PER CURIAM:
Pending before us is a motion, pursuant to Federal Rule of Appellate Procedure 8(a), to stay district court proceedings in a litigation the Plaintiff-Appellant-Petitioner, the United States Securities and Exchange Commission (“S.E.C.”), brought against Citigroup Global Markets Inc. (“Citigroup”). The district court, by order of November 28, 2011, refused to approve the settlement between the S.E.C. and Citigroup and ordered them to a prompt trial. The S.E.C., as plaintiff, and Citigroup, as defendant, each bring interlocutory appeal from that order. The S.E.C. alternatively petitions for a writ of mandamus to set the order aside. Citigroup joins with the S.E.C. in all of its arguments.
The challenge by both parties to the district court’s order raises important questions. These include the division of responsibilities as between the executive and the judicial branches and the deference a federal court must give to policy decisions of an executive administrative agency as to whether its actions serve the public interest (and as to the agency’s expenditure of its resources). They include also the question of a court’s authority to reject a private party’s decision to compromise its case on the ground that the court is not persuaded that the party has incurred any liability by its conduct.
As we are satisfied the criteria for a stay are met, it is hereby ORDERED that the motion for a stay is GRANTED and the motion to expedite the appeal is DENIED.
We recognize that, because both parties to the litigation are united in seeking the stay and opposing the district court’s order, this panel has not had the benefit of adversarial briefing. In order to ensure that the panel which determines the merits receives briefing on both sides, counsel will be appointed to argue in support of the district court’s position.
The merits panel is, of course, free to resolve all issues without preclusive effect from this ruling. In addition to the fact that our ruling is made without benefit of briefing in support of the district court’s position, our ruling, to the extent it addresses the merits, finds only that the movant has shown a
likelihood
of success and does not address the ultimate question to be resolved by the merits panel — whether the district court’s order should in fact be overturned.
Background
We begin by summarizing the proceedings, the proposed settlement, and the district court’s order. As part of an industry-wide investigation into certain abuses that contributed to the recent financial crisis, the S.E.C. undertook an investigation of Citigroup’s marketing of collateralized debt obligations (“CDOs”). After several years of investigation, discovery, and discussions with Citigroup, the S.E.C. filed a complaint charging Citigroup with negligent misrepresentation under 15 U.S.C. § 77q(a)(2) and (3). Simultaneously with the filing of the complaint, the S.E.C. and Citigroup presented a proposed consent judgment to the district court for its approval. The settlement provided in essence the following: Citigroup agreed (1) to pay $285 million into a fund, which the S.E.C. may distribute to investors in a pool of CDOs marketed by Citigroup in compensation of their losses, (2) to the entry of an order enjoining it from violating certain sections of the Securities Act of 1933, and (3) to undertake to establish procedures to prevent future violations and to make periodic demonstrations of compliance to the S.E.C.
The district court rejected the settlement, concluding that it was “neither reasonable, nor fair, nor adequate, nor in the public interest,” primarily because it included no admission by Citigroup of liability.
U.S. Secs, and Exch. Comm’n v. Citigroup Global Mkts. Inc.,
No. 11 Civ. 7387, 827 F.Supp.2d 328, 335, 2011 WL 5903733, at *6 (S.D.N.Y. Nov. 28, 2011). The court explained three main reasons that justified its conclusion:
First, the court expressed strong disapproval of what it called “the S.E.C.’s longstanding policy — hallowed by history but not by reason — of allowing defendants to enter into Consent Judgments without admitting or denying the underlying allegations.”
Id.
at 332, 2011 WL 5903733, at *4. Without the defendant’s admission, such a judgment would have “no collateral estoppel effect” in another litigation brought against the defendant by victims of its alleged wrongdoing.
Id.
“[It] ... leaves the defrauded investors substantially short-changed ... [as they] cannot derive any collateral estoppel assistance from Citigroup’s non-admission/non-denial of the S.E.C.’s allegations.”
Id.
at 334, 2011 WL 5903733, at *5. The court found it “hard[ ] to discern ... what the S.E.C. is getting from this settlement other than a quick headline.”
Id.
Because it “does not involve any admissions and ... results in only very modest penalties [described by the court as “pocket change to an entity as large as Citigroup,”
id.],
[such a consent judgment] is just as frequently viewed,
particularly in the business community, as a cost of doing business.”
Id.
The court also found that the settlement “without any admissions [of liability by Citigroup] serves various narrow interests of the parties,” but not the public interest.
Id.
(emphasis omitted).
The second reason given by the court for rejecting the consent judgment was its perceived unfairness to the defendant, Citigroup.
[The settlement] is not reasonable, because how can it ever be reasonable to impose substantial relief [on Citigroup] on the basis of mere allegations? It is not fair, because, despite Citigroup’s nominal consent, the potential for abuse in imposing penalties on the basis of facts that are neither proven nor acknowledged is patent.
Id.
at 335, 2011 WL 5903733, at *6.
The court’s third reason for concluding that the consent judgment was not in the public interest was that, without admission of liability, a consent judgment involving only modest penalties gives no “indication of where the real truth lies.”
Id.
at 333, 2011 WL 5903733, at *5.
[The settlement] is not adequate, because, in the absence of any facts, the Court lacks a framework for determining adequacy. And, most obviously, the proposed Consent Judgment does not serve the public interest, because it asks the Court to employ its power and assert its authority when it does not know the facts.
An application of judicial power that does not rest on facts is worse than mindless, it is inherently dangerous. The injunctive power of the judiciary is not a free-roving remedy to be invoked at the whim of a regulatory agency, even with the consent of the regulated. If its deployment does not rest on facts — cold, hard, solid facts, established either by admissions or by trials — it serves no lawful or moral purpose and is simply an engine of oppression.
The issue pending before us is whether to grant S.E.C.’s motion for a stay of district court proceedings pending this court’s disposition of the appeals and mandamus petition.
Discussion
Preliminarily, we note that, while it is unclear whether interlocutory appeal lies from an order refusing to approve a proposed consent judgment,
see, e.g., State of New York v. Dairylea Coop., Inc.,
698 F.2d 567, 570 (2d Cir.1983), there is no question that our court has jurisdiction to rule on the issues raised by the petition for mandamus, even if we determine that interlocutory appeal does not lie in these circumstances. We recognize that the standard for grant of mandamus is more onerous than the standard for reversal on appeal.
See Secs, and Exch. Comm’n v. Rajaratnam,
622 F.3d 159, 169 (2d Cir.2010). Nevertheless, we need not resolve whether we have jurisdiction to review the district court’s order on interlocutory appeal because we conclude that the parties have made the required showing of likelihood of success regardless of the applicable standard of review.
In determining whether to issue a stay, the governing precedents direct that we consider the following factors: “(1) whether the stay applicant has made a strong showing that he is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.”
Hilton v. Braunskill,
481 U.S. 770, 776, 107 S.Ct. 2113, 95
L.Ed.2d 724 (1987);
accord In re World Trade Ctr. Disaster Site Litig.,
503 F.3d 167, 170 (2d Cir.2007);
United States v. E. Air Lines, Inc.,
923 F.2d 241, 243-44 (2d Cir.1991).
A.
Likelihood of Success
We first turn to whether the S.E.C. and Citigroup have a strong likelihood of success on the merits in their effort to overturn the court’s ruling. We examine each of the three prongs of the court’s justifications in turn.
1.
Failure of the settlement to serve the public interest.
The court’s first and most important justification for its ruling was that a consent judgment without Citigroup’s admission of liability is bad policy and fails to serve the public interest because defrauded investors cannot use the judgment to establish Citigroup’s liability in civil suits to recover the investors’ losses. In that reasoning, we perceive several problems.
First, it prejudges the fact that Citigroup had in fact misled investors, and assumes that the S.E.C. would succeed at trial in proving Citigroup’s liability. The court appeared to assume that the S.E.C. had a readily available option to obtain a judgment that established Citigroup’s liability, either by trial or settlement, but chose for no good reason to settle for less. The district court’s logic appears to overlook the possibilities (i) that Citigroup might well not consent to settle on a basis that requires it to admit liability, (ii) that the S.E.C. might fail to win a judgment at trial, and (iii) that Citigroup perhaps did not mislead investors.
A still more significant problem is that the court does not appear to have given deference to the S.E.C.’s judgment on wholly discretionary matters of policy. The S.E.C.’s decision to settle with Citigroup was driven by considerations of governmental policy as to the public interest. The district court believed it was a bad policy, which disserved the public interest, for the S.E.C. to allow Citigroup to settle on terms that did not establish its liability. It is not, however, the proper function of federal courts to dictate policy to executive administrative agencies. “[F]ederal judges — who have no constituency — have a duty to respect legitimate policy choices made by those who do. The responsibilities for assessing the wisdom of such policy choices and resolving the struggle between competing views of the public interest are not judicial ones: ‘Our Constitution vests such responsibilities in
the public branches.’ ”
Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc.,
467 U.S. 837, 866, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) (quoting
TVA v. Hill,
437 U.S. 153, 195, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978));
see also Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co.,
463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) (in reviewing whether agency action is arbitrary and capricious under 5 U.S.C. § 706(2)(A), “a court is not to substitute its judgment for that of the agency”). Some courts have gone so far as to conclude that an agency’s assessment of policy factors driving a decision to settle is an “agency action ... committed to agency discretion by law” and therefore not subject to any judicial review under 5 U.S.C. § 706(1)(a)(2).
See, e.g., Ass’n of Irritated Residents v. E.P.A.,
494 F.3d 1027, 1031-33 (D.C.Cir.2007);
New York State Dep’t of Law v. F.C.C.,
984 F.2d 1209, 1214-15 (D.C.Cir.1993). While we are not certain we would go so far as to hold that under no circumstances may courts review an agency decision to settle, the scope of a court’s authority to second-guess an agency’s discretionary and policy-based decision to settle is at best minimal.
The numerous factors that affect a litigant’s decision whether to compromise a case or litigate it to the end include the value of the particular proposed compromise, the perceived likelihood of obtaining a still better settlement, the prospects of coming out better, or worse, after a full trial, and the resources that would need to be expended in the attempt. In the case of a public executive agency such as the S.E.C., the factors include also an assessment of how the public interest is best served. These are precisely the factors that the Supreme Court has recognized as making a discretionary agency decision unsuitable for judicial review.
Heckler v. Chaney,
470 U.S. 821, 831-32, 105 S.Ct. 1649, 84 L.Ed.2d 714 (1985). Based on our preliminary review, and noting that the settlement called for payment by Citigroup of $285 million, which would be available for compensation of investors who lost money, we see no basis to doubt that the SEC’s decision was made in consideration of all of those factors.
See
SEC’s Memorandum of Law in Response to Questions Posed by the Court Regarding Proposed Settlement,
U.S. Secs, and Exch. Comm’n v. Citigroup Global Mkts., Inc.,
No. 11 Civ. 7387(JSR) (Nov. 7, 2011).
While the district court verbally acknowledged its obligation to give deference to the S.E.C.’s decision (stating that it had given “fullest deference to the S.E.C.’s views,” 827 F.Supp.2d at 335, 2011 WL 5903733, at *6), there is no indication in the record that the court in fact gave deference to the S.E.C.’s judgment on any of these questions. The S.E.C. believed, for example, that the public interest was served by the defendant’s disgorgement of $285 million, available for compensation of claimants against Citigroup, plus other concessions. The court simply disagreed. In concluding that the settlement was not in the public interest, the court took the view that Citigroup’s penalty was “pocket change” and the S.E.C. got nothing from the settlement but “a quick headline.”
Id.
at 333-34, 2011 WL 5903733, at *5.
In
addition, the court does not appear to have considered the agency’s discretionary assessment of its prospects of doing better or worse, or of the optimal allocation of its limited resources. Instead, the district court imposed what it considered to be the best policy to enforce the securities laws. In short, we conclude it is doubtful whether the court gave the obligatory deference to the S.E.C.’s views in deciding that the settlement was not in the public interest.
Finally, we question the district court’s apparent view that the public interest is disserved by an agency settlement that does not require the defendant’s admission of liability. Requiring such an admission would in most cases undermine any chance for compromise.
2. Unfairness to Citigroup.
The second basis of the court’s reasoning, as noted, was that the settlement was unfair
to Citigroup
because it imposed on Citigroup “substantial relief on the basis of mere allegations ... that are neither proven nor acknowledged.” 827 F.Supp.2d at 335, 2011 WL 5903733, at *6. The imposition of such a judgment on Citigroup, which did not “rest on facts ... established either by admissions or by trials,” was characterized by the court as “simply an engine of oppression.”
Id.
In the first place, we have difficulty reconciling the court’s concern for the substantiality of the relief being imposed on Citigroup with the court’s earlier observation that the penalties imposed on Citigroup amounted to no more than “pocket change” or a “mild and modest cost of doing business.”
Id.
at 333-34, 2011 WL 5903733, at *5. But a more important concern is whether it is a proper part of the court’s legitimate concern to protect a private, sophisticated, counseled litigant from a settlement to which it freely consents. We doubt that a court’s discretion extends to refusing to allow such a litigant to reach a voluntary settlement in which it gives up things of value without admitting liability.
Cf. Janus Films, Inc. v. Miller,
801 F.2d 578, 582 (2d Cir.1986) (in consent judgment, “[t]he court makes no determination of the merits of the controversy”). And there is no suggestion in the materials we have received that Citigroup’s settlement was anything other than voluntarily given, and, as the district court acknowledged, in the interests of Citigroup.
3.
Absence of basis to assess the fairness of the settlement
The court’s third ground for its refusal to accept the settlement was that it could not properly evaluate the fairness of the settlement unless the underlying facts were conclusively established either by a trial or by binding admission of liability. We doubt whether the absence of proven or admitted liability could justify the refusal to approve the settlement. In the first place, it is not correct that the court had no basis available to assess the underlying facts. The substantial evidentiary record amassed by the S.E.C. over its lengthy investigation was available to the court, and the S.E.C. did provide information to the court regarding how the evidence supported the proposed consent judgment.
See
SEC’s Memorandum of Law in Re
sponse to Questions Posed by the Court Regarding Proposed Settlement,
supra,
at 16-22. Even assuming the doubtful proposition that a court has authority to demand assurance that a voluntary settlement reached between an administrative agency and a private party somehow reflects the facts that would be demonstrated at a trial, the court was free to assess the available evidence and to ask the parties for guidance as to how the evidence supported the proposed consent judgment.
The district court’s reasoning was that the settlement must be deemed to be either insufficiently onerous or excessively onerous unless the liability of Citigroup had been either proved or disproved at trial or one side or the other had conceded the issue. This is tantamount to ruling that in such circumstances, a court will not approve a settlement that represents a compromise. It is commonplace for settlements to include no binding admission of liability. A settlement is by definition a compromise. We know of no precedent that supports the proposition that a settlement will not be found to be fair, adequate, reasonable, or in the public interest unless liability has been conceded or proved and is embodied in the judgment.
We doubt whether it lies within a court’s proper discretion to reject a settlement on the basis that liability has not been conclusively determined.
Having considered the various explanations given by the district court for its refusal to permit the settlement, we conclude that the S.E.C. and Citigroup have a strong likelihood of success in their joint effort to overturn the district court’s ruling. We turn to the other factors to be considered on the question of whether to grant a stay.
B.
Irreparable Harm
The second factor to be considered is “irreparable harm.” Both the S.E.C. and Citigroup will incur significant harm absent a stay if they are prevented from settling their dispute and are ordered to prompt trial, as the district court has directed. We recognize that there are authorities to the effect that the inability to conclude a settlement on the terms agreed to by the parties does not constitute “irreparable harm.”
See Grant v. Local 638,
373 F.3d 104, 109 (2d Cir.2004);
State of New York v. Dairylea Coop., Inc.,
698 F.2d 567, 570 (2d Cir.1983). These cases, however, are substantially different from this one. The principal difference lies in the fact that the refusals to accept settlements that were found not to constitute irreparable harm left the parties free to return to the bargaining table to make reasonable adjustments of terms of settlement or to demonstrate the fairness at a hearing. Here, the court’s posture — requiring a binding admission of liability as a condition of approval of the settlement— virtually precludes the possibility of settlement.
In
Grant,
the defendant union had been found to have discriminated against nonwhites and had further been found to have violated the court’s remedial decree. 373 F.3d at 105-06. While contempt proceedings against the union were pending, the parties reached a settlement that included
a cap on the union’s liability for back pay.
Id.
at 106. After receiving objections to the settlement, the district court declined to approve it.
Id.
Having received an expert’s report to the effect that the union’s potential liability vastly exceeded the cap contemplated in the settlement, and that the union’s ability to pay exceeded the cap, the court directed a hearing on those questions.
Id.
The union appealed, and the parties who had successfully opposed the settlement in the district court contested the appealability of the court’s interlocutory order that refused to accept the settlement.
Id.
Our court considered whether the union would suffer irreparable harm if the appeal were disallowed.
Id.
at 108-11. We distinguished the facts from those in
Carson v. American Brands, Inc.,
450 U.S. 79, 83-84, 101 S.Ct. 993, 67 L.Ed.2d 59 (1981), in which the Supreme Court had found sufficient harm to justify the interlocutory appeal.
Grant,
373 F.3d at 107-09. In particular, in
Carson,
“the district court made clear that it would not enter any decree containing remedial provisions [as contemplated by the proposed settlement] that did not rest solidly on evidence of discrimination and that were not expressly limited to actual victims of discrimination.”
Carson,
450 U.S. at 87 n. 12, 101 S.Ct. 993. The Supreme Court noted, “[i]n ruling so broadly, the court ... effectively foreclosed ... consideration [of the merits of plaintiffs’ injunctive claim embodied in the settlement].”
Id.
For this, and other reasons, the Supreme Court found that the district court’s refusal “to approve the parties’ negotiated consent decree” constituted “serious, perhaps irreparable consequences” and allowed immediate interlocutory appeal.
Id.
at 89, 101 S.Ct. 993.
In our
Grant
ruling, we contrasted the harm suffered by the appealing union with the harm suffered by the appellants in
Carson.
We found that the continued availability of settlement in
Grant
mitigated the harm from the court’s refusal to accept the settlement terms presented. We wrote:
In any case, it is certain that Local 28 [the would-be appellant] has not shown irreparable harm of the type envisioned in
Carson.
The district court’s order does not, as did the order in
Carson,
effectively foreclose the parties from negotiating a settlement.... Because the district court made no comments similar to those of the district court in
Carson,
there is no indication that it would never allow a modification of injunctive relief similar to that in the proposed consent decree.
373 F.3d at 108-09.
Dairylea Cooperative, Inc.,
698 F.2d 567, is to similar effect. There, the parties to an antitrust suit presented a proposed settlement including injunctive provisions to the district court for approval.
Id.
at 568-69. The court, upon consideration of the objections of other defendants who claimed that aspects of the proposed settlement would have anticompetitive impact, upheld the objection on the basis that the settlement would provide the settling defendant with unfair advantages over other competitors and that consumers would not be adequately compensated for Dairylea’s prior overcharging.
Id.
at 569.
The settling parties brought an interlocutory appeal, which we rejected. As in
Grant,
we compared the circumstances to those in
Carson
in which the appeal had been allowed. We noted:
As set forth in
Carson,
the rationale for permitting appeals from denials of settlement agreements which have the “practical effect of denying an injunction” is to allow appellate review only of orders which might result in serious, irreparable harm to the party to whom
injunctive relief is denied. NY and Dairylea [the parties appealing from the district court’s refusal to endorse the settlement] have failed to make such a showing. The parties remain free to return to the bargaining table to devise a settlement which would respond to [the district court’s] objections. Indeed, the district court opinion explicitly expresses a willingness to consider further proposals.
Id.
at 570.
The distinction we drew in
Grant
and
Dairylea
favors a finding of irreparable harm and the grant of a stay in the present case. The position taken by the district court in this case in rejecting the proposed settlement is far more similar to that in
Carson
than that in
Grant
and in
Dairylea.
Here, the district court’s rejection of the settlement cannot be cured by the parties returning to the bargaining table to make relatively minor adjustments to the terms of the settlement to address the district court’s concern. The district court’s intimation that it will not approve a settlement that does not involve Citigroup’s admission of liability, a condition that Citigroup is unlikely to satisfy, substantially reduces the possibilities of the parties reaching settlement. As in
Carson,
“in refusing to approve the parties’ negotiated consent decree, the District Court denied petitioners the opportunity to compromise their claim.... These constitute ‘serious, perhaps irreparable, consequences’ that petitioners can ‘effectually challenge’ only” if our court grants a stay. 450 U.S. at 89-90, 101 S.Ct. 993.
C.
Substantial Injury to Other Persons Interested in the Proceeding
The third factor to be considered is whether issuance of the stay will substantially injure other persons interested in the proceeding. The stay does nothing more than maintain the status quo existing prior to the district court’s order. We see no appreciable harm to anyone from issuing a stay.
D.
Public Interest
The final factor to be considered is the public interest. The S.E.C. asserts that its settlement is in the public interest and that its access to a stay so as to protect the settlement is also in the public interest. We are bound in such matters to give deference to an executive agency’s assessment of the public interest.
See Chevron,
467 U.S. at 866, 104 S.Ct. 2778 (“The responsibilities for ... resolving the struggle between competing views of the public interest are not judicial ones: ‘Our Constitution vests such responsibilities in the political branches.’ ”) (quoting
TVA v. Hill,
437 U.S. 153, 195, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978));
Publicker Indus., Inc. v. United States (In re Cuyahoga Equipment Corp.),
980 F.2d 110, 118 (2d Cir.1992) (“Appellate courts ordinarily defer to the agency’s expertise and the voluntary agreement of the parties in proposing the settlement.”). This does not mean that a court must necessarily rubber stamp all arguments made by such an agency. It does mean at least that a court should not reject the agency’s assessment without substantial reason for doing so.
We have no reason to doubt the S.E.C.’s representation that the settlement it reached is in the public interest. We see no bases for any contention that the S.E.C.’s decision to enter into the settlement was “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). Nor do we find reason to doubt that the stay the S.E.C. seeks, so as to prosecute its challenge to the district court’s disallowance of the settlement, is also in the public interest.
Conclusion
In conclusion, we are satisfied (1) that the S.E.C. and Citigroup have made a strong showing of likelihood of success in setting aside the district court’s rejection of their settlement, either by appeal or petition for mandamus; (2) the petitioning parties have shown serious, perhaps irreparable, harm sufficient to justify grant of a stay; (3) the stay will not substantially injure any other persons interested in the proceeding; and (4) giving due deference to the S.E.C.’s assessment of the importance of its settlement to the public interest, that interest is not disserved by our grant of a stay.
It is hereby ORDERED that the motion to stay the proceedings in the district court is GRANTED pending the outcome of these consolidated appeals, and the motion to expedite the appeal is DENIED. The Clerk of the Court is directed to appoint counsel, who will advocate for upholding the district court’s order, and to set a briefing schedule. Counsel will submit briefs addressing the issues discussed above, as well as any other matters they consider pertinent. These appeals shall be heard by a panel in due course.