BAUER, Circuit Judge.

Mark Serafinn sued his local union and the joint council comprising leaders from his and other regional locals. He claimed that they impaired his free speech and assembly rights, fined him, and suspended his union membership without due process, in violation of the Labor Management Reporting and Disclosure Act (“LMRDA”) (also known as the Landrum-Griffin Act), 29 U.S.C. §§ 401 et seq. The district court granted summary judgment to the joint council, but a jury ruled against the local in favor of Serafinn.

The local appeals its losing jury verdict, arguing that the district court erred in denying it a mixed-motive jury instruction, in instructing the jury to consider witness Timothy Craig’s testimony about his DUI conviction for impeachment purposes only, and in instructing the jury not to consider the correctness of the joint council’s finding that he had violated the union’s work-referral rules. Serafinn cross-appeals the district court’s denial of his motion for relief from his summary-judgment loss against the joint council and its reduction of the attorneys’ fees award he won against the local.

We have reviewed, for an abuse of discretion, four preserved challenges (a fifth *911 challenge was waived) to the district court’s rulings. Finding no prejudicial error, we affirm.

I. BACKGROUND

The International Brotherhood of Teamsters is a labor union comprising 1.4 million members, ranging in occupation “from airline pilots to zookeepers.” See http:// www.teamster.org/contenVteamstersstructure (visited Feb. 16, 2010). Each member belongs to one of several hundred local unions that maintain substantial independence from the international organization. In regions with three or more locals, joint councils of leaders from those locals are set up to “help solve problems and decide some jurisdictional and judicial matters.” Id.

The Teamsters for a Democratic Union (“TDU”) is a “well-known and nationally active dissident faction” comprising thousands of Teamsters. Appellee’s Br. at 6; see http://www.tdu.org/whoweare (visited Feb. 16, 2010). In essence, the TDU functions as a rank-and-file political party within the international organization, opposed to the administration currently led by James P. Hoffa. See Appellee’s Br. at 7.

Mark Serafinn, a TDU member, served three terms as president of his local union, until he lost to Hoffa supporter Steven Mongan in 2001. Serafinn also lost his bid for vice president of his region’s joint council, composed solely of Hoffa supporters and led by president Keith Gleason. What happened next forms the nature of this dispute.

As Serafinn tells the story, his TDU politics were more than Hoffa supporters could bear. Mongan and Gleason colluded to have the local union and joint council bring internal disciplinary charges against Serafinn solely because he met with local union executives and published a newsletter accusing Mongan of cronyism, dumbness, and suppression of free speech. Their charges that he violated union rules by referring himself to a coveted power plant job ahead of others on the referral list were unevenly applied, because Hoffa supporters routinely broke the referral rules to reward their cronies with lucrative work without consequence, and baseless, because he visited the job site only for unpaid training. Mongan had told local union members not to show up at the job site just so that Serafinn could be blamed. The people ahead of him on the referral list were not even eligible because they were already working elsewhere, and two of them, including Timothy Craig, were further barred from complaining because they failed to show up at the power plant. The joint council’s hearing and review of charges that Serafinn violated the referral rules was a prejudging kangaroo court. The presiding officer Gleason was biased against Serafinn and had colluded with Mongan to bring the charges. Mongan told local union members to testify falsely before the joint council, or else they would lose their jobs. The joint council’s order that Serafinn pay restitution and be suspended from the union for six months unjustly penalized Serafinn and chilled union speech.

As the local union and joint council tell the story, Serafinn’s rule-breaking greed was more than anyone could bear. Serafinn assigned himself to work, not training, and even if he assigned himself to training, that also violated the referral rules. Referral-eligible, victimized coworkers in the local who should have got the power plant job — not Mongan and Gleason — brought the charges against Serafinn. The local referred the disciplinary proceeding to the joint council only because a majority of officers who would have presided at the local were also witnesses. Serafinn attended the hearing and was permitted to *912 introduce evidence, testify, and cross-examine witnesses. Gleason presided impartially and in accordance with ordinary rules of evidence and procedure. The panel disciplined Serafínn because he deserved it. The decision had nothing to do with Serafinn’s politics or exercise of free speech. The work-referral rules applied equally to everyone, especially to Serafinn who was in a unique leadership position at the time as lame-duck president.

The district court awarded summary judgment to the joint council. It found that Serafinn provided insufficient evidence for a jury to find that the joint council was involved in bringing the disciplinary charges against Serafinn, or that the joint council afforded Serafinn anything other than a full and fair hearing. Serafinn v. Int’l Bhd. of Teamsters, Local Union No. 722, No. 03 C 9409, 2007 WL 1670360, at *8, 13 (N.D.Ill. June 5, 2007).

But the district court denied the local’s motion for summary judgment, so Serafinn’s case against the local went to trial. Before trial, Serafínn proposed a jury instruction that would have required him to establish that retaliation was only a “motivating factor” in the local’s decision to prosecute the charges against him. Id. at *7. In response, the local proposed a “mixed-motive” jury instruction patterned after Mount Healthy City School District Board of Education v. Doyle, 429 U.S. 274, 287, 97 S.Ct. 568, 50 L.Ed.2d 471 (1977). In the proposed instruction, the local conceded that Serafinn had the initial burden to show that his exercise of free speech was merely “a motivating factor” in the local’s decision to prosecute him, as Serafinn requested. Serafinn, 2007 WL 1670360, at *6. But if the jury found that Serafinn met his burden, then the burden of proof would have shifted to the local to prove that it would have taken the same action even had Serafinn not exercised his free speech, thereby avoiding liability altogether (unlike a typical mixed-motive instruction that limits but does not erase all liability). The district court rejected both parties’ instructions, preferring instead a “but-for cause” instruction, which at all times kept the burden on the plaintiff to prove that the local prosecuted him solely because he exercised his free speech.

To begin the trial, Serafinn called Craig as a witness to prove that when Serafinn was at the power plant, Craig was ineligible to work there because he failed to show up. Craig testified that he didn’t show up because Serafinn had called him at home and told him not to. Serafinn rebutted that Craig’s absence was due to one of his three DUI convictions. Serafinn also argued that one of the later DUI convictions resulted in a revoked driver’s license but that the local nevertheless continued to refer driving-related work to him, thus supporting Serafinn’s theory that Craig was an individual similarly situated to Serafinn against whom the local selectively declined to enforce the referral rules. Serafinn failed to submit evidence showing that any of Craig’s DUI convictions were felonies or involved an act of dishonesty. At the close of trial, the district court instructed the jury to consider “the evidence that Timothy Craig has been convicted of a crime ... only in deciding whether Timothy Craig’s testimony is truthful in whole, in part, or not at all. You may not consider this evidence for any other purpose.” Serafinn, 2007 WL 1670360, at *8.

At the close of trial, the district court also issued the following instruction:

And I’m going to read a limiting instruction to you at this point. There have been a number of discussions from the witness stand and questions by these lawyers regarding the charges and the resolution of those charges. And I am *913 instructing you that in a separate proceeding a union body, known as the Joint Council 65, found that the plaintiff violated union referral practices and suspended and fined the plaintiff as a result. That was the union’s role. Whether the joint council’s decision was correct is not at issue for you in this case. Your role is to determine whether the defendants brought those charges against Serafinn in retaliation for his exercise of rights protected under the law. You are not supposed to be reviewing the decision.

Appellant’s Reply at 9 (quoting Tr. of Trial, p. 668).

A jury found the local liable to Serafinn for retaliating against his exercise of free speech. It awarded Serafinn $50,000 in compensatory damages and $55,000 in punitive damages.

After winning at trial, Serafinn moved for relief from the prior award of summary judgment for the joint council on the ground of newly discovered evidence adduced at trial against the local. The evidence consisted of Gleason’s testimony that within three months before the joint council brought charges against Serafinn, Gleason had met privately with Mongan, had telephoned the international organization, and had conducted a conference call with the joint council. The district court considered this evidence, finding nothing new to “reasonably demonstrate [the joint council’s] bias or role in retaliating against [Serafinn].” Id. at *5.

Serafinn’s attorneys submitted a claim to the district court that their time and expenses totaled $866,063 in this case and that they should be awarded as much. The district court decided that Serafinn should be awarded attorneys’ fees, but it reduced the claimed amount to $181,130.20, considering, among other things, the local’s ability to pay and that “Serafinn prevailed on only two of his four claims against only one of three defendants.” R. 365 at 3.

II. DISCUSSION

We review for an abuse of discretion each of the parties’ challenges to the district court’s decisions that were preserved: to not give a mixed-motive jury instruction, see Alcala v. Emhart Industries, Inc., 495 F.3d 360, 363 (7th Cir.2007) (citing Spiller v. Brady, 169 F.3d 1064, 1066 (7th Cir.1999)), to give a limiting instruction regarding Craig’s testimony, see id., to deny relief from its grant of summary judgment to the joint council, see Gomez v. Chody, 867 F.2d 395, 405 (7th Cir.1989), and to determine the amount of the attorneys’ fees award. See Hensley v. Eckerhart, 461 U.S. 424, 437, 103 S.Ct. 1933, 76 L.Ed.2d 40 (1983). (The local’s challenge to the limiting instruction regarding the joint council’s hearing was waived. We discuss the standard of review to that challenge later.)

Under the abuse-of-discretion standard, “deference ... is the hallmark.” Gen. Elec. Co. v. Joiner, 522 U.S. 136, 143, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). How much deference we afford a district court’s exercise of discretion varies inversely with the strictness of legal rules governing the exercise. When rules, standards, and precedents govern, the district court’s discretion is limited. When multifactor balancing tests and complex fact-determinations govern, the district court’s discretion is greater. See United States v. Approximately 81,454 Cans of Baby Formula, 560 F.3d 638, 641 (7th Cir.2009); Call v. Ameritech Mgmt. Pension Plan, 475 F.3d 816, 822 (7th Cir.2007); Metlyn Realty Corp. v. Esmark, Inc., 763 F.2d 826, 831 (7th Cir.1985).

*914 A. Mixed-Motive Instruction

We will come to the nature of the rule governing the district court’s denial of the local’s proposed “Mt Healthy ‘mixed-motive’ jury instruction.” Appellant’s Br. at 2. But first we note that the local actually argues for a burden-shifting instruction—not a mixed-motive instruction.

A mixed-motive instruction, by definition, follows a prior instruction that permits a plaintiff to establish liability merely by showing that the improper consideration was a “motivating factor” of the defendant’s contested action. Federal Civil Jury Instructions of the Seventh Circuit 3.01 (2008), cmt. c; see Mt. Healthy, 429 U.S. at 287, 97 S.Ct. 568; see also Desert Palace, Inc. v. Costa, 539 U.S. 90, 101, 123 S.Ct. 2148, 156 L.Ed.2d 84 (2003); Price Waterhouse v. Hopkins, 490 U.S. 228, 236-37, 109 S.Ct. 1775, 104 L.Ed.2d 268 (1989); Gooden v. Neal, 17 F.3d 925, 928 (7th Cir.1994). A mixed-motive instruction then permits the defendant to limit its liability if it can prove that it would have taken the contested action regardless of the improper consideration. See, e.g., Mt. Healthy, 429 U.S. at 287, 97 S.Ct. 568.

Here the district court rejected Serafinris proposed motivating-factor instruction, leaving no motivating-factor liability to limit, so there cannot have been the liability-limiting mixed-motive instruction that Mt. Healthy authorizes. Instead the district court instructed the jury that Serafinn had to prove that his exercise of free speech was not merely a motivating factor, but a necessary condition or a “ ‘but-for’ cause.” See Gross v. FBL Fin. Servs., Inc. ,—U.S.-,-, 129 S.Ct. 2343, 2350, 174 L.Ed.2d 119 (2009); see also United States v. Hatfield, 591 F.3d 945, 948 (7th Cir.2010).

So the local argues in essence that the district court should have substituted both a motivating-factor and a mixed-motive instruction for the but-for cause instruction the district court gave. But the local’s composite instruction would have been not only “longer” and “more complicated,” as the district court found, Serafinn, 2007 WL 1670360, at *7 (citing Federal Civil Jury Instructions of the Seventh Circuit 3.01, cmt. c), but also disadvantageous to the local if the evidence was in equipoise. Cf. Appellant’s Br. at 31 (“It was obviously a close case.”). Both the but-for cause instruction and the local’s proposed composite instruction score complete victory for the local if a jury finds that the local would have prosecuted Serafinn regardless of his outspoken politics. But whereas the but-for cause instruction maintains the burden of persuasion on the plaintiff, giving a tie to the local, the local’s proposed composite instruction shifts the burden of persuasion to itself, giving a tie to Serafinn.

So what we are really reviewing is the district court’s decision not to alter the burden of persuasion, which it was not permitted to do merely because of the local’s ill-advised request. See Gross, 129 S.Ct. at 2350-52 (finding that a statute with language similar to the LMRDA does not permit shifting of the burden of persuasion). The district court’s discretion was accordingly confined, and it therefore properly denied the local’s proposed motivating-factor/mixed-motive composite instruction.

The district court also properly denied Serafinris motivating-factor instruction. A mixed-motive theory of liability is never proper in a suit brought under the LMRDA, so any discretion the district court exercised in denying instructions supportive of a mixed-motive theory was thus limited. Some courts have found mixed-motive theories of liability proper in LMRDA cases, see Snyder v. Freight, *915 Constr., Gen. Drivers, Warehousemen, & Helpers, Local No. 287, 175 F.3d 680, 685-86 (9th Cir.1999); Waring v. Int’l Longshoremen’s Ass’n, Local 1414, 665 F.Supp. 1576, 1583 (S.D.Ga.1987), but the Supreme Court overruled their approach last year in Gross v. FBL Financial Services, Inc., 129 S.Ct. at 2348, which applies retroactively to this case. See Harper v. Va. Dep’t of Taxation, 509 U.S. 86, 97, 113 S.Ct. 2510, 125 L.Ed.2d 74 (1993). Mixed-motive theories of liability are always improper in suits brought under statutes without language comparable to the Civil Rights Act’s authorization of claims that an improper consideration was “a motivating factor" for the contested action. Id. at 2350 n. 3 (emphasis in original); see Serwatka v. Rockwell Automation, Inc., 591 F.3d 957, 961 (7th Cir.2010); Fairley v. Andrews, 578 F.3d 518, 525-26 (7th Cir.2009).

The LMRDA contains no such comparable language. It provides, in relevant part, that “[i]t shall be unlawful for any labor organization ... to fine, suspend, expel, or otherwise discipline any of its members for exercising any right to which he is entitled under the provisions of this chapter.” 29 U.S.C. § 529 (emphasis added). The word “for” means “by reason of,” 6 Oxford English Dictionary 25 (1989), and “because of.” Id.; Webster’s Third New Int’l Dictionary 886 (1981). These same definitions from the LMRDA persuaded the Supreme Court to find that the Age Discrimination in Employment Act’s parallel language requires that “the plaintiff bears the burden of persuasion to establish [a] ‘but-for’ cause” such that a mixed-motive theory of liability would be improper in any case brought under the statute. Gross, 129 S.Ct. at 2350-51. And these same definitions parallel the jury instruction the district court gave in this case.

In sum, the district court properly rejected a mixed-motive theory of liability, instructions that would have embraced such a theory, and an instruction that would have impermissibly shifted the burden of proof away from the plaintiff. Instead, it provided an instruction that “when considered in [its] entirety and not in isolation, [was] sufficient to inform the jury of the applicable law.” Alcala, 495 F.3d at 363.

B. Limiting Instruction Regarding Craig’s DUI Convictions

The local is on stabler ground in challenging the district court’s limiting instruction regarding Craig’s misdemeanor DUI convictions as erroneous, but it fails because it cannot show prejudice. Two federal rules of evidence govern the admission of Craig’s misdemeanor DUI convictions. Rule 609 prohibits their admission to attack Craig’s “character for truthfulness,” Fed.R.Evid. 609(a); see United States v. Papia, 560 F.2d 827, 845 (7th Cir.1977), while Rule 404 permits their admission for “other purposes,” Fed. R.Evid. 404(b), including to attack the credibility of a witness’ testimony by means other than attacking the witness’ general character for truthfulness, which is prohibited by Rule 609. See United States v. Lindemann, 85 F.3d 1232, 1243 (7th Cir.1996) (describing five different ways to impeach a witness). So evidence of Craig’s convictions was admissible, for example, to impeach Craig by contradicting the substance of his testimony or showing that he had an impaired ability to recall the event about which he was testifying. Id.

Had the district court so limited evidence of Craig’s convictions, we would find no error although the instruction limited evidence of Craig’s singular “crime,” whereas there was evidence that his *916 crimes numbered three. The court found that all three were relevant to “whether [Craig] was even truthful or able to recall,” App. B. at 8, so it does not matter which of the three DUI convictions the instruction referenced. The first “offered an entirely different reason” for Craig’s absence from the power plant from the one he had given. Serafinn, 2007 WL 1670360, at *8. As for the other two, the local does not explain how the district court erred in finding them admissible for proper impeachment purposes. Specifically, it does not claim that they failed to contradict anything Craig said and failed to call into question Craig’s ability to recall. “It is not the responsibility of this court to make arguments for the parties,” Vaughn v. King, 167 F.3d 347, 354 (7th Cir.1999), and even were we so inclined neither party provided us a copy of the trial transcript. So we are unable to find that the district court abused its discretion in finding that all three DUI convictions were admissible for proper impeachment purposes not barred by Rule 609.

But the district court abused its discretion by admitting the evidence for additional impeachment purposes that were improper. The instruction limited the jury to considering the convictions “in deciding whether Timothy Craig’s testimony is truthful.” This broad category of considering a witness’ testimony for truthfulness subsumes, as we have described, various possible impeachment methods, some of which were proper and others that were not. The instruction allowed the jury to find Craig’s testimony untruthful not only for the proper reason that evidence of his convictions contradicted his statements or called his memory into question, but also for the improper reason that evidence of his convictions called into question his general character for truthfulness. This impermissible use of evidence is exactly what Rule 609 prohibits.

But “even if the jury instruction was patently incorrect, [the local] still must establish that it was prejudiced by the improper instruction.” Gile v. United Airlines, Inc., 213 F.3d 365, 375 (7th Cir.2000). We will not order a new trial unless, “considering all the instructions, the evidence and the arguments, it appears that the jury was misled and its understanding of the issues was seriously affected to the prejudice of the complaining party.” EEOC v. AIC Security Investigations, Ltd., 55 F.3d 1276, 1283 (7th Cir.1995) (quoting McGeshick v. Choucair, 9 F.3d 1229, 1232 (7th Cir.1993)).

The local offers two reasons for why it was prejudiced by the limiting instruction. First, it claims the instruction drew attention to Craig’s convictions, thereby “casting the local ... (of which Craig was a member, and Serafinn was not) in the same unfavorable light.” Appellant’s Br. at 32. But this, claim fails because “[w]e assume that jurors follow[ed] the trial court’s instructions,” Bae v. Peters, 950 F.2d 469, 481 (7th Cir.1991) (citing Francis v. Franklin, 471 U.S. 307, 324 n. 9, 105 S.Ct. 1965, 85 L.Ed.2d 344 (1985)), specifically admonishing the jury not to consider Craig’s convictions for any purpose other than to impeach Craig’s testimony.

The local also claims prejudice because Craig’s convictions should have been admitted for the relevant, non-impeachment purposes of showing Craig’s ineligibility for referral at the power plant and the local’s selective enforcement of the referral rules for Craig and against Serafinn. But Serafinn, not the local, “wanted the jury to consider the evidence for the separate [non-impeachment] purpose.” Serafinn, 2007 WL 1670360, at *8. Indeed, the local spends a large chunk of its brief arguing why the evidence was relevant, but devotes nary a sentence to why the evidence was *917 relevant to its own case. See, e.g., Appellant’s Br. at 32-40. Nor does it explain how the instruction was anything other than advantageous to its case to the detriment of Serafinn. Without being able to show prejudice to its own case, no new trial is warranted.

C. Limiting Instruction Regarding the Joint Council’s Hearing

Finally the local argues that the district court erred by further instructing the jury to disregard the correctness of the joint council’s finding that Serafinn violated the referral rules in deciding the local’s motive for prosecuting him. See, e.g., Appellant’s Br. at 27, 32-33, 39; Reply Br. at 9-11. The district court found this argument “waived,” Serafinn, 2007 WL 1670360, at *5-6 (citing Kramer v. Banc of Am. Sec., LLC, 355 F.3d 961, 964 n. 1 (7th Cir.2004) (‘We have repeatedly made clear that perfunctory and undeveloped arguments that are unsupported by pertinent authority, are waived.”)), a finding that the local does not dispute on appeal, so unlike the other challenged decisions here we do not review this one for an abuse of discretion. Nor do we review the instruction for plain error, because the local has not bothered to explain why it would be proper for us to do so despite the local’s waiver in the district court. See City of Springfield, Mass. v. Kibbe, 480 U.S. 257, 259-60, 107 S.Ct. 1114, 94 L.Ed.2d 293 (1987) (discussing factors bearing on the propriety of reviewing a civil jury instruction for plain error); cf. City of Newport v. Fact Concerts, Inc., 453 U.S. 247, 256, 101 S.Ct. 2748, 69 L.Ed.2d 616 (1981) (allowing for plain-error review of civil jury instructions only in extraordinary circumstances). Arguments not raised on appeal are waived. See, e.g., United States v. Clinton, 591 F.3d 968, 971 (7th Cir.2010).

D. Denial of Relief from Summary Judgment

We move to Serafinn’s cross-appeal. Only Federal Rule of Civil Procedure 60(b) governed the district court’s denial of Serafinn’s post-trial challenges to the joint council’s award of summary judgment, R. 294, although Serafinn also moved to alter or amend the judgment under Rule 59 so that we might review his loss at summary judgment de novo. See Petru v. City of Berwyn, 872 F.2d 1359, 1361 (7th Cir.1989) (detailing the factors necessary to treat an appeal from a Rule 59 motion as an appeal from the underlying judgment). The problem with reviewing Serafinn’s motion under Rule 59 is that he filed it four months after the district court entered summary judgment, rather than within ten days as required by the rules then in effect, so we treat it as a Rule 60(b) motion for relief. See Kiswani v. Phoenix Sec. Agency, Inc., 584 F.3d 741, 742-43 (7th Cir.2009); see also Fed.R.Civ.P. 6(b)(2) (preventing a district court from extending the time to file a Rule 59 motion).

So our focus is on Rule 60(b), under which Serafinn needs to show by clear and convincing evidence that the district court would not have granted summary judgment had it known the additional evidence adduced at trial. See Gomez, 867 F.2d at 405 (finding that Rule 60(b)(2) requires an appellant to show that “the new evidence is likely to change the outcome”); Lonsdorf v. Seefeldt, 47 F.3d 893, 897 (7th Cir.1995) (finding that Rule 60(b)(3) requires the appellant to prove that fraud prevented the full and fair presentation of a meritorious case). In sum, Rule 60(b) requires at a minimum that the previously unavailable evidence be non-cumulative.

Serafinn’s claim fails because he cannot show that evidence revealed at trial against the local was anything other than *918 cumulative to the mix of information previously found by the district court to have created no genuine issue either as to the joint council’s role in retaliating against Serafinn or as to the joint council’s bias.

As for retaliation, the district court had considered the following information when it determined that Serafinn provided insufficient evidence for a jury to find that the joint council charged Serafinn because he exercised his right to free speech: “more than ‘some evidence’ ” that Serafinn had violated the referral rules, no evidence that anyone similarly situated as an executive officer had done so, and evidence demonstrating Serafinn’s “history of antagonism with Gleason.” Serafinn v. Int’l Bhd. of Teamsters, Local Union No. 722, No. 03 C 9409, 2006 WL 2497794, at *8, 10 (N.D.Ill. Aug. 28, 2006); Appellant’s Br. at 37 (quoting the trial transcript). We compare this with the additional evidence adduced at trial that before the joint council brought charges against Serafinn, Gleason had met privately with newly-elected Mongan to discuss Serafinn’s interference with Mongan’s leadership of the local, had called the international organization who then sent Bill Moore to help prevent the interference, and had conducted a conference call with the joint council discussing the interference. In conducting the comparison, we find that the latter set of evidence was cumulative to the former under both the direct and indirect methods of proving retaliation.

Under the direct method, Serafinn needed to show either direct evidence or a “convincing mosaic of circumstantial evidence” that the joint council charged him with violating the referral rules because he exercised his right to free speech. E.g., O’Neal v. City of Chicago, 588 F.3d 406, 410 (7th Cir.2009). None of Serafinn’s evidence is direct. And if a history of antagonism with Gleason did not provide a convincing mosaic of circumstantial evidence when the district court granted summary judgment, none is created now by overlaying evidence suggesting only that Gleason wanted Serafinn to stop undermining Mongan’s authority — not that Gleason wished to accomplish this end by actionable means.

Under the indirect method, Serafinn needed to show at least that he was treated differently from similarly situated union members who did not exercise their right to free speech. See, e.g., Ford v. Minteq Shapes and Servs., Inc., 587 F.3d 845, 848 (7th Cir.2009). Serafinn fails this test because the additional evidence shed no new light on whether the local had previously disciplined other executive officers for referral rule violations. Of course, we do not decide whether an entity may hold members in a “leadership position” to “a higher standard of conduct.” See Luks v. Baxter Healthcare Corp., 467 F.3d 1049, 1056 (7th Cir.2006) (finding this position “arguably” correct). We hold only that the district court considered certain evidence in reaching this position, Appellant’s Br. at 36-38 (quoting the trial transcript), and that additional evidence adduced later was cumulative for purposes of Rule 60(b).

So much for Serafinn’s claim that the additional evidence was non-cumulative evidence of retaliation. He also claims that the additional evidence was non-cumulative evidence of the joint council’s bias, because it revealed improper ex parte contacts and demonstrated that the joint council commingled its prosecutorial and adjudicatory functions. But the contacts between Gleason and Mongan that Serafinn calls “ex parte” took place three months before the disciplinary proceedings began, and Serafinn provides no evidence that the contacts tainted the proceedings much less related to them. Cf. Prof'l Air Traffic Controllers Org. v. Fed. Labor Relations Auth., 685 *919 F.2d 547, 565-66 (D.C.Cir.1982) (discussing relevant factors for finding that ex parte communications tainted an agency’s decision). Moreover, mere commingling of prosecutorial and adjudicatory functions are insufficient to show bias. See Withrow v. Larkin, 421 U.S. 35, 54, 95 S.Ct. 1456, 43 L.Ed.2d 712 (1975). Even if they were sufficient, the additional evidence demonstrates nothing new to indicate Gleason’s prosecutorial role in this case, as we discussed above, much less any new “specific factual allegations from which the operation of bias can be inferred.” See Frye v. United Steelworkers of Am., 767 F.2d 1216, 1225 (7th Cir.1985), rev’d on other grounds, Meyer v. Rigdon, 36 F.3d 1375, 1380 (7th Cir.1994).

E. Attorneys’ Fees

At last we address Serafinn’s challenge to the amount of attorneys’ fees the district court saw fit to award him. The rule governing a district court’s grant of attorneys’ fees lies in its equitable power to do so “when the interests of justice require.” Hall v. Cole, 412 U.S. 1, 4-5, 93 S.Ct. 1943, 36 L.Ed.2d 702 (1973). The district court was free to consider relevant factors including the nature of the attorneys’ work, the benefit of the suit to other union members, and the local’s ability to pay. See id. at 4-5, 93 S.Ct. 1943; Munson v. Friske, 754 F.2d 683, 697 (7th Cir.1985); see also Moriarty v. Svec, 429 F.3d 710, 717 (7th Cir.2005) (listing additional factors). Multiple equitable factors govern the crafting of an attorneys’ fees award, so “[district courts have wide discretion to determine what constitutes reasonable attorneys’ fees.” Moriarty, 429 F.3d at 717.

Serafinn cannot show that the district court abused its discretion. First, it cannot show that the district court erred in finding that the amount of time his attorneys spent was “not appropriate.” R. 365 at 4. Indeed, we see evidence of his attorneys inappropriately wasting everyone’s time in this court by unsupportedly contradicting facts found by the district court. Compare Appellee’s Br. at 14 (“Defendants’ claims are false that Serafinn’s referral of himself for training violated any union or referral rules.”), with Serafinn, 2006 WL 2497794, at *11 (“Plaintiffs own actions and testimony demonstrate that he needed a referral for training.”). More importantly, Serafinn makes no argument that the district court should not have considered the local’s ability to pay in crafting an appropriate award. Indeed, the local contends that even the reduced award, coupled with the judgment rendered in this case, exceeds its net worth. Appellant’s Br. at 10-11. Serafinn does not dispute this, dispute its relevance by showing that the international organization’s,net worth is the more relevant figure to examine, or explain why the district court otherwise erred in considering it. We have considered the district court’s analysis and find its award of attorneys’ fees entirely reasonable.

III. CONCLUSION

The local’s claims of error fail because they dispute jury instructions either that actually benefitted it or for which the local waived any challenge. Serafinn’s cross-appeal of the district court’s denial of relief from summary judgment fails because the new evidence he asks us to consider is cumulative, and his cross-appeal regarding his attorneys’ fees award fails because he neglected to refute the district court’s reasons in crafting it. The parties’ other arguments are without merit and warrant no discussion. Therefore, we Affirm.