SYKES, Circuit Judge.

Trachte Building Systems, Inc., a Wisconsin manufacturer, established an employee stock ownership plan ("ESOP") in the mid-1980s when ESOPs were a popular employee-benefits instrument. In the late 1990s, David Fenkell and Alliance Holdings, Inc., a company he founded and controlled, developed a niche specialty in buying and selling ESOP-owned, closely held companies with limited marketability. In the typical transaction, Fenkell would merge the ESOP of an acquired company into Alliance's own ESOP, hold the company for a few years with its management in place, and then spin it off at a profit (assuming everything went as planned).

In accordance with this business model, Alliance acquired Trachte in 2002 for $24 million and folded its ESOP into Alliance's ESOP. Fenkell projected that the company would fetch around $50 million in five years. When the time came to sell, however, Trachte's profits were flat, its growth had stalled, and no independent buyer would pay anywhere near that price. So Fenkell offloaded the company to its employees in a complicated leveraged buyout. Greatly simplified, the deal involved three steps. First, Fenkell directed the creation of a new Trachte ESOP managed by trustees beholden to him. Next, the accounts in the Alliance ESOP were spun off to the new Trachte ESOP. Finally, the new Trachte ESOP used the employees' accounts as collateral to incur debt to purchase Trachte's equity back from Alliance. Multiple interlocking transactions to that effect closed on the same day in August 2007. When all was said and done, Trachte and the new Trachte ESOP had paid $45 million for 100% of Trachte's stock and incurred $36 million in debt.

The purchase price was inflated and the debt load was unsustainable. By the end of 2008, Trachte's stock was worthless. The losers in this deal — the employee participants in the new Trachte ESOP — sued Alliance, Fenkell, his handpicked trustees, and several other entities alleging breach of fiduciary duty in violation of ERISA. The district court held a bench trial and issued a comprehensive opinion finding the defendants liable. Chesemore v. Alliance Holdings, Inc. (Chesemore I), 886 F.Supp.2d 1007 (W.D.Wis. 2012). After an additional hearing, the judge crafted a careful remedial order making the class and a subclass whole. Chesemore v. Alliance Holdings, Inc. (Chesemore II), 948 F.Supp.2d 928 (W.D.Wis. 2013). The judge later awarded attorney's fees and approved settlements among some of the parties.

Fenkell appealed. He concedes liability but raises many objections to the remedial order, the award of attorney's fees, and the settlements by his codefendants. The only substantial issue is a challenge to the judge's order requiring him to indemnify his cofiduciaries. We held more than 30 years ago that ERISA allows this. Free v. Briody, 732 F.2d 1331, 1337-38 (7th Cir. 1984). Since then a circuit split has arisen on this subject, but we're not persuaded that Free should be overruled. None of Fenkell's other arguments has merit.

The plaintiffs filed a cross-appeal seeking a larger award of attorney's fees and contesting the judge's refusal to award costs against Fenkell. We reject these challenges. Finally, while we've had this case under advisement, the district court found Fenkell in contempt for failing to comply with the remedial order. Fenkell appealed that order as well, but his arguments are frivolous. Accordingly, we affirm in all respects.

I. Background

Trachte Building Systems designs and manufactures steel self-storage systems in Sun Prairie, Wisconsin. In the 1980s Stephen Pagelow, the son-in-law of Trachte's founder, acquired a controlling interest in the company and took over as president and chairman of the board. In 1987 Pagelow directed the establishment of an employee stock ownership plan, or ESOP, as a benefit to employees, selling some of his shares to the plan.1 Throughout the 1990s Trachte experienced significant growth in both sales and operations.

David Fenkell established Alliance in 1994 and at all relevant times was its president, CEO, and sole director. Fenkell also was president, CEO, and sole director of two Alliance subsidiaries, A.H.I., Inc., and AH Transition Corporation. (We'll refer to these companies collectively as "Alliance" unless the context requires otherwise.) Alliance was in the business of buying and selling ESOP-owned, closely held companies that might otherwise be difficult to sell. Alliance's business model was to fold the acquired company's ESOP into its own ESOP, leave the existing management in place, and spin off the company to another buyer a few years later, hopefully at a substantial profit. In short, Fenkell and Alliance made money by flipping ESOP-owned, closely held companies with limited marketability.

By 2002 Pagelow was looking for a way to gradually exit Trachte in anticipation of fully retiring in a few years. Enter Alliance, which that year acquired 80% of Trachte's common stock for $24 million and all of its preferred stock for $2 million. The 2002 transaction — more accurately, a series of interlocking transactions — involved folding the Trachte ESOP into Alliance's own ESOP by transferring the employees' accounts to the Alliance ESOP and exchanging the Trachte stock for Alliance stock. Trachte employees thus became participants in the Alliance ESOP, and the old Trachte ESOP was dissolved. Pagelow retained 20% of Trachte's common stock and a 40% ownership interest in a subsidiary. He also agreed to stay on as chairman for five years. In exchange he received a put option giving him the right to tender his Trachte shares to the company in 2007 at a price keyed to the prior year's appraised value.

After the 2002 transaction, Pagelow resigned as Trachte's president and was replaced by Jeffrey Seefeldt, a longtime Trachte manager. Pagelow immediately reduced his workweek and gradually began to cut back on his day-to-day management of the company. In the fall of 2005, Pagelow exercised part of his put option early. In mid-2006 he broke his hip, which radically reduced his involvement with the company.

During this time, Trachte's sales increased steadily but profits remained flat. Despite its stagnant profitability, the on-paper value of Trachte's stock rose dramatically, from $25.4 million in 2003 to $44.9 million in 2006. Pagelow's put option — coming due in 2007 — was pegged to the 2006 appraised value, but Alliance lacked the liquidity to satisfy it. Faced with the prospect of having to borrow to satisfy Pagelow's option and with serious doubts about Trachte's future performance, Fenkell decided it was past time to sell.

At the time of the 2002 transaction, Fenkell had projected that Trachte would sell for as much as $50 million in 2007. Throughout 2006 he looked for a buyer at or near that price, but he came up empty-handed. Failing to find an independent buyer at his desired price, Fenkell devised and implemented a complicated leveraged buyout to off-load the company onto Trachte's employees. The district court's opinion meticulously describes the history and details of this transaction, as well as the lack of any truly independent due diligence on behalf of Trachte's employees. Chesemore I, 886 F.Supp.2d at 1021-40. Because liability is uncontested here, a radically simplified summary will suffice.

First, on August 22, 2007, Fenkell orchestrated the removal of Trachte's entire board of directors and installed Seefeldt and James Mastrangelo, the chief operating officer, as the sole board members. Id. at 1036. Then, following a plan of Fenkell's devising, Seefeldt and Mastrangelo directed the creation of a new Trachte ESOP, installing themselves and Pamela Klute, the company's vice-president of human resources, as trustees. Id.

The leveraged buyout itself involved 11 separate steps, each of which occurred sequentially and was conditioned on the completion of all previous and subsequent steps. The district judge grouped these steps into three baskets. First, in steps 1-3, the accounts of the Trachte employees in the Alliance ESOP were spun off to the new Trachte ESOP, and their Alliance shares were exchanged for Trachte shares held by A.H.I. Id. at 1037-38. Next, in Steps 4-7, Trachte used the new Trachte ESOP accounts as collateral for loans to pay off the "phantom" stock plan of Alliance employees and redeem Trachte stock held by Alliance and Pagelow. Id. at 1038. Finally, in Steps 8-11, Trachte and the new Trachte ESOP acquired all Trachte equity held by Alliance, Alliance employees, and Pagelow. Id. at 1038-39.

This series of interdependent transactions closed on August 29, 2007. By the end of that day, Trachte and the new Trachte ESOP had paid $45 million in consideration for Trachte's total equity and incurred about $36 million in debt. Id. at 1039.

Trachte did not flourish after the 2007 leveraged buyout. It held its own until May 2008, but at that point projected that it would not meet its loan covenants. By the end of 2008, Trachte's stock was worthless.

Their equity wiped out, a group of current and former Trachte employees filed this class action alleging breach of fiduciary duty in violation of ERISA. The class includes current and former employees who participated in the old Trachte ESOP, the Alliance ESOP, and the new Trachte ESOP. A subclass comprises those participants in the new Trachte ESOP who would have remained employees of Alliance — and thus participants in the Alliance ESOP — but for the August 2007 transaction. Fenkell and Alliance were the primary targets of the suit. The complaint also named the trustees of the new Trachte ESOP as defendants. Pagelow, the new Trachte ESOP, and the Alliance ESOP were named as nominal defendants.2

After extensive litigation and a bench trial, the judge found the defendants liable. Fenkell and Alliance had insisted that they were not fiduciaries because all they did was spin off the Alliance ESOP to the new Trachte ESOP. The judge was not persuaded. He found:

Chesemore I, 886 F.Supp.2d at 1052. The judge continued: "Fenkell and Alliance designed the transaction so that either the accounts of the Trachte participants in the Alliance ESOP would be used as leverage to buy Trachte from Alliance or the accounts would revert to their prior situation with no change." Id. at 1053.

In other words, if there had been an actual independent fiduciary on the other side, Fenkell and Alliance wouldn't have gotten away with it. They installed trustees who "(1) had a conflict of interest that placed them under substantial duress during the negotiation and assessment of the deal; and (2) lacked the experience and the incentive to assess a deal of this type and complexity." Id. at 1054. Although the trustees formally made the decision to use the new Trachte ESOP accounts as collateral for the buyout, Fenkell and Alliance controlled that decision and orchestrated the entire complex transaction. In exercising that control, the judge concluded, they violated fiduciary duties owed to the plaintiffs.

The judge also held, however, that the defendants' fiduciary breach was not wholly responsible for Trachte's total collapse; the 2008 financial crisis also played a role, although the inflated purchase price and excess debt placed tremendous pressure on the company and sealed its fate. In the end, and after an extensive additional hearing, the judge crafted an intricate remedial order making the class and the subclass whole. As relevant here, he ordered the trustees to restore $6,473,856.82 to the new Trachte ESOP, allocated to the class members' accounts according to their shares as of the date of judgment. Chesemore II, 948 F.Supp.2d at 950. He ordered Fenkell and Alliance to restore $7,803,543 to the Alliance ESOP, allocated to the subclass members' accounts according to their holdings as of August 29, 2007. Id. And he ordered Fenkell to restore to Trachte the $2,896,000 he received in "phantom" stock proceeds from the 2007 transaction. Id.

Because Fenkell and Alliance were most at fault, the judge ordered them to indemnify the trustees. Id. at 950. In particular, the judge had this to say about Fenkell:

Id. at 946. Accordingly, the judge found that Fenkell "was far and away the most culpable party." Id.

Finally, the judge assessed prejudgment interest, awarded attorney's fees, and approved settlements between the plaintiffs and the Trachte ESOP trustees, and between the plaintiffs and Alliance.

Fenkell appealed, challenging various aspects of the remedial order, the award of attorney's fees, and the judge's approval of the settlements. The plaintiffs cross-appealed seeking a larger award of fees and costs against Fenkell.

In the meantime while we've had this case under advisement, Fenkell failed to comply with the order to restore the Alliance ESOP, so the judge found him in contempt. Fenkell appealed the contempt order as well. We've consolidated that appeal with the earlier ones.

II. Discussion

Although Fenkell does not challenge his liability, his appeal contests aspects of the judge's remedial order in an attempt to zero out the actual cost of his liability. The only significant legal issue is his challenge to the judge's indemnification order. The remaining issues, the issues raised in the plaintiffs' cross-appeal, and the challenge to the contempt order are more straightforward.

A. Indemnification/Contribution

The judge ordered Fenkell to indemnify Seefeldt, Mastrangelo, and Klute because his culpability vastly exceeded theirs. The judge found that Fenkell orchestrated their installation as trustees and directed their actions. And they in turn did his bidding, both because they were inexperienced as fiduciaries and because he called the shots as controlling owner, sole director, president, and CEO of Alliance. In short, Fenkell had authority over the Trachte trustees and used that authority and his control of the Alliance ESOP assets to orchestrate the inflated leveraged buy-out. As the judge analogized, "Fenkell was the unquestioned conductor and the Trachte [t]rustees mere musicians." Chesemore II, 948 F.Supp.2d at 949.

Fenkell doesn't meaningfully contest the judge's factual findings. He argues instead that ERISA doesn't permit the court to order indemnification or contribution among cofiduciaries.

Although ERISA contemplates the allocation of fiduciary obligations among cofiduciaries (thereby limiting subsequent losses), see 29 U.S.C. § 1105(b)(1)(B), it doesn't specifically mention contribution or indemnity as a remedy. Instead, it broadly permits the court to fashion "appropriate equitable relief" in response to a claim "by a participant, beneficiary, or fiduciary." Id. § 1132(a)(3). The Supreme Court has explained that "appropriate equitable relief" here means "those categories of relief that, traditionally speaking (i.e., prior to the merger of law and equity) were typically available in equity." CIGNA Corp. v. Amara, 563 U.S. 421, 439, 131 S.Ct. 1866, 179 L.Ed.2d 843 (2011) (internal quotation marks omitted).

In this context the Court has interpreted ERISA as generally incorporating the law of trusts. See id. (noting that ERISA "typically treats" a plan fiduciary "as a trustee" and a plan "as a trust"); see also Tibble v. Edison Int'l, ___ U.S. ___, 135 S.Ct. 1823, 1828, 191 L.Ed.2d 795 (2015) ("In determining the contours of an ERISA fiduciary's duty, courts often must look to the law of trusts."); Varity Corp. v. Howe, 516 U.S. 489, 497, 116 S.Ct. 1065, 134 L.Ed.2d 130 (1996) ("[W]e believe that the law of trusts often will inform, but will not necessarily determine the outcome of, an effort to interpret ERISA's fiduciary duties."); Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 110, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989) ("ERISA abounds with the language and terminology of trust law."); Cent. States, Se. & Sw. Areas Pension Fund v. Cent. Transp., Inc., 472 U.S. 559, 570, 105 S.Ct. 2833, 86 L.Ed.2d 447 (1985) ("[R]ather than explicitly enumerating all of the powers and duties of trustees and other fiduciaries [in ERISA], Congress invoked the common law of trusts to define the general scope of their authority and responsibility.").

Thus, the district court's remedial authority under ERISA includes the power of courts under the law of trusts, which vests in them the authority to fashion "traditional equitable remedies." CIGNA, 563 U.S. at 440, 131 S.Ct. 1866. Indemnification and contribution are among those remedies. See, e.g., Marine & River Phosphate Mining & Mfg. Co. v. Bradley, 105 U.S. 175, 182, 26 L.Ed. 1034 (1881) ("[T]he necessity of enforcing[] a trust, marshalling assets, and equalizing contributions[] constitutes a clear ground of equity jurisdiction."); Hatch v. Dana, 101 U.S. 205, 208, 25 L.Ed. 885 (1879) ("[I]f the capital stock should be divided, leaving any debts unpaid, every stockholder receiving his share of the capital would in equity be held liable pro rata to contribute to the discharge of such debts out of the funds in his own hands. This, however, is a remedy which can be obtained in equity only...."); Dupont De Nemours & Co. v. Vance, 60 U.S. 162, 175-76, 19 How. 162, 15 L.Ed. 584 (1856) (explaining the common-law development of contribution as a remedy in equity).

On the other hand, on the subject of fiduciary liability, ERISA says only that a fiduciary "shall be personally liable to make good to such plan" for a breach of his duties. 29 U.S.C. § 1109(a) (emphasis added). If a fiduciary is liable to restore an injured plan, this might imply that he cannot be liable to a cofiduciary. After all, a cofiduciary is not a plan.

We addressed this issue long ago and held that ERISA's grant of equitable remedial power and its foundation in principles of trust law permit the courts to order contribution or indemnification among cofiduciaries based on degrees of culpability. Free, 732 F.2d at 1137. Free involved a profit-sharing plan with two trustees; one fleeced the plan and the other did nothing. Id. The district court found the trustees jointly and severally liable because they both had breached their fiduciary duty. Id. But the court declined to order indemnification. We reversed, holding that ERISA includes the authority to order contribution or indemnification as allowed in the law of trusts. Id.

We noted in Free that § 1105(b)(1)(B) expressly allows fiduciaries to allocate various responsibilities between themselves and thereby insulate themselves from "liability for breaches of duties allocated to another trustee." Id. at 1337. This demonstrates, we said, that "Congress clearly did not intend trustees to act as insurers of co-trustees' actions." Id. The disputed question was not whether cofiduciaries may explicitly allocate and limit their liability under ERISA (they may), but rather whether the protections of § 1105 are the exclusive means of doing so. We concluded that they were not exclusive. We reasoned that "Congress intended to codify the principles of trust law with whatever alterations were needed to fit the needs of employee benefit plans." Id. at 1337-38. Because "[g]eneral principles of trust law provide for indemnification under appropriate circumstances," id. at 1338, we concluded that "courts [have] the power to shape an award so as to make the injured plan whole while at the same time apportioning the damages equitably between the wrongdoers," id. at 1337.

Fenkell argues that Free was "implicitly overturned" in Summers v. State Street Bank & Trust Co., 453 F.3d 404 (7th Cir. 2006). We disagree. True, Summers said in passing that "a right of contribution" under ERISA "remains an open [question] in this circuit." Id. at 413. But Summers did not mention Free, let alone disturb or overturn it. Summers apparently overlooked Free, which had already considered and decided the question. Regardless, Summers specifically said that the issue was "academic" in the context of that case, making its passing reference to contribution pure dicta. Id. at 412.

One judge in the Northern District of Illinois has supposed in dicta that Free has been overturned by the Supreme Court in Massachusetts Mutual Life Insurance Co. v. Russell, 473 U.S. 134, 105 S.Ct. 3085, 87 L.Ed.2d 96 (1985). See BP Corp. N. Am. Inc. Sav. Plan Inv. Oversight Comm. v. N. Tr. Invs., N.A., 692 F.Supp.2d 980 (N.D. Ill. 2010). In Russell the Court held that section 409 of ERISA entitles claimants to equitable relief making them whole under their benefits plan but does not allow recovery of extracontractual damages. The specific issue in Russell was whether a court may award damages for "mental or emotional distress" due to an ERISA violation. 473 U.S. at 138, 105 S.Ct. 3085. The Court said it may not.

Nothing in Russell undermines Free. Indeed, Free was decided specifically in the context of a section 409 action, through which the court fashioned an appropriate equitable remedy keyed to the plan in question. A cofiduciary seeking contribution or indemnification for a plan-related award is not analogous to a plan participant seeking extracontractual damages under an implied right of action for, say, emotional distress or pain and suffering. We think the district court in BP simply overread Russell.

We acknowledge, however, that the circuits are not uniform on the question of contribution and indemnification. Consistent with our holding in Free, the Second Circuit has long maintained that ERISA permits contribution. See Chemung Canal Trust Co. v. Sovran Bank/Maryland, 939 F.2d 12, 15-16 (2d Cir. 1991). The Eighth and Ninth Circuits disagree. See Travelers Cas. & Sur. Co. of Am. v. IADA Servs. Inc., 497 F.3d 862, 864-66 (8th Cir. 2007); Kim v. Fujikawa, 871 F.2d 1427, 1432-33 (9th Cir. 1989).

Fenkell hasn't given us any argument that wasn't already addressed in Free and resolved against his position. And overruling circuit precedent simply to move from one side of a circuit split to the other is disfavored. Buchmeier v. United States, 581 F.3d 561, 566 (7th Cir. 2009). Moreover, we're not convinced that Free was wrongly decided. If we are to interpret ERISA according to the background principles of trust law — as the Supreme Court has repeatedly instructed us to do — then indemnification and contribution are available equitable remedies under the statute.

Accordingly, the district court had the authority to order Fenkell to indemnify the new Trachte ESOP trustees. That remedy is within the court's equitable powers and is consistent with principles of trust law within which ERISA operates.

B. Fenkell's Fiduciary Status

Fenkell argues in the alternative that he can't be ordered to indemnify the trustees because he wasn't a cofiduciary. This argument is highly formalistic. It's true that Fenkell wasn't a trustee or other named fiduciary of the new Trachte ESOP. But the judge found that Fenkell used his position of authority over the Trachte trustees to control the assets spun off from the Alliance ESOP. He orchestrated the resignation of the old Trachte board, directed the creation of the new Trachte ESOP, and installed trustees who were both inexperienced and beholden to him. He then used his control over the trustees to implement a leveraged buyout at an inflated price, saddling Trachte with more debt than it could bear. The whole scheme was set up to ensure that the trustees would do his dirty work and he would keep his hands clean, at least as a formal matter. The judge saw through it, finding that the spin-off "was atypical both in its terms and the position of the parties."

Determining fiduciary status under ERISA is a functional inquiry. Larson v. United Healthcare Ins. Co., 723 F.3d 905, 916 (7th Cir. 2013) ("ERISA ... defines `fiduciary' not in terms of formal trusteeship, but in functional terms of control and authority over the plan, thus expanding the universe of persons subject to fiduciary duties.") (citations omitted). Even if Fenkell kept himself at a safe distance on paper, the whole of the deal was designed to occur only on terms favorable to him. It was arranged so that no one on the other side of the deal would look out for the interests of Trachte or its employees post-spin-off; indeed, the trustees of the new Trachte ESOP reported to Alliance and Fenkell. While Fenkell may not have been a fiduciary on paper, he effectively controlled both sides of the transaction. Either the spin-off and the leveraged buyout would go through together or neither would. That's why any involvement by a truly independent fiduciary looking after the Trachte interests would have scuttled the deal.

As a functional matter, then, Fenkell and Alliance were acting in a fiduciary capacity for the whole of the 2007 transaction, as the judge found. There was no error.3

C. Restoration Order

Fenkell also challenges the court's restoration order. Recall that there are really two classes of plaintiffs here. The main class consists of all participants in the new Trachte ESOP at any time from the transaction on August 29, 2007, to the time of class certification. The subclass comprises Alliance employees who participated in the Alliance ESOP at the time of the 2007 transaction and whose accounts were transferred to the new Trachte ESOP. The judge ordered restitution to the subclass in the amount of $7,803,543, which represents the value of the subclass's Alliance ESOP accounts as of the closing in 2007. Restitution to the main class was set at $6,473,856.82, which represents the amount the participants in the new Trachte ESOP overpaid for the Trachte stock minus the percentage representing the interests of the subclass (because their interests were accounted for in the separate restitution order).

The theory behind the judge's order was that there were two losses that needed restoration. The first is the overpayment in the leveraged buyout, which harmed the entire class. The second is the loss suffered by the subclass: plan participants who would have stayed with the Alliance ESOP or been rolled into a third-party buyer but for the spin-off to the new Trachte ESOP. In either alternative scenario, these participants would still have pension plans. For the subclass the 2007 transaction was the factual cause of their total loss, which is why the court ordered them restored to their 2007 level in the Alliance ESOP.

Fenkell argues that the subclass was only entitled to $1,893,650.61 — its share of the leveraged buyout overpayment. He says that any more would be a "windfall." This argument simply confuses the nature of the respective restitution orders. The subclass restitution order was separate from the class restitution order; the judge subtracted the subclass's share from the overpayment award precisely to avoid double recovery and windfalls.

D. Prejudgment Interest

Moving along, Fenkell mounts two feeble challenges to the award of prejudgment interest. His first claim is that because the plaintiffs assigned their rights to Alliance as part of their settlement and the settlement occurred before final judgment was entered, he is wrongly being required to pay prejudgment interest to a liable party. In other words, he argues that the award of prejudgment interest isn't actually making the plaintiffs whole because the interest accrued to Alliance from the time of settlement until the judgment was entered.

Fenkell cites no authority in support of the proposition that a prejudgment assignment of recovery halts the accrual of prejudgment interest. As a general matter, "[p]rejudgment interest ... is part of the actual damages sought to be recovered." Cement Div., Nat'l Gypsum Co. v. City of Milwaukee, 144 F.3d 1111, 1117 (7th Cir. 1998) (quoting Monessen Sw. Ry. Co. v. Morgan, 486 U.S. 330, 335, 108 S.Ct. 1837, 100 L.Ed.2d 349 (1988)) (emphasis added); see also Morrison Knudsen Corp. v. Ground Improvement Techniques, Inc., 532 F.3d 1063, 1077 (10th Cir. 2008) (calling prejudgment interest "an integral element of compensatory damages").

Here the award of prejudgment interest was a routine part of the plaintiffs' restitution remedy. The plaintiffs, in turn, assigned their right of recovery to Alliance in connection with the court-approved settlements. Alliance now stands in the plaintiffs' shoes. Nothing about the settlement or assignment halted the accrual of prejudgment interest.

Alternatively, Fenkell argues that the prejudgment-interest award amounts to overcompensation because the plaintiffs "reduced" their recovery when they settled. He insists that he should only be held liable for interest on the total damages minus the settlement amount — that is, interest on only about $60,000, which he says is the "actual" damages award.

Fenkell provides no support for this claim. The cases he cites — Zenith Radio Corp. v. Hazeltine Research, Inc., 401 U.S. 321, 91 S.Ct. 795, 28 L.Ed.2d 77 (1971), and Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947 (7th Cir. 1992) — stand for the unremarkable proposition that plaintiffs can't recover more than their actual total damages. The plaintiffs assigned their whole recovery to Alliance. The award of prejudgment interest does not violate this principle.

E. Settlement

Fenkell also objects to the settlements, arguing that the assignment of the plaintiffs' recovery affects his position in future litigation. "The general rule, of course, is that a nonsettling party does not have standing to object to a settlement between other parties. Particularly, non-settling defendants in a [multi]defendant litigation context have no standing to object to the fairness or adequacy of the settlement by other defendants." Agretti v. ANR Freight Sys., Inc., 982 F.2d 242, 246 (7th Cir. 1992) (internal quotation marks omitted). A nonsettling party has standing to object only "when the nonsettling party `can show plain legal prejudice resulting from the settlement.'" Jamie S. v. Milwaukee Pub. Sch., 668 F.3d 481, 501 (7th Cir. 2012) (quoting Agretti, 982 F.2d at 246). "That a settling defendant creates a tactical disadvantage for another defendant is not sufficient to support standing to object; the prejudice to the nonsettling defendant must be legal, such as (for example) interference with contractual or contribution rights or the stripping away of a cross-claim." Id.

The settlements do not prejudice Fenkell's interests in the sense required for standing to object. They do not interfere with any contractual or contribution rights he may have, nor do they eliminate any claim he has asserted in this suit. Fenkell has not established standing to challenge the settlements.

F. Attorney's Fees and Costs

We have cross-appeals before us on the issue of attorney's fees. The judge approved as reasonable almost $8 million in fees and ordered Fenkell to pay about $1.8 million of that total. This figure represents the portion of the approved fees that remained unpaid after the settlements, which included negotiated fee amounts to be paid by the Alliance defendants, the Trachte trustees, and the common settlement fund. These negotiated amounts covered some but not all of the $8 million in approved fees. Fenkell, the remaining liable defendant, was ordered to pay the balance.

District judges have considerable discretion in awarding attorney's fees under ERISA. Hardt v. Reliance Standard Life Ins. Co., 560 U.S. 242, 245, 130 S.Ct. 2149, 176 L.Ed.2d 998 (2010). A court may, in its discretion, award a reasonable attorney's fee "as long as the fee claimant has achieved `some degree of success on the merits.'" Id. (quoting Ruckelshaus v. Sierra Club, 463 U.S. 680, 694, 103 S.Ct. 3274, 77 L.Ed.2d 938 (1983)).

Fenkell makes no independent argument on the issue of attorney's fees. Instead, his challenge rests entirely on the success of his other claims of error. We've rejected every one of these arguments and need say no more.

The plaintiffs, for their part, argue that the judge's order shortchanges them because it confuses fees under section 502(g) of ERISA, which belong to prevailing plaintiffs, and class fees, which belong to their attorneys. See FED. R. CIV. P. 23(h). To the contrary, the judge plainly understood the distinction. Indeed, he said he appreciated the plaintiffs' argument in this regard but would not authorize recovery of fees in excess of the total amount he had approved as reasonable. He said that it would be difficult to differentiate between fees incurred for claims against individual defendants and also that fees were being paid through a complicated system of overlapping settlements and payments by multiple parties. He thought it best to play it safe and avoid redundant recovery.

It's clear to us that the judge fully grasped the difference between ERISA section 502(g)(1) awards and class-counsel awards under Rule 23(h) but simply decided not to award fees according to their separate legal bases because of the remedial complexities of the case. Instead, he set a total reasonable fee award — nearly $8 million — and ordered Fenkell to pay the amount that remained unpaid after the settlements. That cautious approach was not an abuse of discretion.

The same is true of the judge's refusal to assess costs against Fenkell. The plaintiffs asked for costs under ERISA section 502(g) and under Rule 54(d) of the Federal Rules of Civil Procedure. Under the rule "prevailing parties presumptively recover their costs." Loomis v. Exelon Corp., 658 F.3d 667, 674 (7th Cir. 2011). But as we noted in Loomis, "[b]oth [Rule 54(d)] and [section 502(g)] give the district judge discretion to decide whether an award of costs is appropriate," and costs and attorney's fees need not be awarded in tandem. Id. at 675.

Here, although the judge held Fenkell responsible for the attorney's fees that remained unpaid after the settlements, he declined to tax costs against him because the settlements had already covered the plaintiffs' costs in full. In other words, there were no unsatisfied costs to be paid. That was hardly an abuse of discretion.

G. Contempt

Finally, we come to Fenkell's appeal of the judge's contempt order. As we've noted, the judge's approval of the settlements resulted in some adjustments to the restoration order. As relevant here, the final judgment ordered Fenkell to restore $2,044,014.42 to the Alliance ESOP as restitution to the subclass. (This figure accounts for the portion covered by the settlements.) Fenkell neither complied with this order nor posted a bond. So while we've had this case under advisement, Alliance and the Alliance ESOP returned to the district court and initiated contempt proceedings.

After contentious discovery, extensive briefing, and protracted hearings, the judge found Fenkell in contempt. The proceedings were interrupted by Fenkell's premature appeals of several intermediate orders, which we dismissed for lack of jurisdiction. The contempt order is now final, so the issue is properly before us.

Based on abundant evidence, the judge found that Fenkell had substantial assets and "was actually taking affirmative steps to put his assets (at least technically) outside the reach of the [p]lan and other creditors." The evasive steps consisted mainly of transferring ownership of various accounts to his wife. But Fenkell maintained full control over these assets via power of attorney, and his wife testified that she was almost entirely ignorant of their financial affairs. Because Fenkell was fully capable of making the ordered restitution and persisted in failing to do so, the judge found him in contempt, gave him a deadline to comply, and backed up his order with a fine of $500 per day, doubling every seven days. The parties then negotiated the terms of a supersedeas bond, and Fenkell appealed the contempt order.

Fenkell does not challenge the judge's factual findings. Rather, he lodges a host of procedural objections to the contempt proceedings. He argues, for example, that Alliance and the Alliance ESOP lacked standing to pursue contempt sanctions. This argument is frivolous. The judgment requires Fenkell to restore money to the Alliance ESOP, and Alliance is the administrator of the plan. He also argues that it was error for the court to proceed under Rule 70(e) of the Federal Rules of Civil Procedure, which governs contempt, rather than Rule 69, which governs the enforcement of money judgments and incorporates the procedural and other protections of state execution law. This argument too is frivolous. It's well established that an equitable decree of restitution in an ERISA case may be enforced by contempt. See Cent. States, Se. & Sw. Areas Pension Fund v. Wintz Props., Inc., 155 F.3d 868, 876 (7th Cir. 1998); Donovan v. Mazzola, 716 F.2d 1226, 1239 n. 9 (9th Cir. 1983).

Fenkell's remaining arguments have been considered, are likewise frivolous, and do not require comment. The contempt order was procedurally and substantively sound.

AFFIRMED.