but may discharge its burden simply by pointing out “that there is an absence of 24 evidence to support the nonmoving party's case.” See Celotex,

. Once 25 the moving party has done so, the nonmoving party must “go beyond the pleadings and 26 by [its] own affidavits, or by the depositions, answers to interrogatories, and admissions 27 on file, designate specific facts showing that there is a genuine issue for trial.” See

at 1 burden under Rule 56[ ], its opponent must do more than simply show that there is some 2 metaphysical doubt as to the material facts.” Matsushita,

. “If the 3 [opposing party's] evidence is merely colorable, or is not significantly probative, summary 4 judgment may be granted.” Anderson,

(citations omitted). 5 “[I]nferences to be drawn from the underlying facts,” however, “must be viewed in the 6 light most favorable to the party opposing the motion.” See Matsushita,

7 (internal quotation and citation omitted). ANALYSIS 8 9 An ERISA fiduciary has a statutory duty to act for the exclusive purposes of 10 “providing benefits to participants and their beneficiaries” and “defraying reasonable 11 expenses.” See

(a)(1). A fiduciary must “discharge his responsibility 12 ‘with the care, skill, prudence, and diligence’ that a prudent person ‘acting in a like 13 capacity and familiar with such matters’ would use,” see Tibble v. Edison Intern., 575

, 528 (2015) (quoting

(a)(1)). 15 By the instant motion, defendants argue “[t]he undisputed evidence makes clear 16 17 that [p]laintiffs’ prudence claims are meritless, and that [d]efendants are entitled to 18 summary judgment.” (See Defs.’ Mot. at 2:15–16.) 19 I. Breach of Fiduciary Duty of Prudence 20 Defendants argue they are entitled to summary judgment on plaintiffs’ breach of 21 fiduciary duty claim because the record does not show evidence of: (1) defendants’ 22 failure to select the lowest-cost share of JPMorgan Target Date Retirement Funds 23 (“TDFs”); (2) the reasonableness of defendants’ decisions not to replace the JPMorgan 24 25 TDFs and Fidelity Mutual Funds with plaintiffs’ preferred collective investment trusts 26 (“CITs”) before July 19, 2019; (3) damages from the alleged breaches; and (4) 27 deficiencies in defendants’ monitoring process. (See Defs.’ Mot. at 17:3–13.) A. Failure to Select Lowest-Cost JPMorgan SmartRetirement TDFs 1 2 As set forth above, plaintiffs allege defendants “knew or should have known of the 3 existence of cheaper share classes” of JPMorgan’s SmartRetirement Target Date Funds 4 (“TDFs”) and “failed to choose or switch to the R5 share class from 2014 to 2018 and 5 again failed to choose or switch to the R6 class from 2015 to 2018.” (See FAC ¶¶ 11, 78, 6 80.) 7 At the outset, defendants note that the record shows the “Institutional” share class 8 was renamed the “R5” share class as of April 3, 2017, and was offered by the Plan 9 10 throughout the class period. (See Decl. of Eric Serron in Supp. of Defs.’ Mot. for Summ. 11 J. (“Serron Decl.”) Ex. 84 (JPMorgan SmartRetirement Funds, Supplement to Summary 12 Prospectus, Doc. No. 113-28) (providing “[e]ffective on or about April 3, 2017, each 13 Fund’s Institutional Shares will be redesignated as and renamed Class R5 Shares”).) 14 In response, plaintiffs assert that the “heart” of their claim is defendants’ failure to 15 timely switch from the Institutional/R5 to the R6 share class of TDFs and submit evidence 16 that the R6 class had a lower expense ratio2 than the Institutional/R5 class. (See Pls.’ 17 18 Opp’n. to Defs.’ Mot. for Summ. J. (“Pls.’ Opp’n.”) at 6:11–23 citing Serron Decl. Ex. 84 19 (Expert Report of Robert E. Conner, Doc. No. 113-23) ¶ 37; see also Serron Decl. Ex. 86 20 (JPMorgan SmartRetirement 2040 Fund, Summary Prospectus November 1, 2016, Doc. 21 No. 113-30) (showing lower “Total Annual Fund Operating Expenses” in R6 class than 22 Institutional/R5 class).) Although defendants concede the R6 class had a “nominally 23 lower expense ratio,” they submit evidence that the R5 class paid 15 basis points of 24 25 26 2 A fund’s “expense ratio” is “[t]he total percentage of a fund’s assets that goes 27 towards paying annual recurring expenses.” (See Serron Decl. Ex. 37 (Expert Report of revenue sharing,3 which defendants argue rendered it “cheaper on a net cost basis than 1 2 the R6 class.” (See Defs.’ Mot. at 9:1–7, citing Heavner Rpt. ¶ 29.)4 3 Plaintiffs, however, have submitted testimony from their expert, Robert Conner,5 4 that the revenue sharing paid by the R6 class does not directly offset the expense ratio, 5 because, in Conner’s opinion, revenue sharing “reduces the investment returns plan 6 participants receive” and “makes recordkeeping and administrative costs interdependent 7 with plan participant returns.” (See Serron Decl. Ex. 74 (Rebuttal Expert Report of Robert 8 E. Conner (“Conner Rebuttal Rpt.”), Doc. No. 113-22) ¶ 20; see also Serron Decl. Ex. 36 9 10 (Deposition Transcript of Robert Conner (“Conner Dep.”), Doc. No. 112-36) at 188:13–17 11 (stating money for revenue sharing is “coming from the participants as paying expenses 12 for the fund”).) 13 Given the above, the Court finds plaintiffs have produced evidence sufficient to 14 raise a triable issue as to the prudence of defendants’ decision to retain the 15 Institutional/R5 share class of the JPMorgan TDFs until September 13, 2017. 16 B. Failure to Substitute Collective Investment Trusts 17 18 Defendants next argue the evidence does not show their failure to replace the 19 JPMorgan TDFs and the Fidelity Contrafund and Diversified International Fund with 20 21 3 In “revenue sharing” arrangements, “a portion of [a fund’s] expense ratio [is] used 22 to cover plan-related costs like recordkeeping and administrative expenses.” (See Heavner Rpt. ¶ 20.) 23 4 On September 13, 2017, defendants ultimately did switch to the R6 class (see 24 Serron Decl. Ex. 26 (Minutes of Meeting of 401(k) Committee of Salesforce.com, Inc., Sept. 13, 2017, Doc. No. 112-26) at SALESFORCE_0079170), “in connection with the 25 reduction of Fidelity’s recordkeeping fees to $37 per participant.” (See Defs.’ Mot. at 9:15–17, citing Serron Decl. Ex. 24 (2Q 2017 Review of Salesforce 401(k) Savings Plan, 26 June 30, 2017, Doc. No. 112-24) at SALESFORCE_0031877.) 27 5 By separate order filed concurrently herewith, the Court has denied defendants’ collective investment trusts (“CITs”)6 prior to July 19, 2019, was imprudent. (See FAC 1 2 ¶¶ 11–12, 78, 112; Defs.’ Mot. at 20:14–16; 22:22–24.) 3 It is undisputed that the mutual fund versions of the JPMorgan TDFs had higher 4 average expense ratios than their CIT counterparts. (See FAC ¶ 112; see also Serron 5 Decl. Ex. 32 (Target Date Fund Review, Dec. 31, 2015, Doc. No. 115) at 6 SALESFORCE_0038459 (showing average expense ratio is 65 basis points for mutual 7 funds and 59 basis points for CITs).) It is also undisputed that the Plan became eligible 8 for the CIT versions of the JPMorgan TDFs on September 30, 2015 (see id.), and that the 9 10 Committee did not authorize the switch until April 19, 2019 (see Serron Decl. Ex. 48 11 (Minutes of Meeting of 401(k) Committee of Salesforce.com, Inc., Doc. No. 112-47) at 12 SALESFORCE_0060863). 13 Similarly, it is undisputed that the Fidelity mutual funds had higher expense ratios 14 than their CIT counterparts. (See Serron Decl. Ex. 17 (3Q 2016 Review of 15 Salesforce.com, Inc. 401(k) Savings Plan, Sept. 30, 2016, Doc. No. 112-17) at 16 SALESFORCE_0004563 (noting “[f]ees for the Contrafund CIT are 0.43%” and the 17 18 Contrafund K mutual fund are “0.61% with 0.20% in revenue share”); see also Serron 19 Decl. Ex. 89 (Fidelity Transparency Report, Q3 2017, Doc. No. 113-33) (showing 0.92% 20 expense ratio with 0.20% revenue share for International fund); Serron Decl. Ex. 92 21 (Fidelity Transparency Report, Q3 2019, Doc. No. 113-36) (showing 0.58% expense ratio 22 for International CIT).) The Committee was informed the Plan qualified for the Contrafund 23 24

25 6 A CIT is an investment “administered by banks or trust companies, which assemble[s] a mix of assets such as stocks, bonds, and cash.” (See FAC ¶ 109 n.21.) 26 CITs are “[r]egulated by the Office of the Comptroller of the Currency rather than the Securities and Exchange Commission” and have “lower or no administrative costs” 27 because they have “simple disclosure requirements[] and cannot advertise or issue CIT as of September 30, 2016 (see Serron Decl. Ex. 17 at SALESFORCE_0004563), but 1 2 as with the JPMorgan TDFs, did not authorize the switch until April 19, 2019 (see Serron 3 Decl. Ex. 48, at SALESFORCE_0060862).7 4 Defendants argue the Committee’s delay in switching to lower-cost CITs was 5 “within the range of reasonable judgments” to which a prudent fiduciary is entitled. (See 6 Defs.’ Mot. at 20:14–16.) In support thereof, defendants submit evidence that such delay 7 was based on the Committee’s concerns regarding the CITs’ lack of third-party 8 evaluations, such as Morningstar ratings (see Serron Decl. Ex. 29 (Deposition Transcript 9 10 of Linda Ruiz-Zaiko (“Ruiz-Zaiko Dep.”), Doc. No. 112-29) at 71:25–72:14), the differing 11 regulatory regimes governing mutual funds and CITs (see

at 72:20–73:6), and the 12 shorter available track record for CITs, which was not in accordance with the Plan’s 13 Investment Policy Statement (“IPS”) (see Serron Decl. Ex. 49 (4Q 2015 Review of 14 Salesforce 401(k) Savings Plan, Dec. 31, 2015, Doc. No. 116-2) at 15 SALESFORCE_0004754 (noting CITs’ track records ranging from three to five years); 16 Serron Decl. Ex. 12 (2013 IPS, Doc. No. 112-12) at SALESFORCE_0060750 (requiring 17 18 “5 years of verifiable investment performance, unless specifically exempted by the 19 Committee”); Serron Decl. Ex. 13 (2016 IPS, Doc. No. 112-13) at 20 SALESFORCE_0000999 (same)). 21 Additionally, as to the JPMorgan funds, defendants note that the CIT and mutual 22 fund TDFs had different returns (see Decl. of D. Lee Heavner in Supp. of Defs.’ Opp’n. to 23 24 25 7 Neither party submits evidence as to when the Plan qualified for the International 26 CIT, but its inception date is December 13, 2013. (See Serron Decl. Ex. 59 (2Q 2020 Review of Salesforce.com, Inc. Savings Plan, Sept. 15, 2020, Doc. No. 116-6) at 27 SALESFORCE_0049353.) Pls.’ Mot. for Class Cert. at Ex. 1, Doc. No. 82-4 (showing retained funds outperforming 1 2 CITs in 2014, 2017, 2019; showing CITs outperforming retained funds in 2015, 2016, and 3 2018)), and argue such results “refute[] [p]laintiff’s claim that their underlying assets are 4 ‘substantially identical.’” (See Defs.’ Mot. at 22:14–17.) As to the Fidelity funds, 5 defendants submit evidence that the retained mutual funds offered revenue sharing, 6 which, defendants argue, offsets the cost differential between those funds and the CITs. 7 (See Serron Decl. Ex. 17 at SALESFORCE_0004563 (showing for Contrafund CIT, 8 expense ratio of 0.43%, and for Contrafund K mutual fund, expense ratio of 0.61% with 9 10 0.20% revenue share); see also Serron Decl. Ex. 89 (showing for International mutual 11 fund, expense ratio of 0.92% with 0.20% revenue share); Serron Decl. Ex. 92 (showing 12 for International CIT, expense ratio of 0.58%).) 13 In response, plaintiffs submit Conner’s opinion that a prudent fiduciary would not 14 consider either the lack of publicly available information regarding CITs or the differing 15 regulatory regimes sufficient reason to exclude CITs from consideration. (See Conner 16 Dep. at 249:17–250:20; Conner Rebuttal Rpt. ¶ 21; see also Davis v. Salesforce, No. 21- 17 18 15867,

, at *2 (9th Cir. 2022) (noting whether regulatory differences 19 justify defendants’ delay in switching to CITs is “factual issue”).) Further, to the extent 20 defendants argue the two types of funds are not comparable, plaintiffs submit evidence 21 that Bridgebay Financial, Inc., defendants’ investment consultant, itself described 22 JPMorgan’s CITs and mutual funds as having “similar investment profiles and strategies”. 23 (See Serron Decl. Ex. 44 (Minutes of Meeting of 401(k) Committee of Salesforce.com, 24 25 Inc., Nov. 16, 2018, Doc. No. 112-44) at SALESFORCE_0006050–51.) 26 Lastly, as plaintiffs point out and as discussed above, the presence of a revenue 27 sharing arrangement in the retained Fidelity funds does not, in the absence of evidence In light of the above, the Court finds plaintiffs have produced sufficient evidence to 1 2 raise a triable issue as to the defendants’ alleged imprudence in not switching to CITs at 3 an earlier point in time. 4 C. Damages 5 In countering plaintiffs’ assertion of damages in an amount totaling $5,234,737 6 (see Conner Rpt. ¶ 43), defendants argue plaintiffs “claims based on the failure to switch 7 to the R6 class of the JPMorgan TDFs, the Contrafund CIT, and the International CIT” 8 are not supported by the evidence because Heavner, after adjusting Conner’s 9 10 calculations for what Heavner describes as “errors,” determined a “total alleged ‘share 11 class’ damages [figure] of negative $1.1 million.” (See Defs.’ Mot. at 24:20–25, citing 12 Heavner Rpt. ¶¶ 39–40, Ex. 2.A (emphasis in original).) In particular, Heavner criticizes 13 Conner for using a fixed expense ratio when the actual expense ratio of the challenge 14 funds varied, as well as “ignoring” revenue sharing and “inaccurately us[ing] year-end 15 data.” (See Heavner Rpt. ¶ 39.) As discussed above, however, defendants have not 16 offered sufficient evidence that revenue sharing credits directly offset plan expenses, and 17 18 the experts’ disagreement as to the best method of calculating damages reflects a 19 dispute of material fact rather than grounds for summary judgment. 20 Lastly, defendants point out that Conner’s damages calculation does not include a 21 damages figure attributable to the failure to replace the JPMorgan TDFs with plaintiffs’ 22 preferred CITs. (See Conner Rpt. at 31.) Unlike at trial, however, at this stage of the 23 proceedings, defendants, rather than plaintiffs, have the initial burden of production. See 24 25 Nissan Fire & Marine Ins. Co. Ltd. V. Fritz Cos., Inc.,

, 1102 (9th Cir. 26 2000). For purposes of summary judgment, “a moving party without the ultimate burden 27 of persuasion of trial . . . may carry its initial burden of production” either by “produc[ing] 4 that the nonmoving party does not have enough evidence of an essential element. . . to 2 carry its ultimate burden of persuasion at trial.” See id. at 1106. Defendants have done 3 neither here. Heavner does not purport to show plaintiffs suffered no losses from 4 || defendants’ failure to substitute the JPMorgan TDFs with CITs, nor that such damages cannot be supported by the record or otherwise shown; rather, defendants only note that 6 plaintiffs have not yet offered evidence on this point. See id. at 1105 (citing with approval 7 3 Eleventh Circuit's holding in Clark v. Coats & Clark, Inc.,

(11th Cir.

9 1991) that “[e]ven after Celotex,” moving party “must point to materials on file which 10 || demonstrate that a party will not be able to meet” its burden of persuasion at trial). 11 The Court thus finds a triable issue of fact exists as to damages. 12 Il. Failure to Monitor 13 & The Court, having found a triable issue exists as to whether defendants breached

their fiduciary duty of prudence, a triable issue likewise exists as to plaintiffs’ claim for 15 failure to monitor. (See Defs.’ Mot. at 25:5 (describing claim for failure to monitor as 16 “derivative of the underlying breach claim’); Pls.’ Opp’n. at 24:14 (describing claim for Oo 17 = failure to monitor as “derivative of [p]laintiffs’ claims”).) 18 19 IT IS SO ORDERED. 20 21 . Dated: March 20, 2024 : 22 MAXINE M. CHESNEY 53 United States District Judge 24 25 26 27 28