Shipes v. Trinity Industries

*315 E. GRADY JOLLY, Circuit Judge: Forest Henry Shipes, a black male, was employed by Trinity Industries. After he was laid off, Shipes filed suit against Trinity under, inter alia, Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. Shipes alleged that Trinity’s all-white supervisory force had discriminated against him in decisions concerning his job placement, promotions, and lay-off. The district court certified a class that included all black hourly employees at two of Trinity’s plants. The trial was bifurcated, and after the trial on liability Trinity was found to have intentionally discriminated against black hourly employees in initial hiring, promotions, terminations, and lay offs. The district court then appointed an expert who determined damages to individual class members. After an award of attorneys’ fees to Shipes in the amount of $308,-238.05, judgment was entered. Trinity appeals. Still further, Trinity’s trial counsel appeals the district court’s imposition of personal sanctions against him pursuant to Fed.R.Civ.P. 37(b) in the amount of $3,000.00. I Trinity Industries manufactures railcars and structural steel products. Trinity operates over thirty production facilities in thirteen states, including two plants in Long-view, Texas. Shipes was employed by Trinity at its plant in East Longview as a welder’s helper from October 23, 1979, until he was laid off on June 30, 1980. On December 16, 1980, Shipes filed suit against Trinity under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e ei seq. Shipes alleged that Trinity’s all-white supervisory force had discriminated against him personally in decisions concerning his job placement, promotions, and layoff. In addition, the district court permitted Shipes to maintain a class action that included all black hourly workers employed at both of Trinity’s plants in Longview, Texas, between January 10, 1980, to April 1, 1984. The district court bifurcated the trial, and the liability portion was tried in 1984. 1 On October 10, 1985, the district court determined that Trinity intentionally discriminated against plaintiff class members in initial placement and pay and in layoffs at both Longview plants; intentionally discriminated against plaintiff class members in promotions and terminations at the plant in East Longview; and had discriminated against Shipes individually in initial placement and pay, promotions, and layoff. The district court then, sua sponte, appointed an expert to determine damages. On December 6, 1991, the expert submitted a model for damages, and on January 22, 1992, the district court entered partial final judgment and ordered that each class member have and recover from Trinity the amount determined by the court-appointed expert. Trinity had urged the district court to order class-wide relief, but instead the district court instructed the expert to determine damages on an individual-by-individual basis. The district court subsequently awarded Shipes $308,238.05 in attorneys’ fees. Trinity appeals, and Trinity’s trial counsel appeals the imposition of personal sanctions. II On appeal, Trinity challenges class certification, the computation of the back pay award, and the determination of attorneys’ fees. 2 Trinity first appeals the district *316 court’s certification of a class, arguing that Shipes should not have been allowed to represent blacks employed at a different plant from the plant at which he worked. Second, Trinity appeals the district court’s method of calculating the back pay remedy. Trinity argues that the district court erred in determining back pay awards on an individual-by-individual basis. Trinity contends that the district court should have ordered class-wide monetary relief — that is, it should have awarded each and every plaintiff class member his pro rata share of a pre-determined total monetary award. Third, Trinity appeals the district court’s determination of attorneys’ fees, arguing that the district court double counted factors and relied on erroneous considerations in enhancing the award. On the other hand, Shipes first argues that the district court did not err in class certification. Shipes also argues that the district court did not err in formulating the back pay remedy. Furthermore, Shipes argues that the district court correctly calculated the lodestar amount of attorneys’ fees and properly enhanced this amount. In a consolidated action, Trinity’s trial counsel, Robert Rader, Jr., contends that the district court abused its discretion by imposing personal sanctions against him in the amount of $3,000.00. Shipes argues that the sanctions were warranted because of Rader's failure to comply with discovery requests. Ill We first address Trinity’s argument that the district court should not have certified a class that included employees not only at the plant where Shipes was employed, but also employees at an additional plant. The district court’s certification of a class is reviewed under the abuse of discretion standard. Merrill v. Southern Methodist Univ., 806 F.2d 600, 607 (5th Cir.1986). Furthermore, the district court has wide discretion in deciding whether to certify a proposed class. Jenkins v. Raymark Indus., Inc., 782 F.2d 468, 471-72 (5th Cir.1986). Trinity argues that Shipes failed to satisfy the commonality and typicality requirements of Fed.R.Civ.P. 23(a), and thus it was an abuse of discretion for the district court to certify a class. 3 The district court determined that the two plants Trinity operated in Longview utilized the same subjective criteria in making personnel decisions; white supervisors at both plants applied the subjective criteria; employees were transferred between the two plants; the two plants had the same insurance plan, retirement programs, and administrative forms; and the two plants used the same Hourly Employee Handbook. The threshold requirements of commonality and typicality are not high; Rule 23(a) requires only that resolution of the common questions affect all or a substantial number of the class members. Jenkins, 782 F.2d at 472. Allegations of similar discriminatory employment practices, such as the use of entirely subjective personnel processes that operate to discriminate, satisfy the commonality and typicality requirements of Rule 23(a). Carpenter v. Stephens F. Austin State Univ., 706 F.2d 608, 617 (5th Cir.1983). The district court clearly did not abuse its discretion in certifying the class. IV A We next turn to Trinity’s argument relating to damages. The district court’s calculation of a back pay award is reviewed under the clearly erroneous standard. Pegues v. Mississippi State Employment Serv., 899 F.2d 1449, 1455 (5th *317 Cir.1990). At the outset, we observe that fashioning a class-wide back pay award is exceedingly complex and difficult, and the process is fraught with uncertainty. Pettway v. American Cast Iron Pipe Co., 494 F.2d 211, 260 (5th Cir.1974). Two general premises apply to the computation of a back pay award: (1) unrealistic exactitude is not required, and (2) uncertainties in determining what an employee would have earned but for the discrimination should be resolved against the discriminating employer. Claiborne v. Illinois Cent. R.R., 583 F.2d 143, 149 (5th Cir.1978), cert. denied, 442 U.S. 934, 99 S.Ct. 2869, 61 L.Ed.2d 303 (1979). Furthermore, the district court must be granted wide discretion in resolving ambiguities. United States v. United States Steel Corp., 520 F.2d 1043, 1050 (5th Cir.1975), cert. denied, 429 U.S. 817, 97 S.Ct. 61, 50 L.Ed.2d 77 (1976). B The district court ordered that back pay be determined on an individualized basis. The district court ordered that only each class member, whose beginning pay rate at the time he started working for Trinity was below the average beginning rate for white employees with the same qualifications, should receive a back pay award; the award raised the hourly rate of such individual class member by the difference between his beginning rate and the average white beginning rate. Trinity argues that this back pay remedy violates the fundamental rules of calculating such an award because it unfairly excludes some class members and grants a windfall to other class members. Trinity argues that the district court should have determined a total sum that represented the difference between black and white beginning rates as a whole, and then divided this amount to all class members on a pro rata basis. 4 Trinity argues that such an increase to class members would tend to eliminate the racial characteristics reflected in the distribution of pay rates; in other words, black and white employees in representative numbers would occupy the highest pay rates as well as the middle and lower pay rates. Trinity also argues — somewhat vaguely — that the district court’s formula for determining back pay for premature separation from employment was equally flawed. For premature separation from employment, the district court ordered a comparison between the actual length of service of each class member and the median length of service of white employees hired in the same time period. Each plaintiff class member whose service was less than the median length of service for white employees received an award equal to the difference. 5 Again, Trinity argues that the appropriate remedy is to calculate the total difference in the treatment of the classes and then award each class member his pro rata share. C On the other hand, Shipes defends the district court’s remedy. Shipes argues that the district court’s remedy was adopted in order to bring the pay level of each black employee who was paid less than the average salary of white employees with equal qualifications up to the level of that average. Shipes argues that the method pro *318 posed by Trinity is not intended to, and does not, compensate individual black employees to the extent to which they were underpaid relative to white employees with their qualifications; instead, Trinity’s method is concerned only with equalizing the amounts paid to whites and blacks as a group. Shipes argues that under Trinity’s plan, each class member would receive the same increment to his initial pay, regardless of the amount by which he was actually underpaid or regardless of whether he was underpaid at all. Shipes points out that some blacks were hired at rates equal to comparable whites. Thus, there is no basis to find that these black employees were discriminated against. Consequently, Shipes argues that Trinity’s proposed pro rata award to all black employees would in effect deduct from the wage rate awarded each black employee who was underpaid and, in effect, grant that money as a windfall to non-discriminatees. D Let us begin our evaluation of these arguments by observing that a Title VII class action suit imposes on the plaintiff a bifurcated burden of proof. First, the plaintiff must establish invidious class-based treatment; next, he must prove damages caused to class members by that illegal conduct. A finding of racial discrimination'against a class “does not necessarily mean that every member of the class is entitled to back pay.” Johnson v. Goodyear Tire & Rubber Co., 491 F.2d 1364, 1375 (5th Cir.1974). Once a prima facie case of discrimination against a class is made, a presumption of back pay arises in favor of all class members; this presumption does not, however, per se entitle a class member to back pay without some individual clarification. Pettway, 494 F.2d at 259. Once it has been determined that the class has been subjected to unlawful racial discriminatory practices, only those individuals who have suffered a loss of pay because of the illegal discrimination are entitled to compensation. Johnson, 491 F.2d at 1376. Title VII does not require a remedy for those not discriminated against. Gamble v. Birmingham So. R.R. Co., 514 F.2d 678, 686 (5th Cir.1975). The complexity of the case is a determining factor in what method the district court should utilize to formulate a back pay award. Pettway, 494 F.2d at 261. If the class is small, the time period short, or if the effect of the discrimination is straightforward, an individual-by-individual determination of what each claimant’s position would have been but for the discrimination is possible. Id. If, however, the class is large, the promotion or hiring practices are ambiguous, or the illegal practices continued over an extended period of time, a class-wide approach to the measure of back pay may be necessitated. Id. The process of computation is not, however, an “either, or” approach, and the determination of a back pay model is not a choice between one approach more precise than another. Id. at 261 n. 151. Methods for determining back pay possessed of superior certainty, such as the approach adopted by the district court, should be exhausted before resorting to racially drawn class-wide comparisons or pro rata approaches. United States v. Steel Corp., 520 F.2d at 1055. The fact that a class is large does not mean that pro rata relief should automatically be ordered. As noted earlier, not all members of a class are automatically entitled to back pay and there should be, if possible, a determination on an individual basis as to which class members are entitled to damages and the amount of such recovery. Johnson, 491 F.2d at 1375. E To reverse the district court, we would have to conclude that its calculations of back pay were clearly erroneous. On December 27, 1990, the district court entered an order that established the guidelines that the court-appointed expert was to follow in constructing the final model for back pay damages. With respect to discrimination occurring in wage rates upon initial hiring, the district court ordered that *319 back pay be awarded to particular class members based on the difference between the average starting wage paid to white employees with the class member’s qualifications and the actual starting wage paid to such class member. On July 29, 1991, the district court entered a second order relating to damage calculations. In this order, the district court addressed Trinity’s concern that calculations for the class should be determined as a total sum, and then damages awarded to each class member on a pro rata basis. The district court stated that Trinity’s proposal rested on the assumption that a class-wide finding of discrimination implies that each class member was affected equally by the discrimination and therefore damages should be determined by comparing the class of white employees with the class of black employees. The district court, however, expressly rejected this argument, stating that Trinity’s proposed method of awarding damages was inappropriate. Specifically, the district court stated that although Trinity was found liable for discrimination against the plaintiff class generally, it could not be assumed that such discrimination was uniformly felt by all class members. The back pay model adopted by the district court has two sections. First, for back pay remedying discrimination in initial starting salary, each individual class member was compared to white employees who had the same educational background and work experience. If a class member was paid a starting rate below the average starting rate of white employees with the same qualifications, the class member was awarded that difference as back pay. Second, for back pay in premature separation from employment, each individual class member was compared to white employees who were hired in the same time period. An average length of employment for these white employees was determined. If a class member was laid off or terminated before the average length of employment for the group of comparable white employees, he was awarded the difference as back pay. Under the back pay formula relating to initial pay rates, those individual class members who were paid starting rates above the average starting rates of white employees with the same qualifications received no back pay award. Similarly, those individual class members who were not laid off or terminated prior to the average length of employment for white employees hired in the same time period received no back pay award. A Title VII plaintiff, however, is not entitled to recover an amount greater than his pecuniary loss. Pegues, 899 F.2d at 1457. Because these class members did not suffer pecuniary losses as a result of Trinity’s discrimination, it follows that they are not entitled to an award of back pay. After carefully reviewing the record and the back pay remedy ordered by the district court, it is our opinion that the remedy ordered was not clearly erroneous. The district court’s back pay remedy is therefore affirmed. V A We now address Trinity’s argument concerning the attorneys’ fees awarded to Shipes. The district court’s determination of attorney’s fees is reviewed for abuse of discretion, and the findings of fact supporting the award are reviewed for clear error. Von Clark v. Butler, 916 F.2d 255, 258 (5th Cir.1990). As the first step in determining the amount of attorneys’ fees to award, the district court must determine the compensable hours from the attorneys’ time records, including only hours reasonably spent. Alberti v. Klevenhagen, 896 F.2d 927, 930 (5th Cir.), vacated in part, 903 F.2d 352 (5th Cir.1990). As a second step, the district court must select an appropriate hourly rate based on prevailing community standards for attorneys of similar experience in similar cases. Id. The number of compensable hours is then multiplied by the selected hourly rate to produce the “lodestar” amount. Shipes sought recovery of attorneys’ fees for 1,306.88 hours of time at a rate of $175.00 per hour for lead counsel *320 and $150.00 per hour for associate counsel. The district court, after determining beyond doubt that the plaintiffs had indeed prevailed on the central issue of the lawsuit and were thus prevailing parties, ordered payment of 100% of the time requested at a rate of $165.00 per hour for lead counsel and $140.00 per hour for associate counsel, for a total fee award of $144,712.70. The district court stated that only four of the Johnson v. Georgia Highway Express, 488 F.2d 714 (5th Cir.1974), factors were subsumed in the lodestar amount: (1) time and labor required; (2) customary fee; (3) counsel’s experience and ability; and (4) awards in similar cases. 6 After determining that the hours accurately reflected the time expended in this action, the district court further determined that the rates proffered by Trinity were too low, but that Shipes’s attorney had failed to show that attorneys with abilities comparable with her actually received $175.00 per hour. For this reason, the district court ordered payment at the hourly rate of $165.00, which represented the amount recently applied by the district court for work performed by lead counsel in employment discrimination cases. The district court did not err in its calculation of the lodestar amount. The district court reviewed the attorneys’ time records and found them to be more than adequate. The district court did not simply accept the hourly rate suggested by Shipes’s attorneys, but instead lowered it to an hourly rate that had been applied in the community for similar cases. The district court further ordered that the hourly rates for the attorneys be reduced for the time the attorneys spent traveling. We find that the district court did not err in its determination of the lodestar amount of $144,-712.70. B After determining the lodestar amount, the district court may adjust the lodestar up or down in accordance for relevant Johnson factors not already included in the lodestar. After calculating the lodestar amount, the district court must then apply the remaining Johnson factors to determine if the lodestar should be adjusted; the district court must be careful, however, not to double count a Johnson factor already considered in calculating the lodestar when it determines the necessary adjustments. Von Clark, 916 F.2d at 258. Furthermore, the district court must explain with a reasonable degree of specificity the findings and reasons upon which the award is based, including an indication of how each of the Johnson factors was applied. Id. Four of the Johnson factors— the novelty and complexity of the issues, the special skill and experience of counsel, the quality of representation, and the results obtained from the litigation — are presumably fully reflected in the lodestar amount. Alberti, 896 F.2d at 930. Although upward adjustments of the lodestar figure based on these factors are still permissible, such modifications are proper only in certain rare and exceptional cases supported by both specific evidence on the record and detailed findings by the lower courts. Id. After calculating the lodestar amount, the district court enhanced the fee on five additional Johnson factors. First, the district court determined that the lodestar amount should be enhanced for the novelty and difficulty of the case, stating that this action involved representation of over three hundred employees at two manufacturing plants, challenged the entire spectrum of Trinity's employment decisions, and involved complex, highly technical data. Second, the district court enhanced the lodestar amount based on the skill required, stating that Shipes’s claim necessitated ex *321 tensive analysis, which Shipes’s counsel competently presented. Third, the district court enhanced the lodestar amount for the preclusion of other employment, stating that Shipes’s counsel’s total time commitment to this case for several months adequately demonstrated that this case was inordinately time-consuming and precluded the acceptance of other requests for representation. Fourth, the district court enhanced the lodestar amount based on special time limits imposed, stating that Shipes’s counsel’s claim of nearly total time commitment before trial supported an upward fee adjustment because of the exceptional skill needed to present the last-minute data in an intelligible manner. Fifth, the district court enhanced the lodestar amount based on the amount of money involved and the obtained, stating that the outcome of this lawsuit represented a tremendous victory based upon the large number of claimants and the probable enormity of the relief. 7 On December 7, 1987, the district court ordered an 80% enhancement of the lodestar award based on these five Johnson factors, increasing the attorneys’ fee award from $144,112.70 to $260,482.86. 685 F.Supp. 607. C We find that four of these five Johnson factors used by the district court to enhance the lodestar amount — the novelty and difficulty of the case, the skill required, special time limits imposed, and preclusion of other employment — are unsupported, and thus enhancement based on these factors was unwarranted. We do think, however, that enhancement due to the results obtained may be warranted. The district court enhanced the lodestar amount based on the novelty and difficulty of the ease because it found that there were over three hundred plaintiffs, an entire spectrum of employment decisions was being challenged, the case was complex and highly technical, and Trinity’s obstinate conduct caused additional difficulty. These factors — not uncommon in much present-day litigation — simply do not render a case “rare” or “exceptional” for purposes of enhancing the lodestar amount. All counsel competent to handle a case such as this one are expected to be able to deal with complex and technical matters; this expertise is reflected in their regular hourly rate, based on fees for counsel of similar experience and ability. Still further, the difficulty in the handling of the ease is adequately reflected in the number of hours billed — hours for which the attorney is compensated in the lodestar amount. Similarly, obstinate conduct by opposite counsel is compensated by the additional number of hours that are required to prevail over such obstinacy. The district court’s enhancement based on two other factors — the skill required and special time limits imposed— was, we think, also unwarranted as each is accounted for in the lodestar amount. Even though Shipes’s counsel presented extensive statistical data with competence, nothing less should be expected of counsel; consequently, this factor alone does not support enhancement. We also find as an unwarranted basis for enhancement “special time limits imposed” by the defendant’s necessitating that Shipes’s attorney evaluate last-minute data shortly before trial. Again, we emphasize that this factor is not an abnormal occurrence — especially in a trial involving statistics and complex data — and is accounted for by the additionally required hours that are reflected in the lodestar. We also conclude that the district court improperly enhanced the lodestar based on the preclusion of other employment. The district court found that Shipes’s counsel was totally committed to this case for several months prior to trial and concluded that this demonstrated that counsel was precluded from accepting other requests for representation. The district court did not make a finding that Shipes’s attorney had indeed been required to refuse other employment because of this *322 case, and enhancement based on this factor is therefore unsupported. Furthermore, this factor will ordinarily be subsumed within the lodestar amount: If, for example, Shipes’s attorney worked on nothing but this case, then this potential loss of income in refusing other employment is compensated for in the number of hours she billed in the instant case. 8 We do think, however, that the district court may have been warranted in enhancing the lodestar amount because of the amount involved and results obtained. 9 Shipes’s victory was complete on all issues. Furthermore, the victory resulted in a substantial award of monetary damages for class members — plus, and very importantly, future protection against discrimination in the form of injunctive relief. For these reasons, enhancement of the lodestar amount on account of the results obtained may have been warranted. On remand, the district court must determine whether it is customary in the area for attorneys to charge an additional fee above their hourly rates for an exceptional result after lengthy and protracted litigation. If Shipes’s attorneys can demonstrate this area custom, the district court will be warranted in enhancing the lodestar in an appropriate amount based on this factor. D On August 4, 1988, the district court ordered an additional 33% enhancement of the lodestar amount based on the risk of not prevailing, or the contingent nature of the case. The district court found that the clear improbability of plaintiffs’ ability to find counsel to represent them in similar actions within the district warranted the contingency enhancement. In Islamic Center of Mississippi v. Starkville, 876 F.2d 465 (5th Cir.1989), this court adopted the approach to enhancement for the contingent nature of the case set forth in Justice O’Connor’s concurrence in Pennsylvania v. Delaware Valley Citizens’ Council (Delaware Valley II), 483 U.S. 711, 107 S.Ct. 3078, 97 L.Ed.2d 585 (1987). In City of Burlington v. Dague, — U.S.-, 112 S.Ct. 2638, 120 L.Ed.2d 449 (1992), however, the Supreme Court rejected Justice O’Connor’s approach. In Burlington, the Court first noted that, “[although different fee-shifting statutes are involved, the question is essentially identical” to the question it had addressed, but which it had not resolved in Delaware Valley II: whether a court may enhance the fee award above the lodestar amount in order to reflect the fact that the party’s attorneys were retained on a contingent-fee basis and thus assumed the risk of receiving no payment at all. In Burlington, the Court was urged to adopt the approach set forth by Justice O’Connor in Delaware Valley II. The Court expressly declined to do so, however, stating that “we do not see how it can intelligibly be applied.” Burlington, — U.S. at-, 112 S.Ct. at 2642. The Court instead adopted the approach reflected in Justice White’s plurality opinion in Delaware Valley II and held that enhancement for contingency was not permitted under the fee-shifting provisions of the Solid Waste Disposal Act and the Clean Water Act. Burlington, — U.S. at- -, 112 S.Ct. at 2643-44. Thus, Burlington eviscerates our holding in Islamic Center in which we stated that Justice O’Connor’s position was “the only view to which a majority of the [Supreme] Court, and perhaps even the whole [Supreme] Court could subscribe.” Islamic Center, 876 F.2d at 471. Burlington now tells us, however, how wrong we were. Accepting the error of previous thinking, *323 and following the clearly lighted path of Burlington, we now hold that the contingent nature of the case cannot serve as a basis for enhancement of attorneys’ fees awarded to prevailing plaintiffs under traditional fee-shifting provisions. Accordingly, it was error for the district court to enhance the lodestar amount by 33% based on this factor. E In summary, we conclude that the district court correctly calculated the lodestar amount to be $144,112.70. We also conclude that one of the enhancement factors applied by the district court — the results obtained — may have been warranted; we therefore remand this issue to the district court for a determination of whether this factor is supported by custom in the legal community and, if so, the appropriate amount. VI. We now turn to the issue of sanctions which were accessed against Trinity’s trial counsel, Robert Rader, Jr., pursuant to Rule 37(b), the appeal of which has been consolidated with this action. The imposition of sanctions is a matter of discretion for the district court; we review only for abuse. Frame v. S-H, Inc., 967 F.2d 194, 202 (5th Cir.1992). We hold that the district court did not abuse its discretion in sanctioning Rader. In its order of March 5, 1984, the district court noted that Trinity had been recalcitrant in relation to the discovery sought by Shipes. As early as November 24, 1981, the district court observed that Trinity’s entire posture was infected with a tone of indifference and disrespect for clearly established rules of procedure. The district court warned that such lassitude would not be countenanced. On February 12, 1982, the district court stated that the recurring nature of Trinity’s indifference to time limits substantially limited the credibility of Trinity’s protestations of good faith. The district court further stated that Trinity’s objection to and failure to answer two interrogatories directly violated an earlier court order; the district court cautioned that Trinity’s failure to answer was tantamount to contempt of court. In 1982, the district court also noted that Trinity continued to adhere to its posture of disobedience to court orders and again warned that this attitude would not be countenanced further. At this point, the district court cautioned Rader that Rule 37(b) prescribed harsh actions against parties who failed to comply with orders concerning discovery. Notwithstanding these warnings, still Shipes had to file numerous other motions to compel, which the district court granted. The district court held a hearing on March 16, 1984, to allow Rader to appear to show cause why he should not be sanctioned. After the hearing, the district court concluded that there had been a deliberate effort by Rader to delay discovery. The district court did not, however, impose any sanctions at that time, but stated that it did not expect to be presented with any additional problems concerning discovery. After the hearing, Shipes was given the information to which Rader had referred. Then, however, on March 20, 1984, Shipes filed yet another motion for sanctions. At this time, the district court awarded sanctions. The court found that the information given to Shipes by Rader after the earlier hearing did not represent most of the outstanding discovery and that Rader had been disingenuous in his representations to the court that he had complied with the court’s orders. The district court did not abuse its discretion in sanctioning Rader. On the contrary, the district court gave Rader several warnings and on numerous occasions cautioned Rader to comply with discovery requests. The court did not err in concluding that Rader had failed to heed its warnings and had failed to comply in good faith. Neither did the court err as to the amount of $3,000.00. Under Rule 37(b), Rader may be personally liable for reasonable expenses, including attorneys’ fees, caused by his failure to comply with a discovery order. Batson v. Neal Spelce Assoc., Inc., 765 F.2d 511, 516 (5th Cir.1985). The affidavit presented by Shipes’s attorney along with the January 23, 1984 motion for sanctions showed attorneys’ fees and expenses in the amount of $1,617.50. On March 16, 1984, Shipes’s attorneys had to attend a *324 hearing for Rader to show why he should not be sanctioned, and on March 20, 1984, Shipes’s attorneys had to prepare yet another motion for sanctions. It was reasonable for the district court to assume that the expenses and fees involved in the second motion were similar to those incurred in preparing the first motion. We therefore affirm the district court in sanctioning Rader in the amount of $3,000.00. VII. For the reasons stated above, we affirm the district court’s certification of a class. Furthermore, we affirm the district court on the issue of damages. We reverse, vacate, and remand, however, on the issue of attorneys’ fees; the district court correctly determined the lodestar amount, but enhancement should not have been ordered except possibly on one Johnson factor — the results obtained — with respect to which we remand. In addition, we affirm the district court’s sanctioning of Trinity’s trial counsel. Accordingly, the decision of the district court is AFFIRMED in part; REVERSED, VACATED, and REMANDED in part. Footnotes [1] During the course of discovery, the district court imposed personal sanctions against Trinity’s trial counsel, Robert Rader, Jr. Rader’s appeal of these sanctions has been consolidated with this appeal. [2] Trinity also argues on appeal that the district court erred by not including voluntary resignations in its analysis of discrimination in layoffs. Trinity itself originally excluded from its layoff analysis those employees who had resigned. Its analysis, however, showed that a statistically significant higher number of black employees than whites had been laid off. Thus, Trinity then revised its study and included those employees who had resigned. Trinity, however, presented no evidence of actual employee motivation in resigning. Furthermore, Shipes’s analysis still showed an unfavorable result to black employees even when employees who had resigned were included in the statistical data. The district court found Trinity’s proposal to *316 include employees who had resigned in its analysis of discrimination in layoffs unsupported and illogical; we find no reversible error in the district court’s decision. [3] Rule 23(a) provides that one or more members of a class may sue on behalf of all if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class. [4] Under Trinity’s proposed method, every class member would receive the same monetary award. In the calculation of the back pay award for initial rate, for example, the average starting rate for white employees as a group would be compared to that of plaintiff class members. Simply illustrated, if five white employees had beginning rates as a group of $5.00, $6.00, $7.00, $8.00, and $9.00, the average would be $7.00. If five class members had beginning rates of $3.00, $4.00, $5.00, $6.00, and $7.00, the average would be $5.00. The total difference in rates between the groups would be $2.00, (the difference in averages) multiplied by 5, the number of class members, or $10.00. This amount, $10.00, would then be distributed pro rata to the class members, each class member receiving a back pay award of $2.00, or $10.00 divided by 5. Trinity’s method would thus shift the entire group of plaintiff class members upward by that amount. [5] For example, if the average length of employment for white employees hired during a particular time period was two years, but a black employee hired during this same time period was laid off or terminated after one year, the black employee would be entitled to back pay for one year. [6] The twelve Johnson factors are (1) the time and labor involved; (2) the novelty and difficulty of the questions; (3) the skill requisite to perform the legal services properly; (4) the preclusion of other employment by the attorney due to this case; (5) the customary fee; (6) whether fee is fixed or contingent; (7) time limitations; (8) the amount involved and results obtained; (9) the experience, reputation, and ability of counsel; (10) the undesirability of the case; (11) the nature and length of the professional relationship with the client; and (12) awards in similar cases. Johnson v. Georgia Highway Express, 488 F.2d 714, 717-19 (5th Cir.1974). [7] The district court further determined that enhancement should not be granted for two remaining Johnson factors: the undesirability of the case and the nature and length of professional relationship with the client. [8] We also note that the complaint in this case was filed on December 16, 1980, the trial began on November 19, 1984, and partial final judgment on liability was entered on October 10, 1985; for this extended period of time, however, Shipes's attorneys sought fees for only 1,306.88 hours. Certainly during this time period Shipes’s attorneys were not precluded from accepting other employment because of this case. [9] As we stated earlier, this factor is presumably reflected in the lodestar amount and enhancement based on results obtained is proper only in rare and exceptional cases supported by specific evidence and detailed findings by the district court. See Alberti, 896 F.2d at 930. On remand the district court should take cognizance of this requirement in deciding whether to enhance the lodestar amount based upon the results obtained.