Hornady v. Outokumpu Stainless USA, LLC, 2024 U.S. App. LEXIS 25712 (11th Cir. Oct. 11, 2024):

*1 In this labor dispute, the district court ordered defendant Outokumpu Stainless to produce key time and pay records. For more than two years, Outokumpu begged for more time and promised both the court and the plaintiffs that it would produce the records—but time after time, it failed to comply. And as it repeated this pattern, Outokumpu began to paint its third-party payroll processor as the true culprit. Until, that is, the payroll processor caught wind of Outokumpu’s misrepresentations and corrected the record. Confronted with a merry-go-round of broken promises and blatant misrepresentations, along with an upcoming wage-and-hour trial for which no wages or hours were known, the district court issued the only sanction remaining in its arsenal: default judgment.

Outokumpu now turns to this Court for relief, but it will find none. First, we conclude that the district court did not err when it found Outokumpu’s intentionally subversive approach to discovery worthy of the sanction of last resort. Second, because district courts have plenary power to reconsider their interlocutory orders, we conclude that the court here did not abuse its considerable discretion when it declined to revisit the default judgment sanction. Third, we determine that the district court properly found that the plaintiffs alleged a sufficient basis for their claims, entitling them to relief on all counts. Fourth, and finally, we conclude that the record does not enable us to analyze the last issue raised—the district court’s application of the statute of limitations—and so we remand for more explanation.

I.

Outokumpu Stainless, USA, is the domestic subsidiary of Outokumpu Oyj—a multinational steel fabricator and manufacturer headquartered in Finland. It has operated a steel mill in Calvert, Alabama for over a decade. As Outokumpu admits, at least some of its employees are covered by the Fair Labor Standards Act, 29 U.S.C. § 201 et seq.

Four of those employees sued Outokumpu under the FLSA and Alabama common law. Alleging a series of wage-and-hour violations, they sought relief for themselves and all other similarly situated employees who opted in to the suit. Specifically, in Count I, the named plaintiffs alleged that Outokumpu violated the FLSA by (a) failing to pay wages for the entire time they were clocked in and working or available to work, (b) failing to pay overtime for all qualifying hours at one and one-half times an hourly rate that included monthly bonuses, and (c) failing to pay overtime on the regular payment date and instead paying overtime as “trued up” wages calculated at a later date. Count II alleges common law claims for quantum meruit, or unjust enrichment in the alternative. And Count III brings the same FLSA claims as Count I, but on behalf of the collective employees.

A.

Several of the FLSA’s requirements—and how they relate to Outokumpu’s pay practices—are relevant to understanding the plaintiffs’ claims. The FLSA requires covered employers to pay an employee “one and one-half times the regular rate at which he is employed” for hours worked beyond forty per week. 29 U.S.C. § 207(a)(1). The employee’s “regular rate” of pay must reflect all compensation, excluding overtime payments, that the employee ordinarily receives during the workweek. Id. § 207(e)(5). The correct balance between overtime and so-called “straight time” depends on multiple considerations, including the temporal relationship between an employee’s work schedule and the employer’s fixed workweek, whether the employer has paid the employee a nondiscretionary bonus, and whether the employer has adopted a “rounding policy” that creates discrepancies between the time employees are clocked in and the time for which wages are paid.

*2 Outokumpu’s pay practices implicated all of the above. As an employer subject to the FLSA, the company was required to “make, keep, and preserve” all records of its employees, including those related to “the wages, hours, and other conditions and practices of employment.” 29 U.S.C. § 211(c). Indeed, it would be nearly impossible for an Outokumpu employee to determine on his own whether he was correctly paid for his work. The pay rate changed based on the shift worked, the way time was rounded, the level of work, and the company’s monthly incentive plan. Whether overtime was paid correctly depended on these factors as well as how Outokumpu defined its workweek.

The paychecks themselves did not reflect this level of nuance—they categorized total straight time, overtime, and holiday pay, but did not identify any step-up rates. Nor did Outokumpu otherwise provide this information to its employees, who would have needed to record their own daily clock-in and clock-out times, along with the applicable pay rates, to confirm that they were paid appropriately for their overtime. Outokumpu was the only party with access to this information, which was necessary to understand if the “regular rate” of pay was calculated correctly. See 29 C.F.R. § 778.115.

Outokumpu’s records were similarly indispensable to evaluating the merits of the plaintiffs’ challenge to the rounding policy. Rounding clock-in and clock-out times is generally permitted, so long as the practice “will not result, over a period of time, in failure to compensate the employees properly for all the time they have actually worked.” 29 C.F.R. § 785.48(b). Outokumpu used two different rounding policies during the relevant time period, both of which the plaintiffs claim resulted in chronic underpayment. Neither policy can be adjudged without Outokumpu’s time and pay records.

Finally, whether the plaintiffs were paid properly also hinges on the precise details of Outokumpu’s incentive plan. According to this plan, Outokumpu pays its employees a nondiscretionary bonus every month. These incentive-based payments could implicate the calculation of an employee’s “regular rate of pay.” See 29 C.F.R. § 778.209. And when they did, any overtime would need to be recalculated to reflect the revised regular rate. Id. Here too, only Outokumpu had the information necessary to make that calculation.

The bottom line is that Outokumpu was the only party who had the evidence at the heart of the plaintiffs’ claims.

B.

Over the course of nearly two years, the district court ordered Outokumpu to produce the key pay, time, and incentive plan records twelve separate times. Faced with these orders, Outokumpu repeatedly told the court that it would produce the necessary records. But when the time came to comply, somehow Outokumpu always had another excuse—leaving the plaintiffs without any way to prove their case. To cap it all off, Outokumpu followed up on these discovery violations with outrageous misrepresentations to both the plaintiffs and the court.

At the suit’s inception in the fall of 2018, Outokumpu stipulated that it would promptly “produce a copy of all time sheets and payroll records” and assured the court that the parties were “exchanging information effectively.” This promise was memorialized in the first scheduling order, which required Outokumpu to produce—in Microsoft Excel format—daily clock-in and clock-out times, information about how it calculated bonuses, and detailed pay records that included the pay period and all applicable hourly rates. On the deadline, Outokumpu produced verified pay records for one batch of plaintiffs in an Excel file, and PDF spreadsheets of daily clock-in and clock-out times. Although the plaintiffs relied on these documents to estimate damages for a settlement conference, they would later learn that the records were full of false or incomplete data.

*3 After settlement negotiations failed, the court again ordered Outokumpu to produce the necessary time, pay, and bonus records. In a pattern of conduct that would repeat itself many times over, Outokumpu failed to comply by the deadline, assured the plaintiffs that its failure was not intentional but a product of “some sort of mistake, confusion, or misunderstanding,” promised to “provide responses as soon as reasonably possible”—and then did nothing to change its behavior. The company made no disclosures, offered no responses to discovery, and gave no estimate of when the information would be provided. These deficiencies prompted the plaintiffs’ first of many motions to compel. After Outokumpu failed to respond to the motion, the magistrate judge sanctioned Outokumpu by striking five of its affirmative defenses.

The sanction did not end Outokumpu’s streak of noncompliance—it again refused to produce the necessary records, convinced the plaintiffs and the court that it needed more time, and then failed to meet the extended deadline too. Faced with another motion to compel, Outokumpu stressed that court intervention was unwarranted and promised to provide the time and pay records “within a reasonable time period” if mediation failed to resolve the dispute. Mediation failed, and Outokumpu promised to deliver the records by the date of the next settlement hearing. Continuing the pattern, the settlement conference came and went without any documents being produced, and the plaintiffs filed yet another motion to compel. True to form, Outokumpu agreed to supplement its production but then failed to produce the records.

Outokumpu followed-up on that failure by attempting to move the goalposts—representing that, even after sixteen months of litigation, it would be too burdensome to produce the time and pay records without yet another extension. The magistrate judge, in a fit of generosity, accommodated this request and ordered Outokumpu to produce the records over the next three months.

During that time, the plaintiffs deposed Outokumpu’s payroll specialist and learned that the verified pay summaries and time records originally produced in 2018 were incorrect and incomplete—ultimately meaningless. The Outokumpu-created pay records did not show the correct pay rates, and there was no way to tell from the daily clock-in and clock-out records whether Outokumpu had applied its rounding policy or paid the employee for the full time.

The records that Outokumpu produced in response to the court’s latest order fared no better. For each employee and for each pay period, the spreadsheets produced included over 120 columns of data. Remarkably, none of these columns contained any information about pay rates—neither base rate of pay, nor step-up, holiday, overtime, or nighttime pay rates. At the next discovery hearing, the plaintiffs accused Outokumpu of bad faith, but Outokumpu again pleaded innocence and asked for another four months to produce the time and pay records. The conference ended with a new scheduling order designed to cure Outokumpu’s recurring discovery deficiencies the next month—now the ninth discovery order related to Outokumpu’s pay and timekeeping practices.¹

*4 Outokumpu, acting consistently if nothing else, flouted this order too, failing to produce the required time, pay, and incentive plan records by the June 1 deadline. The plaintiffs moved for sanctions in mid-June, incensed by the success Outokumpu’s repeated discovery violations had in stymying their case. For its part, Outokumpu maintained that all of the previous “issues with the production,” were inadvertent “errors or oversight,” insisting that there was “clearly not any willful or bad faith conduct.”

C.

While the plaintiffs’ latest motion for sanctions was pending, several troubling misrepresentations came to light. These misrepresentations centered on Outokumpu’s attempt to blame its third-party payroll processor, Automatic Data Processing, Inc., for its own discovery violations.

Outokumpu first raised the specter of ADP’s alleged misconduct in the fall of 2018, describing the requested records as “ADP pay records.” Then, after it was caught producing false records, Outokumpu again blamed ADP, arguing that part of the necessary records came “not from the company but from the ADP system.” And again—when faced with the possibility of sanctions—Outokumpu painted ADP as the real culprit and accused it of having “not been very helpful.”

That was just the start of the ADP narrative. Outokumpu successfully delayed the plaintiffs’ motion for summary judgment by representing that it was still waiting on information from ADP and that it was “just as frustrated as [the plaintiffs] on this particular situation.” Outokumpu later deflected the plaintiffs’ motion for sanctions by asking the magistrate judge for time to subpoena ADP. Then Outokumpo claimed the subpoena proved fruitless. When the plaintiffs renewed their motion for sanctions, Outokumpu once again shifted the blame onto ADP.

In reality, ADP had diligently complied with every request. ADP responded to the subpoena within one day, explaining that it did not ordinarily create time reports, but still offering PDF records of the limited data it had. It also explained that Outokumpu itself could create the required reports in Excel format using its payroll software.² Following these exchanges, ADP rightfully believed that it had complied with the terms of the subpoena. It did not know that Outokumpu had not produced any documents to the plaintiffs or that the plaintiffs had renewed their motion for sanctions in response to that continued failure of production.

Troubled by ADP’s apparent role in the never-ending discovery dispute, the district court instructed Outokumpu to confer with ADP in advance of a hearing set later that month. At that next hearing, Outokumpu reassured the court that it had indeed “reach[ed] out” to ADP but “ha[d] not gotten a response yet.” Although the court could not have known it, Outokumpu had not contacted ADP until three hours before the start of the hearing—despite having twelve days to comply. ADP, for its part, responded promptly; five hours after Outokumpu made contact, it offered to create a responsive Excel report customized for the case. No matter—Outokumpu never produced the PDFs or even admitted that ADP had offered them. Instead, Outokumpu expressed frustration at the hearing and told the court that “it would be ... helpful to hear [ADP’s] explanation to the Court.”

Believing—understandably—that ADP had intentionally flouted the subpoena, the district court entered a show cause order, directing an ADP corporate representative to appear fourteen days later. The court simultaneously directed Outokumpu to serve ADP with the order. It will likely come as no surprise that Outokumpu did not comply. First, it waited ten days to tell ADP about the hearing. And even then, it led ADP to believe that the hearing would be canceled if ADP could provide the necessary data before the hearing date.

*5 Believing Outokumpu’s representations, ADP produced an Excel report two days later that included pay data going back to 2018—everything still retained in its system. But not until the night before the hearing did Outokumpu tell ADP that it was still going forward as planned. The hearing marked the first time that ADP learned what Outokumpu had been telling the court all along, which left it understandably unprepared to answer the court’s questions. The court set another hearing for a week later.

At the second show cause hearing, ADP was able to correct the record and shine a light on Outokumpu’s misrepresentations. In a desperate attempt to avoid a “blame game,” Outokumpu tried to walk back its earlier statements, insisting that it had not attributed the discovery delays to ADP. The district court, with assistance from ADP, rejected Outokumpu’s attempts to rewrite the record. After hearing arguments on the issue, the district court informed the parties that “a sanction has to happen here.”

D.

Although the district court said that it was “loath to enter a default as to liability,” it did not see “any other recourse.” On November 18, 2021, the court formally entered default judgment on liability against Outokumpu under Federal Rule of Civil Procedure 37(b)(2)(A)(vi). After detailing Outokumpu’s never-ending discovery violations, the court found clear and convincing evidence that Outokumpu had acted in bad faith. A review of the record showed a “clear picture of willful and prejudicial discovery abuse,” including unrebutted evidence that Outokumpu doctored and produced false records, refused to produce other records it had previously agreed to produce, and made substantial misrepresentations to the plaintiffs and the court in an attempt to foist the blame for its discovery failures on anyone but itself. Finally, the district court concluded that no sanction other than default judgment would be sufficient to address Outokumpu’s bad faith.

Outokumpu fired their old counsel and moved for reconsideration. New counsel, however, appeared to continue singing from the same song sheet, telling the court that the old attorney had “accurately characterized the need for ADP’s help in producing documents in Excel format.” While that motion was pending, the parties disputed how damages should be calculated. Outokumpu raised a series of objections to the plaintiffs’ proposed damages calculation—some of which were sustained—and the district court settled on a final methodology. The district court denied the motion to reconsider and awarded the plaintiffs $13,171,958.56 in damages, $12,606,408.58 of which were for the collective FLSA action and $565,549.98 of which were for the 276 individual claims under Alabama law.

Both parties appealed. Outokumpu takes issue with the district court’s entry of default judgment as a sanction, denial of the motion to reconsider sanctions, and its determination that certain allegations made by the plaintiffs were well pleaded. In their cross appeal, the plaintiffs argue that the district court erred by limiting damages to those claims that fell within the FLSA’s statute of limitations.

II.

“This Court reviews sanctions orders for abuse of discretion.” J.C. Penney Corp. v. Oxford Mall, LLC, 100 F.4th 1340, 1346 (11th Cir. 2024). That standard means that “we review the district court’s legal conclusions de novo and its subsidiary factual findings for clear error.” Skanska USA Civ. Se. Inc. v. Bagelheads, Inc., 75 F.4th 1290, 1311 (11th Cir. 2023). We also review for abuse of discretion a district court’s decision whether to reconsider its own interlocutory order. Region 8 Forest Serv. Timber Purchasers Council v. Alcock, 993 F.2d 800, 805–06 (11th Cir. 1993). This Court reviews the sufficiency of the allegations in a complaint de novo. Eagle Hosp. Physicians, LLC v. SRG Consulting, Inc., 561 F.3d 1298, 1303 (11th Cir. 2009).

III.

*6 On appeal, Outokumpu does not argue that this conduct was not sanctionable—and how could it. Instead, it says that for two reasons the district court should not have awarded the ultimate sanction of default judgment. Neither carries the day.

First, Outokumpu contends that it should not be sanctioned for the conduct of its attorneys. This argument is wrong several times over. To start, it is generally “not an abuse of discretion to charge [parties] personally with ‘the consequences of the acts or omissions of their freely selected agent.’ ” Jochum v. Schmidt, 570 F.2d 1229, 1232 (5th Cir. 1978) (alteration adopted and quotation omitted).³ “There is certainly no merit to the contention that dismissal of petitioner’s claim because of his counsel’s unexcused conduct imposes an unjust penalty on the client.” Link v. Wabash R.R. Co., 370 U.S. 626, 633, 82 S.Ct. 1386, 8 L.Ed.2d 734 (1962). To hold otherwise “would be wholly inconsistent with our system of representative litigation.” Id. at 634, 82 S.Ct. 1386.⁴

What’s more, the district court sanctioned Outokumpu only after specifically finding that it had acted in bad faith. Indeed, the district court specifically “rejected the notion” that Outokumpu was “an innocent victim of its former counsel.” Outokumpu, not its attorney, was responsible for producing the time and pay records required for the plaintiffs’ suit. And Outokumpu, not its attorney, willfully and repeatedly refused to do so. In its nearly 100-page order, the district court painstakingly detailed every way in which both Outokumpu and its counsel intentionally sabotaged the judicial process. Outokumpu presents no evidence on appeal that the district court erred at all in making those factual findings, much less did so clearly. The harsh sanction of default judgment is appropriate here where Outokumpu itself is culpable. See Betty K Agencies, Ltd. v. M/V MONADA, 432 F.3d 1333, 1338 (11th Cir. 2005).

Second, Outokumpu argues that the district court should have imposed lesser sanctions because the plaintiffs eventually received all the records they sought. According to Outokumpu, ADP’s production of certain records after the second show cause order, eighth scheduling order, and fifth motion to compel cured any discovery violation and rendered the sanction of default unnecessarily severe.

To call that a remarkable contention is exercising a great deal of judicial restraint. To start, any cure would have been only partial because Outokumpu has still failed to produce every missing piece of evidence. It has not provided the plaintiffs with the time and pay records from 2015 to 2017—the very years when its exposure was greatest. Those records no longer exist, and it was Outokumpu who failed to maintain them. The district court determined that this failure was spoliation, and we have no qualms about that conclusion.

*7 Even if Outokumpu were correct that it had lived up to its discovery obligations, the district court’s sanction would not have outstripped the company’s disclosure violations. Generally speaking, sanctions of last resort—such as dismissals or default judgments—are appropriate only “when less drastic sanctions would not ensure compliance with the court’s orders.” Malautea v. Suzuki Motor Co., 987 F.2d 1536, 1542 (11th Cir. 1993). But “[w]hen lesser sanctions would be ineffective, Rule 37 does not require the vain gesture of first imposing those ineffective lesser sanctions.” Id. at 1544. Outokumpu points to no authority supporting its argument that eventual production of the required discovery displaces sanctions for earlier noncompliance. That ask-for-forgiveness-rather-than-permission rule would strip sanctions of any deterrent effect—malicious actors could repeatedly violate court orders with peace of mind, knowing that any future sanction could be nullified after the fact by simple compliance with the initial order.

 The record shows that the district court did not abuse its discretion by deciding that no sanction short of default judgment was appropriate. In fact, lesser sanctions had already been tried—and to no avail. After one of Outokumpu’s many discovery violations, the magistrate judge struck five of Outokumpu’s affirmative defenses. Neither this nor any of the many other orders against Outokumpu deterred Outokumpu’s “unrelenting campaign to obfuscate the truth.” See id. As the district court determined, Outokumpu’s “deceitful, subversive and manipulative conduct”—set forth above in unflattering detail—would not have changed with a sanction less harsh than default judgment. The district court properly considered and rejected lesser sanctions. It did not abuse its discretion by resorting to the sanction of last resort.

IV.

We next consider the district court’s denial of Outokumpu’s motion for reconsideration. That turns out to be a somewhat trickier question, if only because our precedents are less than clear about what standard of review district courts should employ when faced with such a motion in the context of a non-final order.

The answer is Rule 54(b), which governs a district court’s reconsideration of interlocutory orders. Entry of a non-final default judgment is one such order. Under Rule 54(b), district courts retain plenary power to reconsider an interlocutory order before the entry of final judgment. And when a district court exercises its power, this Court reviews only for abuse of discretion. Here, we find no abuse of discretion and affirm the district court’s denial of Outokumpu’s motion to reconsider sanctions.

A.

This Court has yet to precisely define the standards governing Rule 54(b) motions to reconsider. In our silence, some district courts have applied the more definite standards from Rule 59(e) or Rule 60(b). See, e.g., Reynolds v. Calhoun, 650 F. Supp. 3d 1272, 1275 (M.D. Ala. 2023) (collecting cases); Grier v. Griffin Moving & Storage, Inc., 452 F. Supp. 3d 1325, 1333 (S.D. Fla. 2017). But those standards—whether addressing newly discovered evidence or manifest errors under Rule 59(e), or mistake, inadvertence, fraud, and the like under Rule 60(b)—do not apply here. Rules 59(e) and 60(b) only come into play after a final, appealable judgment is entered. See Alcock, 993 F.2d at 806 & n.5; Toole v. Baxter Healthcare Corp., 235 F.3d 1307, 1315 (11th Cir. 2000).

Rule 54(b) is relevant at earlier stages of a case. It provides that an order “that adjudicates fewer than all the claims or the rights and liabilities of fewer than all the parties ... may be revised at any time before the entry of a judgment adjudicating all the claims and all the parties’ rights and liabilities.” Fed. R. Civ. P. 54(b). When, as here, a district court enters a non-final order, it should evaluate motions to reconsider that order under the standards inherent in Rule 54(b)—plenary authority “to reconsider, revise, alter or amend” a non-final order before the entry of final judgment. Toole, 235 F.3d at 1315 (quotation omitted); see also Hardin v. Hayes, 52 F.3d 934, 938 (11th Cir. 1995) (“[S]ince an order granting a new trial is an interlocutory order, the district court has plenary power over it” and may modify it “at any time prior to final judgment.” (quotation omitted)).

*8 A federal court’s inherent authority to correct its own errors, so long as it possesses jurisdiction over the action, is firmly rooted in the common law. See Doss v. Tyack, 55 U.S. (14 How.) 297, 312–13, 14 L.Ed. 428 (1852); Bucy v. Nev. Constr. Co., 125 F.2d 213, 216–17 (9th Cir. 1942). The Supreme Court too has long recognized that “if an interlocutory decree be involved, a rehearing may be sought at any time before final decree.” John Simmons Co. v. Grier Bros., 258 U.S. 82, 90–91, 42 S.Ct. 196, 66 L.Ed. 475 (1922); see also Dietz v. Bouldin, 579 U.S. 40, 46, 136 S.Ct. 1885, 195 L.Ed.2d 161 (2016) (“[T]he Court has recognized that a district court ordinarily has the power to modify or rescind its orders at any point prior to final judgment in a civil case.”). For that reason, until final judgment, a district court may “reconsider any portion of its decision and reopen any part of the case.” Marconi Wireless Tel. Co. of Am. v. United States, 320 U.S. 1, 47, 63 S.Ct. 1393, 87 L.Ed. 1731 (1943); accord 18B Charles Alan Wright & Arthur R. Miller, Federal Practice & Procedure § 4478 (3d ed. June 2024 update).

This inherent power was recognized—not changed—by Rule 54(b). See City of Los Angeles, Harbor Div. v. Santa Monica Baykeeper, 254 F.3d 882, 886–89 (9th Cir. 2001). Indeed, when Rule 54(b) was amended in 1946 to add that non-final orders are “subject to revision at any time” before final judgment (substantially the same language in today’s rule), the Committee explained that this change (along with allowing district courts to certify immediate appeals) “re-establish[ed] an ancient policy with clarity and precision.” Fed. R. Civ. P. 54(b) (1946); 1946 Advisory Committee Notes on Rule 54.

Though district courts enjoy plenary power to reconsider non-final rulings, they need not employ plenary review when doing so. Indeed, in most instances district courts should hesitate before revisiting their earlier interlocutory orders; important interests of finality, stability, and predictability underly that justifiable caution. Cf. United States v. Williams, 728 F.2d 1402, 1406 (11th Cir. 1984); Ins. Grp. Comm. v. Denver & Rio Grande W. R.R. Co., 329 U.S. 607, 612, 67 S.Ct. 583, 91 L.Ed. 547 (1947).

A district court’s discretion in these matters is governed by the law-of-the-case doctrine, which in this context functions as a guide for courts rather than “a limit to their power.” See Christianson v. Colt Indus. Operating Corp., 486 U.S. 800, 815–17, 108 S.Ct. 2166, 100 L.Ed.2d 811 (1988). Operating with more or less force depending on the stage of litigation, the doctrine as a whole “expresses the practice of courts generally to refuse to reopen what has been decided.” Id. at 817, 108 S.Ct. 2166 (quotation omitted). So district courts should gently keep in mind the general point that “when a court decides upon a rule of law, that decision should continue to govern the same issues in subsequent stages in the same case.” Arizona v. California, 460 U.S. 605, 618, 103 S.Ct. 1382, 75 L.Ed.2d 318 (1983).⁵ “Common sense, not a rigid set of rules,” governs when there is no higher court mandate. 18 James Wm. Moore et al., Moore’s Federal Practice § 134.21[1] (3d ed. June 2024 update); see also White v. Murtha, 377 F.2d 428, 431–32 (5th Cir. 1967).

A district court’s decision to reconsider an interlocutory order is thus committed to its sound judgment, which we review for abuse of discretion. See Alcock, 993 F.2d at 805–06. If the movant is able to meet a significantly higher showing for reconsideration—for example, the standards applicable to Rules 59(e) or 60(b), or the exceptions to the mandate rule—the district court should not hesitate to revisit its prior ruling. But just because a movant fails to meet these higher standards does not mean that reconsideration cannot be had—the district court may still have room to conclude that reconsideration is appropriate.

*9 On the flip side, a district court typically would not abuse its discretion when rejecting a motion to reconsider an interlocutory order if the movant simply rehashed arguments already considered and rejected. Cf. Am. Home Assurance Co. v. Glenn Estess & Assocs., Inc., 763 F.2d 1237, 1239 (11th Cir. 1985). And finally, the more time that has passed between a district court’s ruling and a party’s motion to reconsider that ruling, the less willing the court ought to be to entertain the party’s request. Parties must “be able to rely on the rulings that progressively direct proceedings toward trial.” 18B Wright & Miller, supra, § 4478.1.

In short, district courts have discretion to revisit their prior interlocutory orders, considering both the weight of the moving party’s arguments and the disruption that a change would cause in light of the time that has passed since the decision was initially made. Only when the district court abuses that considerable discretion will we set aside a Rule 54(b) decision.

Viewed against this backdrop, the district court here did not abuse its considerable discretion by denying Outokumpu’s motion to reconsider the sanction order. The district court recognized its plenary power to revisit its order and refused to do so for reasons well within its discretion. In fact, the district court (though understanding that it need not reconsider all aspects of its decision) entertained most of Outokumpu’s arguments in full. And Outokumpu—after years of misbehavior—simply failed to provide the court with an adequate reason to revise the sanction order. Instead, the company almost exclusively raised arguments that had already been rejected. The district court’s rejection of Outokumpu’s unashamed rehash was no abuse of discretion.

B.

Outokumpu’s principal argument on appeal is that the district court should have applied Rule 55(c)’s “good cause” standard to the motion for reconsideration, and it should have found that Outokumpu met that standard. That Rule provides that a “court may set aside an entry of default for good cause, and it may set aside a final default judgment under Rule 60(b).” Fed. R. Civ. P. 55(c). What Outokumpu fails to realize, however, is that the good cause standard applies to entries of default, not entries of default judgment. That may sound like a distinction without a difference, but it is not. And the second half of Rule 55(c)—directing courts to apply the Rule 60(b) standard—applies only to final default judgments. The order here is neither an entry of default nor a final default judgment; it is the entry of a non-final default judgment. The court entered judgment against Outokumpu but did not finalize the damages. Accordingly, Rule 55(c) does not govern.⁶

The text of Rule 55 draws a distinction between defaults and default judgments. Rule 55(a), titled “Entering a Default,” provides that “[w]hen a party against whom a judgment for affirmative relief is sought has failed to plead or otherwise defend ... , the clerk must enter the party’s default.” By contrast, Rule 55(b), titled “Entering a Default Judgment,” requires the clerk to “enter judgment ... against a defendant who has been defaulted for not appearing,” so long as “the plaintiff’s claim is for a sum certain or a sum that can be made certain by computation.” Otherwise, Rule 55(b) requires the plaintiff to “apply to the court for a default judgment,” which may be entered against a party if certain other conditions are met. So a Rule 55 “default” is when a party does not defend, and section (b) directs the court or the clerk to enter a default judgment when a party defaults on that basis. Rule 55(c) continues this distinction between defaults and default judgments by requiring courts to use a different standard when setting aside “an entry of default” as opposed to “a final default judgment.”

*10 The district court here never entered a Rule 55(a) default against Outokumpu. Instead, it relied on Rule 37(b)(2)(A)(vi), which authorized it to “render[ ] a default judgment against the disobedient party” as a sanction for disobeying “an order to provide or permit discovery.” Because the district court entered a default judgment against Outokumpu rather than a default, Rule 55(c)’s “good cause” standard does not apply.

The second half of Rule 55(c)—permitting courts to “set aside a final default judgment under Rule 60(b)”—is similarly inapt. A non-final default judgment is “an interlocutory order and thus not subject to being vacated under Rule 60(b).” Bon Air Hotel, Inc. v. Time, Inc., 426 F.2d 858, 862 (5th Cir. 1970) abrogated on other grounds by Little v. Liquid Air Corp., 37 F.3d 1069 (5th Cir. 1994). At the time Outokumpu moved for reconsideration, the entry of default judgment was not final because the amount of damages to be awarded was still outstanding. Guy v. Dzikowski (In re Atlas), 210 F.3d 1305, 1307 (11th Cir. 2000). Instead, the district court had only entered default judgment on Outokumpu’s liability. In 2015, Rule 55(c) was amended “to insert the word ‘final’ before ‘default judgment.’ ” 10A Wright & Miller, supra, § 2692. This change was intended “to make plain the interplay between Rules 54(b), 55(c), and 60(b).” 2015 Advisory Committee Notes on Rule 55. “A default judgment that does not dispose of all of the claims among all parties is not a final judgment unless the court directs entry of final judgment under Rule 54(b).” Id. So Rule 60(b)’s heightened standards “apply only in seeking relief from a final judgment”; until then, “Rule 54(b) allows revision of the default judgment at any time.” Id. (emphasis added).

V.

We now turn to Outokumpu’s final contention, that the district court should not have credited the plaintiffs’ legal arguments in support of several claims.

After a default, the defendant cannot contest the plaintiff’s factual allegations. Nishimatsu Constr. Co., v. Houston Nat’l Bank, 515 F.2d 1200, 1206 (5th Cir. 1975). But the same is not true for legal arguments; those can be challenged as not offering a sufficient basis for the judgment. Id. When deciding whether the claims are sufficiently pleaded, we apply the same analysis we use when evaluating Rule 12(b)(6) motions to dismiss. Surtain v. Hamlin Terrace Found., 789 F.3d 1239, 1245 (11th Cir. 2015). By that standard, the complaint must contain “sufficient factual matter, accepted as true, to state a claim to relief that is plausible on its face.” Ashcroft v. Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009) (quotation omitted). And a claim is plausible “when the plaintiff pleads factual content that allows the court to draw the reasonable inference that the defendant is liable for the misconduct alleged.” Id. Under 12(b)(6), as here, legal theories contained in a complaint are not accepted as true. Id. And that makes sense—defendants can forfeit the ability to contest the truth of facts asserted, but a default does not give the court powers beyond those established by law.

Outokumpu challenges the district court’s conclusion that the plaintiffs’ workweek and bonus claims were well pleaded. In the Third Amended Complaint, two of the plaintiffs’ allegations were that Outokumpu violated the FLSA by (1) failing to calculate its workers’ wages on a fixed, 168-hour workweek and (2) improperly structuring its incentive bonus plan such that the bonus percentage was not applied to the gross amount of pay the employee earned during the calendar month in which the bonus was calculated. Both allegations state plausible claims for relief.

*11 First, the workweek claim. The FLSA requires employees to be paid overtime at one and one-half times the regular rate of pay for any hours worked in excess of forty per workweek. 29 U.S.C. § 207(a)(1). The “workweek is a fixed and regularly recurring period of 168 hours—seven consecutive 24-hour periods”—that “need not coincide with the calendar week but may begin on any day and at any hour of the day.” 29 C.F.R. § 778.105. “Once the beginning time of an employee’s workweek is established, it remains fixed regardless of the schedule of hours worked by him.” Id.

The plaintiffs’ complaint alleges two separate violations of this workweek requirement. Each allegation—supported by testimony from Outokumpu’s payroll specialist—states a plausible claim for relief. One is that from 2015 until January 27, 2019, Outokumpu’s workweek ran from Monday to Sunday, but the company paid overtime based on the number of hours worked between a different sequence of days.⁷ That alone is a sufficient basis for judgment in their favor; if Outokumpu’s workweek was Monday to Sunday, but it calculated overtime based on the hours worked between Sunday and Saturday, miscalculations would have inevitably occurred.

The plaintiffs also allege that Outokumpu allocated all the time associated with an employee’s shift to a single workweek regardless of when the shift began or ended. So, for example, a shift that spanned the end of one workweek and the beginning of the next would have counted only toward the hours worked in the first workweek. This too presents a sufficient basis for judgment in favor of the plaintiffs on their workweek claim. If Outokumpu allocated all hours worked during a shift to only one workweek, even if that shift spanned multiple workweeks, overtime miscalculations would have been unavoidable.

Second, the bonus claim. The FLSA requires employers to calculate overtime pay as a function of the employee’s “regular rate.” 29 U.S.C. § 207(a)(1). The “regular rate” includes “all remuneration for employment,” save for several exemptions. Id. § 207(e). Sometimes when an employer pays a bonus, that bonus is considered part of the employee’s total remuneration for a given workweek, which means the employer must recalculate the overtime rate to take that bonus into account. 29 C.F.R. §§ 778.208, 778.209. Other bonuses, however, do not trigger a recalculation. Id. When a bonus is a “[p]ercentage of total earnings,” employers need not recalculate the overtime rate: a “bonus based on a percentage of total wages ... increases both straight time and overtime wages by the same percentage, and thereby includes proper overtime compensation as an arithmetic fact.” 29 C.F.R. §§ 778.210, 778.503. These bonuses, in other words, “provide for the simultaneous payment of overtime compensation due on the bonus.” Id. § 778.210.

Every month, Outokumpu paid its employees a nondiscretionary incentive bonus if the mill met certain production criteria. The plaintiffs allege that this bonus did not qualify as an FLSA-approved percentage bonus because the “percentage is applied to the gross amount of pay received by the employee during the calendar month” and not “the gross amount of pay the employee earned during the calendar month.” Paychecks received during a 30-day calendar month, for example, may have only covered 4 workweeks, or 28 days. And if the mill met production goals during that 30-day period, a bonus was applied to only 28 days’ worth of paychecks.

*12 The plaintiffs say that this mismatch between pay periods and calendar months means Outokumpu’s incentive bonus plan did not qualify as a percentage bonus. Because the bonus was neither an exemption under § 207(e) nor a proper percentage bonus, they argue, the FLSA required Outokumpu to recalculate the plaintiffs’ regular rate. And because Outokumpu failed to account for the incentive bonuses in the regular rate, Outokumpu failed to accurately calculate the overtime rate. As plaintiffs’ counsel explained to this Court, the rounding and bonus claims overlap—there is no way to separate the two.

Outokumpu tries to rebut the plaintiffs’ arguments, pointing to evidence uncovered during discovery that it says contradicts the plaintiffs’ stated facts. That evidence is irrelevant. The question here is whether the plaintiffs’ complaint states a plausible claim for relief. And in answering that question, we assume all well-pleaded factual allegations are true. See Surtain, 789 F.3d at 1245. Given the sparse factual record, we cannot say that the plaintiffs have failed to allege a sufficient basis for their claims. It may be that Outokumpu could have prevailed on the merits had it chosen to litigate this case, and in the process produced records that rebutted the plaintiffs’ factual allegations. But Outokumpu gave up that right, thumbing its nose at the judicial process and forcing the district court to rule on the issue at a premature stage in the litigation. The district court committed no error when it awarded judgment on the plaintiffs’ workweek and bonus claims.⁸

VI.

On cross-appeal, the plaintiffs argue that the district court improperly credited Outokumpu’s statute of limitations defense when it calculated the damages owed.⁹

The FLSA’s statute of limitations provides that an action must commence no later than three years after a willful violation of the statute. 29 U.S.C. § 255(a). For party plaintiffs the action commences on the date the complaint is filed, and for opt-in plaintiffs the action commences on the date their consent is filed in court. Id. § 256. Here, the class complaint was filed on July 30, 2018, meaning that only paychecks received on or after July 30, 2015, are actionable for named party plaintiffs. For any plaintiffs who later opted in to the class, the § 255(a) three-year statute of limitations would limit claims to only those paychecks received no more than three years before the opt-in date.

The plaintiffs also sought recovery for the later opt-in plaintiffs back to July 30, 2015. To get around § 255(a)’s bar, they argued that the statute of limitations should be equitably tolled. The district court rejected the attempt. In a normal case, that would be the end of it—but this is no normal case.

*13 The statute of limitations provided by § 255(a) is an affirmative defense, which means it must be specifically pleaded. Day v. Liberty Nat’l Life Ins., 122 F.3d 1012, 1014–15 (11th Cir. 1997); see also Fed. R. Civ. P. 8(c)(1). “[F]ailure to plead the bar of the statute of limitations constitutes a waiver of the defense,” even when the “complaint itself interjected the issue into the case.” Day, 122 F.3d at 1015. And a forfeited affirmative defense “is excluded from the case.” Wood v. Milyard, 566 U.S. 463, 470, 132 S.Ct. 1826, 182 L.Ed.2d 733 (2012) (alteration adopted and quotation omitted).

Here, Outokumpu asserted the statute of limitations defense in each of its answers. The magistrate judge first struck the defense from Outokumpu’s second amended answer as a sanction. And although Outokumpu re-pleaded the defense in its third amended answer, the district court struck that entire answer as part of its final sanction. As a result of these two sanctions, the statute of limitations defense was effectively unpleaded.

Even so, when the plaintiffs submitted proposed damages that included claims dating back to July 2015 for opt-in parties, Outokumpu objected, arguing that the scope of relief was inconsistent with the court’s earlier ruling rejecting equitable tolling and applying § 255(a)’s statute of limitations. Although the plaintiffs pointed out that any affirmative defense would have needed to be asserted in Outokumpu’s now-nonexistent answer, the district court sustained Outokumpu’s objection.

Reviewing the record, we cannot determine why the district court decided to sustain the objection. “A court has an obligation to assure that there is a legitimate basis for any damage award it enters.” Anheuser-Busch, Inc. v. Philpot, 317 F.3d 1264, 1266 (11th Cir. 2003). When a court enters default judgment, the award entered must be at the very least “a reasonable estimate of damages.” Adolph Coors Co. v. Movement Against Racism & the Klan, 777 F.2d 1538, 1544 (11th Cir. 1985). While the default judgment may not “exceed in amount what is demanded in the pleadings,” a court “may award substantially less tha[n] what was requested whenever the damage proceeding indicates that only a smaller award is warranted.” 10A Wright & Miller, supra, § 2688.

Although we review a district court’s award of damages under a clearly erroneous standard, the sparse record here renders this question incapable of meaningful appellate review. See Meader ex rel. Long v. United States, 881 F.2d 1056, 1060 (11th Cir. 1989). It could be the case that the district court limited the plaintiffs’ claims as an attempt to estimate damages more faithfully. Or perhaps the court rejected the plaintiffs’ equitable tolling arguments on the merits. Or perhaps in the flurry of the case it was not clear that the affirmative defense had been stricken. Or perhaps it was something else—the record does not say. Consequently, we remand to the district court to explain or reconsider its reasoning for sustaining Outokumpu’s objection on the statute of limitations defense.

* * *

The “most severe in the spectrum of sanctions provided by statute or rule must be available to the district court in appropriate cases.” Nat’l Hockey League v. Metro. Hockey Club, Inc., 427 U.S. 639, 643, 96 S.Ct. 2778, 49 L.Ed.2d 747 (1976). It is hard to imagine a more appropriate case than this one. Years of obstinance, dozens of discovery violations, and unceasing attempts to blame others finally caught up with Outokumpu. Even on appeal, the company displays a remarkable lack of contrition. The district court did not err when it entered default judgment, when it refused to reconsider that decision, or when it determined which of the plaintiffs’ claims were well pleaded. Accordingly, we AFFIRM those decisions. We cannot, however, properly evaluate the plaintiffs’ arguments about § 255(a)’s statute of limitations. We therefore VACATE AND REMAND the calculation of damages for the district court to more thoroughly explain its reasoning on that issue.