SUSQUEHANNA INTERNATIONAL GROUP, LLP, et al., Petitioners v. SECURITIES AND EXCHANGE COMMISSION, Respondent Options Clearing Corporation, Intervenor
No. 16-1061
United States Court of Appeals, District of Columbia Circuit.
Argued March 20, 2017 Decided August 8, 2017
442
David H. Thompson, Washington, DC, argued the cause for petitioners. With him on the briefs were Howard C. Nielson Jr., Washington, DC, Peter A. Patterson, Ada, MI, and Harold S. Reeves, Washington, DC.
Robert Battalio and Robert Jennings, pro se, were on the brief for amicus curiae Robert Battalio and Robert Jennings in support of petitioners.
Tracey A. Hardin, Assistant General Counsel, Securities and Exchange Commission, argued the cause for respondent. With her on the brief were Anne K. Small, General Counsel, Sanket J. Bulsara, Deputy General Counsel, Michael A. Conley, Solicitor, and Emily T.P. Rosen, Senior Counsel.
William J. Nissen, argued the cause for intervenor. With him on the brief were Steven E. Sexton and Kristen E. Rau, Chicago, IL.
Before: GARLAND, Chief Judge, GRIFFITH, Circuit Judge, and SENTELLE, Senior Circuit Judge.
Seeking to increase its capital reserves, the Options Clearing Corporation proposed a change in its rules. That change was subject to approval by the Securities and Exchange Commission, which granted approval without itself making the findings and determinations prescribed by the
I
The Options Clearing Corporation (OCC), a Delaware corporation, is a clearing agency that facilitates trades in options and other financial instruments. It is the only clearing agency for standardized U.S. options listed on U.S. national securities exchanges. Given its significant role, OCC has been designated a systemically important financial market utility and is closely regulated by the Securities and Exchange Commission (SEC). See Order Approving Proposed Rule Change Concerning the Options Clearing Corporation‘s Capital Plan, 81 Fed. Reg. 8294, 8294 (Feb. 18, 2016) (“Order“).
At the time of the events in this case, there were twelve national securities exchanges on which listed options were traded. Five were equal shareholders in OCC; seven were nonshareholders, lacking any ownership interest. All of the exchanges clear their trades in listed options through OCC. In addition to the exchanges, OCC has “clearing members” that clear and settle options trades for their customers through the exchanges. See Order, 81 Fed. Reg. at 8294; OCC, Bylaws Art. V (amended 2009).
This case concerns OCC‘s attempt to boost its capital reserves and, in order to do so, to alter how fees and refunds are calculated. In 2014, OCC began evaluating its capital level and eventually determined that it did not have enough to cover “business, operational, and pension risks.” Order, 81 Fed. Reg. at 8296. While these capital needs exclude counterparty and on-balance-sheet risks, which are covered by billions of dollars in other funds, they are still significant. OCC determined that on top of its existing capital reserves of $25 million, it needed an additional $222 million of capital immediately on hand, plus another $117 million in backup “Replenishment Capital” that it could call upon if necessary. See Notice of Proposed Rule Change, 80 Fed. Reg. at 5172; Order, 81 Fed. Reg. at 8295-96.
To amass those reserves, OCC developed a Capital Plan. Under the Plan, OCC‘s five shareholder exchanges would make immediate capital contributions to reach OCC‘s current capital target and also pledge to provide Replenishment Capital upon request. The Plan compensates those contributions with dividends paid out of OCC‘s fees. In particular, after fees are applied to OCC‘s operating expenses, and then used to restore capital reserves if they have dipped, the remaining unused fees are split between dividends and refunds. Approximately half of the unused fees go to shareholders as dividends; approximately half are refunded to clearing members. In other words, whereas clearing members previously received all of the excess fees as refunds, the Plan diverts roughly half of those refunds to dividends. See Notice of Proposed Rule Change, 80 Fed. Reg. at 5173-75.
The Plan makes other changes to OCC‘s fee practices as well. The buffer used to calculate each year‘s fees—that is, the amount by which that year‘s projected expenses are inflated to arrive at the amount to be charged as upfront fees—decreases under the Plan from 31% to 25%. And the Plan provides for a permanent end to refunds (but not dividends) if Replenishment Capital becomes necessary and is not repaid in 24 months or if the target capital requirement is not restored within that period. See id.
OCC‘s Plan cannot go into effect unless approved by the SEC because OCC is a “self-regulatory organization” under the
Petitioners—two nonshareholder exchanges (Miami International Securities Exchange, LLC and BOX Options Exchange LLC), a clearing member (KCG Americas LLC, a subsidiary of Petitioner KCG Holdings), and a market participant (Susquehanna International Group, LLP)—sought judicial review. They also moved to stay the SEC‘s Order to prevent the OCC‘s Plan from going into effect, but a panel of this court denied the stay. Susquehanna Int‘l Grp., LLP v. SEC, No. 16-1061 (D.C. Cir. Feb. 23, 2016) (order denying motion for stay). As a consequence, OCC currently operates according to the Plan.
II
We have jurisdiction to review the SEC‘s Order pursuant to
OCC is registered as a clearing agency with the SEC, and is therefore classified as a “self-regulatory organization.” See
First, the Exchange Act requires that a clearing agency‘s rules “not impose any burden on competition not necessary or appropriate in furtherance of the purposes of” the Act.
Second, the Act requires that a clearing agency‘s rules be “designed . . . , in general, to protect investors and the public interest.”
Third, the same subsection requires that rules not be “designed to permit unfair discrimination . . . among participants in the use of the clearing agency.”
Fourth, the Act requires that a self-regulatory organization “comply with . . . its own rules.”
We do not reach any of those arguments, all of which contend that the OCC‘s Plan is inconsistent with the above-described requirements of the Exchange Act. We do not reach them because, as Petitioners also argue, the SEC‘s Order approving the Plan fails in a more basic respect: the Commission did not itself “find[]” or “determin[e],”
We have rejected this kind of agency “decisionmaking” before. In Gerber v. Norton, we held:
When a statute requires an agency to make a finding as a prerequisite to action, it must do so. Merely “[r]eferencing a requirement is not the same as complying with that requirement.” . . . Nor may the agency delegate its responsibility to the regulated party. See State of Idaho v. ICC, 35 F.3d 585, 596 (D.C. Cir. 1994) (holding that the ICC failed to meet its responsibilities under [the National Environmental Policy Act] by “deferr[ing] not only to the judgments of other agencies, but also to that of Union Pacific, the licensee“).
294 F.3d 173, 185-86 (D.C. Cir. 2002) (quoting Sugar Cane Growers Coop. v. Veneman, 289 F.3d 89, 97 (D.C. Cir. 2002)). To be sure, SEC appellate counsel acknowledges that the SEC cannot “simply accept what [a self-regulatory organization] has done,” but rather is “obligated to make an independent review.” Oral Arg. Tr. 24. And the SEC‘s Order states that the Commission did make the necessary findings. See Order, 81 Fed. Reg. at 8300-04. But “‘[s]tating that a factor was considered‘—or found—‘is not a substitute for considering’ or finding it.” Gerber, 294 F.3d at 185 (quoting Getty v. Fed. Savings & Loan Ins. Corp., 805 F.2d 1050, 1055 (D.C. Cir. 1986)). As explained below, stating, not finding, is what the Commission did here.
A
The Order‘s shortcomings are apparent in its discussion of whether the Plan pays dividends to shareholder exchanges at a reasonable rate. That is a central issue: if the dividend rate represents an unnecessary windfall for shareholders, as Petitioners argue, then the Plan may run afoul of the Exchange Act‘s prohibitions by unnecessarily or inappropriately burdening competition, harming the interests of investors and the public, or unfairly discriminating against nonshareholders and clearing members. See
That explanation raises more questions than it answers. Who were those independent experts? How does the SEC know they were independent? What analysis did they and OCC‘s Board perform? How did they measure the “level” of the dividends? How did they measure the “cost and risks“? And how did they determine that the dividend level was reasonable for the associated cost and risks? The Order is silent on all counts. Instead, the SEC candidly admits that it simply “rel[ied] on the Board‘s analysis” of “the rate of return the Stockholder Exchanges were receiving for their capital investment.” SEC Br. 30. That is, to decide whether the dividend level was reasonable, the SEC took OCC‘s word for it.
This is not the reasoned analysis that the Exchange Act and the APA require. The Exchange Act permits the SEC to approve OCC‘s proposed rule change only “if it finds that” the proposal is consistent with the requirements of the Act.
Nor may the SEC reach a conclusion that is “unsupported by substantial evidence” or “arbitrary [and] capricious.”
On appeal, the SEC defends the Order‘s approach to the dividend rate in several ways. The SEC‘s principal argument is that it was reasonable for it to trust “the process” undertaken by OCC. Oral Arg. Tr. 20, 29. Restating the words of the Order, the SEC‘s brief says that the process included “independent analysis” by outside consultants. SEC Br. 31. But the SEC does not appear to have identified the consultants or seen their analysis, so it came as no surprise that the SEC disclaimed reliance on that outside analysis at oral argument. Oral Arg. Tr. 18. Instead, the SEC ultimately relies on another aspect of OCC‘s decisionmaking process: what it describes as “arm‘s-length negotiations” between those OCC Board directors who represented clearing members and those who represented shareholders, which culminated in a “supermajority” Board vote in favor of the plan. SEC Br. 31, 38 & n.16; see also Oral Arg. Tr. 19-20.
“Trust the process” may be a reasonable slogan for the hometown basketball team of lead petitioner Susquehanna In-ternational Group.3 But the process alone cannot justify the dividend rate in this case. For one thing, it is hardly accurate to describe the negotiations between Board members as “arm‘s length.” OCC‘s shareholders have effective veto power over certain proposals, giving them outsize bargaining power compared to clearing members represented on the Board. See Letter from James E. Brown, Executive Vice President, OCC, to Brent J. Fields, Secretary, SEC at 6 n.10 (Feb. 23, 2015) (J.A. 85 n.10). Indeed, only four of nine directors representing clearing members voted in favor of the Plan, see id. at 7 n.12 (J.A. 86 n.12), making it less than clear that the process struck an appropriate balance between the interests of shareholders and clearing members. What is more, not all of the interested parties were even part of the negotiations among Board members. Only a small fraction of clearing members are on the Board, and none of the nonshareholder exchanges are. See Order Approving Proposed Rule Change Concerning a Proposed Capital Plan, 80 Fed. Reg. 13058, 13060 n.8 (Mar. 12, 2015). So, as to any agreement between OCC and shareholders regarding the dividend rate, the shareholder exchanges were on both sides of the transaction (because they were both OCC Board members and recipients of the dividends), while nonshareholder exchanges were on neither.
More fundamentally, the SEC cannot rely on OCC‘s process totally divorced from any examination of the substance of the Plan, especially when the procedural features that the SEC cites are little more than the general elements of OCC‘s governance structure. Indeed, were we to accept this justification, that would mean the SEC could or should approve nearly any plan OCC proposes—hardly the result the Exchange Act envisions. See, e.g.,
The SEC also defends its approval of the Plan‘s dividend rate by arguing that the Plan‘s structure guarantees reasonable dividends. See Order, 81 Fed. Reg. at 8301 (“The Commission believes that various components of the Capital Plan operate to set reasonable dividends for the cost and risks associated with the Stockholder Exchanges’ contributed and committed capital.“); id. at 8303; Oral Arg. Tr. 21. In particular, the SEC notes that the Plan sets fees based on a fixed percentage of projected costs, and allocates unused fees between dividends and refunds in a fixed proportion. Those inputs, the SEC argues, ensure that the dividend rate will turn out to be reasonable. Oral Arg. Tr. 21, 23.
That reasoning begs the question. The resulting dividend rate will only be reasonable if the Plan‘s structure (most significantly, the portion of unused fees allocated to dividends) is reasonable. But the Order does not analyze whether—or explain why—it is reasonable to allocate roughly half of unused fees to dividends, as opposed to using a different percentage or a formula other than a fixed proportion of unused fees. In other words, the SEC defends its unquestioning reliance on OCC‘s claim that the dividend rate is reasonable by its unquestioning reliance on OCC‘s claim that the Plan‘s structure is reasonable. That is no defense at all.
B
The SEC‘s lack of reasoned decisionmaking in assessing the dividend rate is enough to make its Order arbitrary and capricious. But that is not the Order‘s only flaw. A similar lack of reasoned decisionmaking recurs throughout the Order, including in the four additional instances that follow.4
In the Order, the SEC states that it “review[ed] the process used by OCC to establish” the target. Order, 81 Fed. Reg. at 8301. But what was that process? OCC states that “[a]n outside consultant conducted a ‘bottom-up’ analysis of OCC‘s risks” and, “[b]ased on internal operational risk scenarios and loss modeling,” OCC quantified its operational and pension risk. Notice of Proposed Rule Change, 80 Fed. Reg. at 5172-73. But since there is no indication that the SEC knew who the consultant was, what analysis he or she conducted, or what additional analysis OCC performed, the Commission was in no position to make a reasoned finding that OCC‘s process was sound—let alone that the resulting capital target was reasonable. In short, the SEC‘s Order reflects the same lack of reasoned decisionmaking when assessing the capital target as it displays when assessing the dividend rate. “Trusting the process” is simply not enough.
Second, the SEC was also too quick to accept OCC‘s claims that the Plan would not increase fees for customers. In determining that the Plan is designed to protect investors and the public interest,
The only thing the Order says to justify ignoring net fees in favor of focusing on upfront fees is that the Plan is “designed to give market participants the benefit of lower upfront transaction costs, especially those customer end users who do not receive passed through refunds from the clearing member.” Order, 81 Fed. Reg. at 8302-03 (emphasis added). The underlying assumption of this justification must be that there are so few market participants who do receive passed-through refunds that net fees are irrelevant. But the SEC‘s Order does not provide record support for that assumption. Instead, it cites, see id. at 8303 n.127, a portion of the Notice of Filing of a Proposed Rule Change, prepared by OCC, stating that “OCC is aware that a portion—possibly a significant portion—of those refunds are not passed through by the clearing members to their end user customers.” 80 Fed. Reg. at 5175. That is, the SEC Order‘s only basis for not taking passed-through refunds into account is a statement by OCC, prepared to justify OCC‘s own proposal, that a “possibly” significant portion of refunds are not passed through. That does not qualify as a “satisfactory explanation,” State Farm, 463 U.S. at 43. See NetCoalition, 615 F.3d at 540-41 (criticizing the SEC‘s reliance on the “self-serving views of the regulated entities” in that case).
Third, the Order fails to give any explanation at all for rejecting one of Petitioners’ objections. Petitioners charged the Plan with unfair discrimination for treating refunds to clearing members differently from dividends to shareholders under the following scenario: If Replenishment Capital becomes necessary and is not repaid in 24 months (or if the target capital requirement is not restored in the same period), refunds will end permanently but dividends can resume. See Notice of Proposed Rule Change, 80 Fed. Reg. at 5174-75 (explaining this feature); Letter from James E. Brown, Executive Vice President, OCC, to Brent J. Fields, Secretary, SEC at 4-5 (Feb. 24, 2015) (J.A. 105-06) (discussing objections). The Order does not explain why that criticism is unavailing. In fact, as the SEC concedes, the Order misstates the Plan feature at issue, incorrectly explaining at one point that both refunds and dividends will end permanently if Replenishment Capital goes unpaid for 24 months. Order, 81 Fed. Reg. at 8297; Oral Arg. Tr. 13-14 (concession by SEC counsel that “that is mistaken“). Hence, not only does the Order give no indication that the SEC analyzed this Plan feature for compliance with the Act, but it instead suggests that the SEC may have misunderstood the feature entirely. That double deficiency demonstrates a lack of reasoned decisionmaking.
Finally, the Order gives short shrift to Petitioners’ objection that OCC, by failing to notify nonshareholder exchanges earlier in its development of the Plan, violated its own bylaws. Primarily at issue is OCC‘s bylaw requirement that nonshareholder exchanges “be promptly provided with information that [OCC‘s] Executive Chairman considers to be of competitive significance.” OCC, Bylaws Art. VIIB, Interpretations & Policies § 1.01 (amended 2014). Despite recognizing that there was a dispute over whether the Plan was competitively significant, the Order merely notes that “OCC represented that it” had “completed all action required to be taken under its . . . bylaws.” Order, 81 Fed. Reg. at 8305. On remand, the SEC must resolve Petitioners’ argument that OCC could not reasonably have considered the Plan to be competitively insignificant. Or, if the SEC concludes that this does not matter, it must give a reasoned explanation why.
III
Having concluded that the SEC‘s Order is arbitrary and capricious, unsupported by substantial evidence, and otherwise not in accordance with law, we are left with the question of the proper remedy. “An inadequately supported rule . . . need not necessarily be vacated.” Allied-Signal, Inc. v. U.S. Nuclear Regulatory Comm‘n, 988 F.2d 146, 150 (D.C. Cir. 1993). “The decision whether to vacate depends on ‘the seriousness of the order‘s deficiencies (and thus the extent of doubt whether the agency chose correctly) and the disruptive consequences of an interim change that may itself be changed.‘” Id. at 150-51 (quoting Int‘l Union, UMW v. Fed. Mine Safety & Health Admin., 920 F.2d 960, 967 (D.C. Cir. 1990)).
Here, the SEC may be able to approve the Plan once again, after conducting a proper analysis on remand. Moreover, unwinding the Plan in the interim would be—in Petitioners’ own words—a “logistical nightmare.” Pet‘rs’ Emergency Mot. for Stay 9. Indeed, the parties may well awaken from that nightmare just as the SEC decides to rewind the unwinding.
Thankfully, we can avoid that restless night. The SEC and OCC assure us that it will be possible to unwind the Plan at a later time, see SEC‘s Opp‘n to Pet‘rs’ Emergency Mot. for Stay 2-3; Oral Arg. Tr. 12 (SEC); id. at 33-34 (OCC), and no party contends that the task would be materially more difficult if done then rather than now. Thus, with no immediate need to vacate the Order, we will remand the case to give the SEC an opportunity to properly evaluate the Plan. See Allied-Signal, Inc., 988 F.2d at 151 (remanding rather than vacating “because of the possibility that the Commission may be able to justify the Rule, and the disruptive consequences of vacating“).
So ordered.