Summary
In Vosdingh v. Qwest Dex, Inc., No. Civ. 03-4284, 2005 WL 914732 (D. Minn. Apr. 21, 2005), a district court held that the plaintiff employees had shown that an evaluation system had a disparate impact on women who took leave, and accordingly applied the three-step framework of 42 U.S.C. § 2000e-2(k), shifting the burden to the employer to demonstrate a legitimate, business-related justification for the evaluation system, which the plaintiff could defeat by showing that an alternative approach — such as one normalizing the evaluation scores to compensate for absence — was comparably effective.
Summary of this case from Velez v. Novartis Pharmaceuticals Corp.Opinion
Civil No. 03-4284 ADM/AJB.
April 21, 2005
Bruce C. Recher, Esq., and Richelle Wahi Reiff, Esq., Henson Efron, P.A., Minneapolis, MN, appeared for and on behalf of Plaintiffs.
Elizabeth I. Kiovsky, Esq., Baird Kiovsky LLC, Denver, CO, and Melissa Raphan, Esq., Dorsey Whitney LLP, Minneapolis, MN, appeared for and on behalf of Defendants.
MEMORANDUM OPINION AND ORDER
I. INTRODUCTION
On February 3, 2005, oral argument was heard before the undersigned United States District Judge on Melanie Vosdingh ("Vosdingh") and Holly Nicholls' ("Nicholls") (collectively, "Plaintiffs") Motion for Partial Summary Judgment [Docket No. 44] on Counts One and Two of Plaintiffs' Amended Complaint [Docket No. 7]. Oral argument was also heard on Qwest Dex, Inc., Qwest Corporation and Dex Media East LLC's (collectively, "Qwest Dex" or "Defendants") Motion for Summary Judgment [Docket No. 40]. Plaintiffs' Amended Complaint alleges Defendants' performance evaluation plan discriminated on the basis of sex and pregnancy, had a disparate impact on female employees, and violated both the Family Medical Leave Act ("FMLA") and Minnesota Parenting Leave Act ("MPLA"). For the reasons set forth below, Plaintiffs' Motion for Partial Summary Judgment is denied and Defendants' Motion for Summary Judgment is granted in part and denied in part.
Andrea Fannan ("Fannan") settled her claims against Defendant and is no longer an active Plaintiff in this case. As a result, Count VII of Plaintiffs' Amended Complaint, which alleges violations of Fannan's rights under the Americans with Disabilities Act, 42 U.S.C. § 12101, et seq., is no longer before this Court. After the issuance of this Order, the case caption will reflect only the remaining Plaintiffs.
II. BACKGROUND
For purposes of the instant Motion, the facts are viewed in the light most favorable to the nonmovant. See Ludwig v. Anderson, 54 F.3d 465, 470 (8th Cir. 1995).
Plaintiffs Vosdingh and Nicholls worked as sales consultants in the Minneapolis-St. Paul offices of Qwest Dex. Vosdingh served as a "telephone" sales consultant until she was discharged in February 2002. Nicholls served as a "premise" sales consultant until her alleged constructive discharge in March 2003. Dex Media, formerly Qwest Dex, is an information directory business that sells traditional Yellow Page and Internet advertising. During the period in question, 1999-2002, the company maintained offices in St. Paul, Minneapolis, Duluth and Rochester, Minnesota.
Telephone sales consultants are internal salespeople who generally call on accounts with low dollar volume, while premise sales consultants are external salespeople who visit customers with higher end accounts. Premise sales consultants generally have greater perquisites, flexibility, opportunity for commission and higher incomes than telephone sales consultants. Cocchiarella Dep. (Recher Aff. [Docket No. 50] Ex. C) at 80.
A. Sales Performance Evaluation System
At the epicenter of this lawsuit is the performance evaluation system used by Defendants to evaluate their sales consultants during the time Plaintiffs were employed. Known as the sales performance evaluation system ("SPE"), the system assigned point totals to track performance in seven different categories. SPE System (Recher Aff. [Docket No. 57] Ex. 75). The results of these categories were weighted and cumulatively constitute a sales consultant's overall SPE score. In order to accommodate short-term market inconsistencies, the system used sales performance data for individual sales consultants generated over a "rolling 24-month" (or eight quarter) period to calculate the individual components that comprise the SPE score. SPE System at 126; Scott Dep. (Recher Aff. [Docket No. 54] Ex. G) at 34; Roberts Dep. (Recher Aff. [Docket No. 50] Ex. B) at 17-18.
Advertisers net percentage gain, known as combined monthly billing ("CMB"), constitutes 40% of the total SPE score. SPE System at 4-6. Internet sales, next issue monthly billings, new customer conversion and new customer revenue make up 45% of the SPE score. Id. at 7-14. Work flow and accuracy comprise 15% of the SPE score. Id. at 15-18.
Employees received performance ratings based solely on how their SPE score compared to the "universe" median. Roberts Dep. at 24-25; Scott Dep. at 37. A "universe" was comprised of sales consultants who held the same position in a specific sales office. A sales person received a satisfactory rating if her score was within 25% of the universe median. Roberts Dep. at 24-25; Scott Dep. at 37. A score 25% or more below the universe median equated to an unsatisfactory score. Roberts Dep. at 24-25; Scott Dep. at 37. To receive a performance rating, employees must have worked a certain number of sales days (190 days). Roberts Dep. at 24-25. Employees who worked less than the requisite number of days would not receive a performance rating, although the system would continue to generate an SPE score. SPE Rating Timeframe Modification Memo (Recher Aff. [Docket No. 57] Ex. 77). The system continued to input data, or the absence thereof, for sales consultants who took any kind of company-approved leaves of absence, including vacation days, sick days, FMLA leaves, maternity/parenting or child care leave and disability leave. Roberts Dep. at 28-30, 68-69. This absence of data was used to calculate a sales consultant's SPE score for a 24-month period before it eventually dropped out of the system. Id. at 94-95.
A sales consultant who received four unsatisfactory ratings out of the preceding eight quarters, including the last two quarters, was subject to discipline, including termination. SPE System at 28-29; Scott Dep. at 46-48. A case management meeting was required to "insure consistency across all universes" for those employees who have received a third or fourth unsatisfactory rating. SPE System at 28. Management also considered "extenuating circumstances," including "[t]ime away from the job" before taking disciplinary action. Id.
The system was altered in July of 1999 to include an internet sales component and to change point allocations. Redesigned SPE Memorandum (Recher Aff. [Docket No. 57] Ex. 76). SPE scores were generated under both the old and new version for one year to avoid short-term market inconsistencies. Id. In 2003, Defendants adopted a "normalization" process, which numerically adjusted four of the SPE categories (internet sales, new customer conversions, new customer revenue, next month billing) to account for the entire time a sales consultant was on a leave of absence. Roberts Dep. at 129-30, 139-177.
B. Care of Newborn Children Leave of Absence
Depending on their assigned office, sales consultants were represented by either the Communications Workers of America or the International Brotherhood of Electrical Workers. The terms of the sales consultants' employment were governed by a collective bargaining agreement that granted Defendants the right to establish and administer a performance evaluation system. The agreement permitted sales consultants to take a parenting or child care leave for up to one year. Bradford Dep. at 13-21 (Recher Aff. [Docket No. 54] Ex. E); International Brotherhood of Electrical Workers Agreement (Recher Aff. [Docket No. 56] Ex. 73) at 23; Communications Workers of America Agreement (Recher Aff. [Docket No. 61] Ex. 145) at 39.
The terms of the leave were set forth in a policy, entitled "Care of Newborn/Adopted/Foster Children Leave of Absence" ["CNC leave"]. Section 3.01-3.20 (Recher Aff. [Docket No. 57] Ex. 74). Under the policy, any natural, adoptive, or foster parent may take a "care of newborn child or parenting leave." Id. The length of the leave of absence was governed by Section 3.05: "[a]n employee may take up to twelve (12) months from date of birth . . . inclusive of all extensions. This leave will not be granted, nor may this leave be taken beyond twelve (12) months from the date of birth, adoption, or placement for foster care."Id. § 3.05. Although Defendants had no discretion to deny the first six months of the leave, the last six months were subject to Defendants' agreement, as dictated by "needs of the business."Id. § 3.06; Bradford Dep. at 26-27. Defendants testified they were aware of no instance in which a request for the subsequent six months of leave was denied. Id. at 22-24; Scott Dep. at 20-21.
C. Melanie Vosdingh
Vosdingh began working for Qwest Dex's predecessor in 1996 and became a telephone sales consultant in 1997. Vosdingh Dep. (Recher Aff. [Docket No. 55] Ex. H) at 53-54. Vosdingh performed well in the position and won, among other awards, Qwest Dex's highest sales award for a two-year span ending on December 31, 1999 for having the highest performance evaluation in the St. Paul telephone universe. Scott Dep. at 166; 12/30/00 Annual Review (Recher Aff. [Docket No. 56] Ex. 25) at 9; 12/31/99 Annual Review (Recher Aff. [Docket No. 55] Ex. 11). Vosdingh became pregnant in the latter half of 1999 and missed time from work in March and April 2000 because of medical complications associated with her pregnancy. Vosdingh Dep. at 91-92, 105. In mid-April 2000, Vosdingh went on a company-approved leave of absence. Id. This leave included both FMLA leave and CNC leave. Id. at 92-93; FMLA Status (Recher Aff. [Docket No. 55] Ex. 16); July 17, 2000 Letter (Recher Aff. [Docket No. 55] Ex. 17). Vosdingh was scheduled to return to work on May 9, 2001. 10/17/00 E-mail (Recher Aff. [Docket No. 61] Ex. 188). When Vosdingh began her leave of absence, at the end of the first quarter of 2000, Vosdingh had an overall SPE score of 55.13, which was satisfactory when compared to the median universe score of 57.22. 3/25/00 Review (Recher Aff. [Docket No. 55] Ex. 12).
Under the older version of the SPE system, which was still being calculated at this time, Vosdingh had a score of 84.35, as compared to a universe median of 78.38. 12/30/00 Annual Review.
Although the parties dispute the reason why, Vosdingh returned to work for one month in September 2000 before her leave of absence was reinstated effective October 9, 2000. Vosdingh Dep. at 99-100; Hanf Dep. (Kiovsky Aff. [Docket No. 42] Ex. I) at 22, 26. For various work and health related reasons, Vosdingh was in the office for only one week of this month and generated no sales during this period. Vosdingh ultimately returned to work on January 15, 2001. On February 15, 2001, Vosdingh received her annual performance review for the period ending December 30, 2000. 12/30/00 Annual Review. Vosdingh received an SPE score of 36.12, which was unsatisfactory when compared to the universe median of 57.76. Id. In her year end performance review, her manager noted Vosdingh's CMB was low because she was on maternity leave.Id. He also noted Vosdingh had "achieved the Directors award in the past and I know it is possible for her to achieve it again."Id. In April 2001, Vosdingh was told she would not receive a performance rating for this period because, due to her leave of absence, she had not worked the number of days necessary to generate such a rating. Vosdingh Dep. at 182-83.
Although Vosdingh's SPE score improved in the first and second Quarters of 2001, it remained unsatisfactory when compared to the universe median. Her 2001 first Quarter SPE score was 46.26; the universe median was 72.06. 4/07/01 Review (Recher Aff. [Docket No. 56] Ex. 26). Similarly, her 2001 second Quarter SPE score was 51.73, compared to a universe median of 68.97. 6/30/01 Review (Recher Aff. [Docket No. 56] Ex. 29). During this time, Vosdingh sought mental health treatment and was diagnosed by a licensed psychologist and a psychiatrist with an adjustment disorder with anxiety and depressed mood. Selmo Dep. (Recher Aff. [Docket No. 55] Ex. J) at 21; Hedberg Dep. (Recher Aff. [Docket No. 55] Ex. K) at 54-55, 62-71 The two clinicians concluded that Vosdingh's condition was a result of work pressures. Selmo Dep. at 21; Hedberg Dep. at 54-55, 62-71 Vosdingh's SPE scores did not improve over the next two Quarters. In the third Quarter of 2001 she generated an unsatisfactory SPE score of 42.88, compared to the universe median of 71.09. 10/06/01 Review (Recher Aff. [Docket No. 56] Ex. 33). Her fourth Quarter 2001 SPE score was 33.65, also well below the universe median of 67.56. 12/31/01 Annual Review (Recher Aff. [Docket No. 56] Ex. 35).
On February 2, 2002, following her fourth unsatisfactory rating, Defendants convened a case management conference to discuss her performance. Scott Dep. (Kiovsky Aff. Ex. B) at 182; Nielsen Dep. (Kiovsky Aff. Ex. J) at 143. In a February 7, 2002 letter notifying Vosdingh of her termination, Defendants' explained, "[b]ased upon your four consecutive unsatisfactory ratings under the SPE, the decision has been made to terminate your employment with DEX effective today." 2/7/02 Letter (Recher Aff. [Docket No. 56] Ex. 37).
D. Holly Nicholls
Nicholls began working for Qwest Dex's predecessor in 1994 and became a St. Paul telephone sales consultant in 1996. Nicholls Dep. (Recher Aff. [Docket No. 55] Ex. I) 10, 49, 51. In late 1999, Nicholls was subsequently promoted to St. Paul premise sales consultant, a position she held until July 1999. Id. at 51-52. Nicholls twice won the Vice President's Circle of Excellence Award and by the end of the second quarter of 1999 had posted six consecutive quarters of satisfactory SPE scores. Id. at 40, 90, 93-94, 215-16. In July 1999, Nicholls transferred to the Minneapolis premise office. Id. at 95-103. After the transfer, her SPE scores for the third and fourth Quarter of 1999 were unsatisfactory. Scott Dep. at 140-41, 154-55. Nicholls' performance review for the third Quarter reflected an overall SPE score of 30.76 in relation to a universe median of 71.69. 9/25/99 Review (Kiovsky Aff. Ex. F-158). Similarly, her SPE score for the fourth Quarter was 33.32, as compared to the universe median of 72.11. 12/31/99 Annual Review (Recher Aff. [Docket No. 58] Ex. 99).
On December 14, 1999, Nicholls took a leave of absence after being placed on bed rest due to pregnancy-related complications. Nicholls Dep. at 126-27, 148. After her child was born on January 18, 2000, Nicholls began a one year CNC leave. Id. at 123-25. Nicholls returned to work in January 2001. Id. at 125.
Nicholls' SPE score fell from 33.32 to 22.9 during the time she was on leave, a 16% decline relative to the universe median. (Recher Aff. [Docket No. 59] Ex. 100). Although Nicholls' score trended upward and she showed improvement in specific areas such as internet sales, her SPE score remained in the unsatisfactory range in the first three quarters of 2001. 4/7/01 Review (Recher Aff. [Docket No. 59] Ex. 101); 6/30/01 Review (Recher Aff. [Docket No. 59] Ex. 103); 10/6/01 Review (Recher Aff. [Docket No. 59] Ex. 104). In the first quarter, Nicholls received a SPE score of 25.94 as compared to the universe median of 61.34. 4/7/01 Review. In the second quarter, she generated a SPE score of 27.87 in relation to the universe median of 62.72. 6/30/01 Review. In the third quarter, she garnered an SPE score of 36.13 while the universe median was 63.13. 10/6/01 Review.
Nicholls' overall SPE score further improved in the fourth quarter of 2001 and she received handwritten comments from her manager encouraging her to keep "up the good work" and noting she was "doing all the right things." 12/31/01 Annual Review (Recher Aff. [Docket No. 59] Ex. 106). Nevertheless, her SPE score of 46.48 was still unsatisfactory when compared to the universe median of 70.42. Id. During her year end review, Nicholls requested she receive additional market in hopes of increasing her net advertiser gain score. Id.; Nicholls Dep. at 148-49. Her request was denied based on an unwritten company policy that consultants receiving unsatisfactory SPE rating were prohibited from obtaining reassigned accounts. Id. at 140; 12/31/01 Review; Cocchiarella Dep. (Recher Aff. [Docket No. 50] Ex. C) at 137.
At Nicholls' February 2002 review meeting, management agreed Nicholls had made "good strides" and had a "good attitude." 2/5/02 Meeting Notes (Recher Aff. [Docket No. 61] Ex. 147). In March 2002, Nicholls told her manager, Mary Cocchiarella, that she was pregnant. Nicholls Dep. at 32-34. Cocchiarella responded by asking "what are you going to do about your job?" She also stated "It's hard to come back after having a child" and "It's hard to do this job with two kids, look what happened to Andrea." Id. In April 2002, Cocchiarella also stated that Nicholls hit a "bump in the road" when she took maternity leave. Id. at 154-55; Stanton Aff. [Docket No. 48] ¶ 13.
The Andrea reference is to former Plaintiff Andrea Fannan, a St. Paul premise sales consultant who left Defendants' employment after receiving several unsatisfactory ratings. Fannan had taken six months of maternity leave. Fannan Leave Application (Kiovsky Aff. Ex. F-66).
Nicholls received another unsatisfactory rating in the first quarter of 2002, after generating an overall SPE score of 49.26 compared to a universe median of 70.29. 4/6/02 Review (Recher Aff. [Docket No. 59] Ex. 109). On April 26, 2002, Defendants held a case management conference to discuss Nicholls' performance. Because Nicholls had received four consecutive unsatisfactory ratings, she was required to leave the premise position. Cocchiarella Dep. at 164, 180-81. However, Defendants agreed to offer Nicholls a telephone sales consultant post given her strong past performance in that position. Cocchiarella Dep. at 176. Nicholls was given 24 hours to decide between termination or accepting a demotion to a telephone sales consultant position.Id. at 180; Nicholls Dep. at 155; April 30, 2002 Meeting Notes (Recher Aff. [Docket No. 59] Ex. 112). Nicholls chose to accept the telephone sales consultant position.
On July 9, 2002, Nicholls took a company approved leave and on August 25, 2002, gave birth to her second child. Nicholls Dep. at 16-17. Nicholls was scheduled to return from CNC leave on May 1, 2003. Id.; FMLA Status Review (Kiovsky Aff. Ex. F-151). On March 17, 2003, however, Nicholls tendered her letter of resignation, effective June 1, 2003, stating she did not want to return to the telephone sales consultant position and face an SPE system that incorporated the lack of data generated by her six months of CNC leave. Nicholls Dep. at 39-41; Resignation Letter (Recher Aff. II [Docket No. 88] Ex. 154).
In its Amended Complaint, Plaintiffs assert a number of claims against Defendants, including: (1) violation of the FMLA, 29 U.S.C. § 2601, et seq.; (2) violation of the MPLA, Minn. Stat. § 181.940, et seq.; (3) sex discrimination under Title VII, 42 U.S.C. § 2000e, et seq.; (4) sex discrimination under the Minnesota Human Rights Act ("MHRA"), Minn. Stat. § 363.01, et seq.; (5) reprisal under Title VII; and (6) reprisal under the MHRA. Plaintiffs seek summary judgment on Counts I and II only. Defendants move for summary judgment on all counts.
III. DISCUSSION
A. Standard of Review
Federal Rule of Civil Procedure 56(c) provides that summary judgment shall issue "if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to judgment as a matter of law." Fed.R.Civ.P. 56(c); see Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 587 (1986); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 252 (1986); Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986). On a motion for summary judgment, the Court views the evidence in the light most favorable to the nonmoving party. Ludwig v. Anderson, 54 F.3d 465, 470 (8th Cir. 1995). The nonmoving party may not "rest on mere allegations or denials, but must demonstrate on the record the existence of specific facts which create a genuine issue for trial." Krenik v. County of Le Sueur, 47 F.3d 953, 957 (8th Cir. 1995).
B. FMLA
The FLMA provides eligible employees up to 12 work weeks of unpaid leave during any 12-month period. 29 U.S.C. § 2612. Employees may take FMLA leave, inter alia, to give birth, to care for a newborn, or because a serious health condition makes the employee unable to perform the employee's job functions. Id. (a)(1). An employee must provide notice to an employer that she plans to take FMLA leave. 29 U.S.C. § 2612(e)(2). The statute contains two provisions that protect an employee from discrimination based on her exercise of rights guaranteed by the FMLA. See 29 U.S.C. §§ 2615(a)(1) (2). The first refers to "interference" or (a)(1) claims, in which an employee asserts an employer denied or otherwise interfered with her substantive rights under the FMLA. § 2615(a)(1). The second concerns "retaliation" or (a)(2) claims, which protects an employee from discrimination for exercising those rights.
An employee is eligible for FMLA leave if she "has been employed — (i) for at least 12 months by the employer with respect to whom leave is requested under section 2612 of this title; and (ii) for at least 1,250 hours of service with such employer during the previous 12-month period." 29 U.S.C. § 2611(2)(A). The parties do not dispute that both Vosdingh and Nicholls are eligible employees.
Similarly, there is no dispute that Vosdingh and Nicholls provided Defendants with appropriate notice.
§ 2615(a)(1). Plaintiffs allege Defendants actions violated both provisions. Plaintiffs request summary judgment on their claims under § 2615(a)(1) while Defendants seek summary judgment on both the § 2615(a)(1) and § 2615(a)(2) claims.
1. Interference Claims
Section 2615(a)(1) makes it "unlawful for any employer to interfere with, restrain or deny the exercise of or the attempt to exercise, any right provided under this title." The regulation interpreting this provision provides, "[e]mployers cannot use the taking of FMLA leave as a negative factor in employment actions, such as hiring, promotions or disciplinary actions. . . ." 29 C.F.R. § 2615(a)(1) § 825.220(c). The Eighth Circuit has held § 2615(a)(1) claims should be analyzed under an objective test rather than theMcDonnell Douglas burden-shifting scheme. Rankin v. Seagate Tech., Inc., 246 F.3d 1145, 1148 (8th Cir. 2001). As a result, an employee must show, by a preponderance of the evidence, that she was entitled to the benefit denied. See Strickland v. Water Works and Sewer Bd., 239 F.3d 119, 1206-07. Once such an entitlement is shown, an employer can only avoid liability by showing it would have made the same contested decision even if the employee had not exercised her FMLA rights. Throneberry v. McGehee Desha County Hosp., 2005 U.S. App. Lexis 5865, *10 (8th Cir. April 11, 2005).
Plaintiffs argue Defendants violated § 2615(a)(1) by including in their SPE scores the absence of performance data generated while they were on FMLA leave. Plaintiffs contend these scores form the basis for the unsatisfactory performance ratings that resulted in Vosdingh's termination and Nicholls' demotion, and therefore Defendants used the taking of FMLA leave as a negative factor in employment actions. Defendants argue Plaintiffs' claims should not be considered as interference or (a)(1) claims because Plaintiffs each received 12 weeks of FMLA leave and were subsequently reinstated to their positions, as required by the FMLA.
Although certain circumstances may raise either (a)(1) or (a)(2) claims, the Eighth Circuit has confined interference claims to those sets of facts where employees have been denied their substantive rights to FMLA leave or to reinstatement following that leave. See Rankin, 246 F.3d at 1148; Smith v. Allen, 302 F.3d 827 (8th Cir. 2002); Throneberry, 2005 U.S. App. Lexis at *11; Dillway v. Ferrante, 2003 U.S. Dist. LEXIS 23468, *3 (D. Minn. Dec. 9, 2003). In the instant case, it is undisputed that both Plaintiffs received all of the FMLA leave to which they were entitled and were subsequently reinstated to their positions upon their return. Vosdingh took her FMLA leave from April 10, 2000 until July 9, 2000. Vosdingh Dep. at 92-93. When she eventually returned to work in January 2001, she was reinstated to her telephone sales consultant position. Nicholls took 12 weeks of FMLA leave from December 14, 1999 until March 7, 2000 and was reinstated to her premise sales consultant position on her return in January 2001. Nicholls Dep. at 127. Nicholls began a second 12 weeks of maternity related FMLA leave on July 8, 2002. Id. at 10-11. On March 17, 2003, before returning to work, she submitted a letter of resignation. Id. at 15. There is no allegation Nicholls did not receive her 12 weeks FMLA leave before resigning. For reasons discussed infra, the Court rejects Nicholls' contention that she was constructively discharged and thereby denied reinstatement. Therefore, Plaintiffs cannot claim they were denied an entitlement under the FMLA. See Sarno v. Douglas Elliman-Gibbons Ives, Inc., 183 F.3d 155, 161 (2d Cir. 1999).
Vosdingh was not terminated until February 7, 2002, 17 months after completing her FMLA leave. Similarly, Nicholls was not demoted to a telephone sales consultant position until February 2002, nearly two years after her first FMLA leave ended. Adopting Plaintiffs' interpretation of (a)(1) would swallow (a)(2) as every act of FMLA discrimination would also constitute interference. Plaintiffs would be able to prevail merely by satisfying the objective test of (a)(1) rather than the more demanding McDonnell Douglas test of (a)(2). In Darby v. Bratch, the Eighth Circuit established that (a)(2)'s prohibition against discriminating for asserting rights under the Act "necessarily includes consideration of an employee's use of FMLA leave as a negative factor in an employment action." 287 F.3d 673, 679 (8th Cir. 2002). Therefore, because Plaintiffs received the full amount of FMLA leave to which they were entitled and were subsequently reinstated to their same positions, their (a)(1) interference claims fail.
Furthermore, the Eighth Circuit recently held that "an employer who interferes with an employee's rights will not be liable if the employer can prove it would have made the same decision had the employee not exercised the employee's FMLA rights."Throneberry, 2005 U.S. App. Lexis at *11. The employer bears the burden of establishing this defense. Id. at *18-19. Even if Defendants interfered with an FMLA right, Defendants have shown Plaintiffs would have been terminated had they not exercised their FMLA rights. Defendants have consistently claimed Plaintiffs were terminated for sustained under-performance for multiple quarters and not because they took FMLA leave. Vosdingh's SPE score remained in the satisfactory range after 12 weeks of FMLA leave. Nicholls' SPE score was in the unsatisfactory range before taking FMLA leave, but increased in relation to the median SPE score after the 12 week period. Defendants presented evidence that both Vosdingh and Nicholls remained in the unsatisfactory range for at least four quarters after returning from FMLA even if the 12 weeks of FMLA leave are excluded from subsequent calculations. Plaintiffs both took between 6 and 10 months of CNC leave upon the completion of their FMLA leaves. Their SPE scores were also lowered by the absence of performance data generated during this period. They then worked for another year before they suffered their respective adverse employment actions. During this time, they continued to generate unsatisfactory SPE scores. As a result, Defendants have met their burden of demonstrating they would have terminated Vosdingh and demoted Nicholls even if they had not taken FMLA leave.
2. Retaliation Claims
Section 2615(a)(2) makes it "unlawful for any employer to discharge or in any manner discriminate against any individual for opposing any practice made unlawful by this subchapter." The Eighth Circuit has held "[t]his prohibition necessarily includes consideration of an employee's use of FMLA leave as a negative factor in an employment action." Darby, 287 F.3d at 679. "An employee can prove FMLA retaliation circumstantially, using a variant of the burden shifting test established in McDonnell Douglas." McBurney v. Steve Hansen's Dodge City, Inc., 398 F.3d 998, 2005 U.S. App. LEXIS 2638, *10 (8th Cir. Feb. 16, 2005). Under the McDonnell Douglas framework, the plaintiff employee must initially establish a prima facie case of discrimination. This showing creates a presumption that the employer acted unlawfully. See Kiel v. Select Artificials, Inc., 169 F.3d 1131, 1134-35 (8th Cir. 1999) (applying framework to Title VII claim). The burden of production then shifts to the employer who must provide legitimate, non-discriminatory reasons for the adverse employment action. Id. If the defendant meets this requirement, the burden returns to the plaintiff to show that the employer's explanation is pretextual. Id. To establish a prima facie case of FMLA retaliation, an employee must show: (1) she exercised rights protected under the Act; (2) she suffered an adverse employment action; and (3) a causal connection existed between the employee's action and the adverse employment action. See Darby, 287 F.3d at 679.
There is no dispute that Plaintiffs both exercised rights protected under the Act by taking 12 weeks of FMLA leave. To satisfy the second prong of a prima facie case, Plaintiffs assert several acts which they argue are adverse employment actions. Adverse employment actions are defined as "tangible changes in duties or working conditions that constituted a material employment disadvantage." Manning v. Metropolitan Life Ins. Co., Inc., 127 F.3d 686, 692 (8th Cir. 1997).
First, Plaintiffs Vosdingh and Nicholls suffered adverse employment actions through their respective termination and demotion. Second, Plaintiffs correctly note, "[a] negative review is actionable . . . where the employer subsequently uses the evaluation as a basis to detrimentally alter the terms or conditions of the recipient's employment." LaCroix v. Sears, 240 F.3d 688, 692 (8th Cir. 2000). Plaintiffs were terminated and demoted as a result of receiving four unsatisfactory performance evaluations. Therefore, their negative reviews may also qualify as adverse employment actions.
Third, Vosdingh claims Defendants discussed promoting her to the premise sales consultant position, but these discussions ceased upon her announcement that she was pregnant. Vosdingh Dep. at 103. Although the denial of promotional opportunities can constitute an adverse employment action, Vosdingh did not demonstrate that premise sales consultant positions were available during the time in question or submit an application for such a position as required by the collective bargaining agreement. Id. at 79, 103-04. Consequently, as a matter of law, she cannot show she was denied promotional opportunities. See Dotson v. Delta Consol. Indust., Inc., 251 F.3d 780, 781 (8th Cir. 2001).
Finally, Nicholls contends she was constructively discharged. "An employee is constructively discharged when an employer renders the employee's working conditions intolerable and thus forces her to quit her job." West v. Marion Merrell Dow, Inc., 54 F.3d 493, 497 (8th Cir. 1995) (citations omitted). As a result, an employee must prove both "intent" and "intolerable conditions" to show she was constructively discharged. Whether an employee is constructively discharged is evaluated by an objective standard. Bergstrom-Ek v. The Little Stores, 153 F.3d 851, 858 (8th Cir. 1998). "An employee may not be unreasonably sensitive to her working environment. A constructive discharge arises only when a reasonable person would find her working conditions intolerable." West, 54 F.3d at 497 (citations omitted). An employee is not constructively discharged if she quits without giving her employer a reasonable chance to remedy the problem. Id. at 498.
Nicholls' constructive discharge claim fails because she did not give Defendants a reasonable chance to remedy the problem. Although Nicholls did complain to management about the perceived unfairness of having the lack of data generated while she was on company approved leave tabulated in her SPE score, she filed neither a grievance with the union nor an internal EEO complaint. Beginning August 25, 2002, Nicholls was on a company approved leave of absence and was not scheduled to return to work until May 1, 2003. Nicholls tendered her resignation in a letter dated March 17, 2003. At deposition, she confirmed that nothing occurred between July 2002 and March 2003 to cause her to resign. Nicholls Dep. at 41. Nicholls initiated no contact with Defendants about proposed remedial measures prior to tendering her resignation. The circumstances here are not the type a reasonable person would find to be objectively intolerable.
However, Vosdingh's termination, Nicholls' demotion and both Plaintiffs' negative reviews constitute adverse employment actions sufficient to satisfy the second prong of a prima facie FMLA retaliation claim.
To satisfy the final prong, Plaintiffs must show a causal connection existed between the protected activity and the adverse employment action. In general, mere temporal proximity is insufficient to establish causation. Smith v. Mem'l Hosp. Corp., 302 F.3d 827, 832-33 (8th Cir. 2002). Here, the extended time period between events (17 and 23 months elapsed between the time Vosdingh and Nicholls respective FMLA leaves ended and the time their adverse employment actions occurred) is too attenuated to establish a causal connection through temporal proximity. See Gagnon v. Sprint Corp., 284 F.3d 839, 851 (8th Cir. 2002).
Plaintiffs alternatively argue the SPE's inclusion of data from the period when they were on FMLA leave lowered their overall SPE scores for the subsequent quarters and thereby constituted a negative factor in the resulting adverse employment actions. Although Plaintiffs' argument sounds logically correct, they are hamstrung by the Eighth Circuit's holding that an employee "must demonstrate FMLA leave was the determinative factor in the employment decision at issue." Hatchett v. Philander Smith College, 251 F.3d 670, 677 (8th Cir. 2001) (emphasis added). This standard recognizes that the "FMLA does not require an employer to allow an employee to stay in a position that the employee cannot perform." Id. Plaintiffs have not presented any evidence to show their 12 weeks of FMLA leave was the reason they suffered adverse employment actions. Although their subsequent SPE scores were undoubtedly affected by absence of data from the 12 week period of FMLA, there is no evidence from which a reasonable jury could conclude FMLA leave was the determinative factor in Vosdingh's termination and Nicholls' demotion. For the same reasons, Plaintiffs cannot show their 12 weeks of FMLA leave was the determinative factor in generating any of the negative performance reviews that detrimentally altered the terms and conditions of their employment. As a result, Plaintiff's (a)(2) retaliation claim must be dismissed.
C. Minnesota Parenting Leave Act
The MPLA allows eligible employees to take an unpaid leave of absence in conjunction with the birth or adoption of a child. Specifically, the MPLA provides:
An employer must grant an unpaid leave of absence to an employee who is a natural or adopted parent in conjunction with the birth or adoption of a child. The length of the leave shall be determined by the employee, but may not exceed six weeks unless agreed to by the employer.
Minn. Stat. § 181.941, subd. 1. The statute also explains "[n]othing in Sections 181.940 to 181.943 prevents any employer from providing leave benefits in addition to those provided in Sections 181.940 to 181.944." Minn. Stat. § 181.943(b). Finally, the MPLA contains a provision prohibiting employers from retaliating against employees who exercise the rights provided under the Act. Minn. Stat. § 181.941, subd. 3. There is no dispute that Vosdingh and Nicholls were eligible for MPLA leave.
Minn. Stat. § 181.941, subd. 3 provides:
No Employee Retribution. An employer shall not retaliate against an employee for requesting or obtaining a leave of absence as provided by this section.
Plaintiffs argue Defendants violated the MPLA's retaliation provision by factoring into their rolling SPE scores the absence of performance data generated when Plaintiffs were on company approved CNC leaves. Defendants claim the MPLA entitles the employee to only six weeks of parenting leave. Any parenting leave provided by the employer beyond that six weeks is voluntary and not protected by the MPLA's retaliation provision. Furthermore, Defendants contend any claim the CNC policy could give rise to a cause of action under the MPLA is preempted by § 301 of the Labor Management Relations Act ("LMRA"). Under § 301, "[a]n application of state law is preempted . . only if such application requires the interpretation of a collective bargaining agreement." Lingle v. Norge Div., Magic Chef, 486 U.S. 399, 413 (1988).
As an initial matter, this Court finds Defendants waived any argument that Plaintiffs' claims are preempted under § 301 of the LMRA by failing to raise the affirmative defense in a timely manner. As § 301 of the LMRA dictates only the law to be applied rather than the forum in which a claim is litigated, it is well established that § 301 preemption may be waived and that "parties do not have an absolute right to raise that argument at any stage they wish in the proceedings." Sweeney v. Westavco Co., 926 F.2d 29, 39-40 (1st Cir. 1991) (Breyer, J.); see also, Johnson v. Armored Transport of California, Inc., 813 F.2d 1041, 1044 (9th Cir. 1987). Defendants did not raise § 301 preemption prior to submitting their Memorandum in Opposition to Plaintiffs' Motion for Partial Summary Judgment [Docket No. 79]. At oral argument, Defendants' argued they did not raise § 301 preemption as an affirmative defense because they were not aware until Plaintiffs filed their Memorandum in Support of Plaintiffs' Motion for Partial Summary Judgment [Docket No. 47] that Plaintiffs sought to include CNC leave in their MPLA claim. A cursory reading of the Amended Complaint refutes this argument. In Count II (the MPLA claim), Plaintiffs allege:
135. Defendants Qwest and Qwest Dex agreed under Minn. Stat. 181.941, subd. 1, to provide plaintiffs a period of leave up to one year in conjunction with the birth of their children.
137. Defendants Qwest and Qwest Dex's actions in using the absence of performance data while plaintiffs were on a parenting leave agreed to by defendants . . . which operates as a negative factor in their performance evaluations, constitutes unlawful retaliation or retribution against plaintiffs because they took parental leaves in violation of Minn. Stat. § 181.941 subd. 3.
Amended Compl. ¶¶ 135, 137. These allegations provided Defendants with notice of Plaintiffs' intent to include their CNC leaves in their MPLA claims. Consequently, a failure to timely plead LMRA preemption as an affirmative defense constitutes a waiver of the application to that defense. See Dueringer v. General American Life Ins. Co., 842 F.2d 127, 130 (5th Cir. 1988); Williams v. Ashland Eng'g Co., Inc., 45 F.3d 588, 593 n. 7 (1st Cir. 1995); Florida Health Sciences Center, Inc. v. Tampa General Hosp., 190 F. Supp. 2d 1297, 1303 (M.D. Fl. 2001) (holding ERISA preemption was an affirmative defense that was waived if not timely plead).
The U.S. Supreme Court has stated Congress modeled ERISA preemption on preemption under § 301 of the LMRA. Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 52-55 (1987). As a result, the analysis of the scope of preemption under one statute generally applies with equal force to the other. See Wolf v. Reliance Standard Life Ins. Co., 71 F.3d 444, 448 (1st Cir. 1995).
In the alternative, Defendants have moved the Court for permission to amend their Answer to plead preemption under § 301 of the LMRA. Whether to grant leave to amend the pleadings is a matter committed to the discretion of the trial court. Although leave to amend "shall be generally freely given when justice so requires," there is no absolute right to amend. Fed.R.Civ.P. 15(a). Bediako v. Stein Mart, Inc., 354 F.3d 835, 840-41 (8th Cir. 2004). Leave to amend may be denied when delay causes undue prejudice to the non-moving party. Id. An affirmative defense must be pled in the answer to provide Plaintiffs with sufficient notice and opportunity to refute the defense. Florida Health Sciences Center, Inc., 190 F. Supp. 2d at 1304. In the instant case, Defendants did not raise the affirmative defense of preemption under the LMRA until this defense of a partial summary judgment motion, two months after the close of discovery. The Amended Complaint was filed more than a year ago. Further, Defendants did not raise the preemption defense in the context of a Rule 12(b) Motion to Dismiss. Given Defendants' failure to raise the affirmative defense of § 301 preemption until this late stage in the proceedings, the Court finds Plaintiffs would be unduly prejudiced. As a result, Defendants request is denied and the preemption defense will not be considered.
Next it must be determined whether the MPLA's retaliation provision applies to voluntary, company approved parenting leaves beyond the statutory six week period. Defendants offer legislative history to support their contention that Plaintiffs' interpretation of the Act "would impose extras duties and potential penalties on employers who desired to grant parenting leaves beyond the six weeks protected under the Act and would inhibit rather than promote employer flexibility to maintain marketplace-driven policies" in contravention of the legislature's intent. Defs.' Resp. to Pls.' Mot. for Partial Summ. J. at 21.
Careful examination of the legislative history, however, reveals the legislature's concern was that mandating employers provide a maximum of more than six weeks parenting leave would radically alter then-current business practices and significantly burden employers. The plain text of the Act itself clearly contemplates that employers may voluntarily approve additional parenting leave for employees. Having done so, however, an employer may not penalize an employee for taking approved parenting time. The purpose of the MPLA was to provide employees with the opportunity to care for newborn or newly adopted children. Allowing employers to discriminate against employees for taking companyapproved leave time is contrary to the purpose of the MPLA. The Act bars retaliation against an employee for requesting or obtaining a leave of absence, "as provided by this section." Minn. Stat. § 181.941, subd. 3. The MPLA clearly defines the duration of the parenting leave: "[t]he length of the leave shall be determined by the employee, but may not exceed six weeks, unless agreed to by the employer." Minn. Stat. § 181.941, subd. 1. Defendants were under no statutory obligation to provide CNC leaves to their employees. Having agreed to extended parenting leaves, however, the MPLA prohibits Defendants from retaliating against employees who exercise that leave.
Although the MPLA does not define retaliation, the prohibition against retaliation also applies to the Minnesota Human Rights Act, Minn. Stat. § 363A.15 and the Minnesota Whistleblower Act, Minn. Stat. § 181.932, subd. 1. To establish a prima facie case of retaliation under Minnesota law, plaintiffs must show: (1) they engaged in protected conduct; (2) they experienced an adverse employment action; and (3) a causal connection exists between the two. Bersch v. Rgnonti Associates, Inc., 584 N.W.2d 783 (Minn.Ct.App. 1998) (whistleblowing); Bradley v. Hubbard Broadcasting, Inc., 471 N.W.2d 670 (Minn.Ct.App. 1991) (MHRA retaliation). If Plaintiffs establish a prima facie case of retaliation, they must still satisfy the McDonnell Douglas burden shifting framework by showing any proffered business justification for the adverse employment action is merely pretextual. Bradley, 471 N.W.2d at 676. Consideration of the protected activity by the employer need not be the sole reason for the employment action, but a "discernable, discriminating and causative factor." Id. at 676.
Plaintiffs have satisfied a prima facie case of retaliation under the MPLA. It is undisputed that Plaintiffs engaged in protected conduct by taking company approved CNC leaves. Vosdingh took approximately 8 months of CNC leave while Nicholls took approximately 12 months. Furthermore, it is clear Plaintiffs suffered adverse employment actions through their respective termination and demotion. Finally, it is undisputed the SPE incorporates the absence of data recorded while an employee is on a leave of absence into an employee's overall SPE score. This lack of data is reflected in the employee's score for a 24 month period and this score directly translates to an employee's performance rating. The lack of data generated by Plaintiffs' extended CNC leaves exerted a negative pressure on their overall SPE scores and resulting performance ratings. Therefore, the taking of their parenting leave may constitute a "discernable, discriminating and causative factor" in Plaintiffs' respective adverse employment actions.
Having established a prima facie case, the burden shifts to Defendants to demonstrate a legitimate business reason for the adverse employment action. Defendants contend Plaintiffs suffered the adverse employment actions as a result of their sustained poor performance. Defendants' Minneapolis office maintained a longstanding policy to terminate any telephone or premise sales consultant whose overall sales performance was unsatisfactory for four quarters. Furthermore, since an SPE score within 25% of the universe median was still considered satisfactory, the SPE system was designed with the potential for every sales consultant to succeed. Defendants argue Vosdingh was terminated following a case management meeting based on her poor performance, sub-par work flow and lack of follow-through on coaching plans. See 2/4/02 Case Management Notes (Recher Aff. [Docket No. 58] Ex. 93); April 2, 2002 Meeting Notes. Defendants cite Nicholls' unsatisfactory SPE scores for several quarters prior to her demotion. Despite her performance, Defendants offered her the telephone sales consultant position in which she had previously been successful, rather than terminate her. Defendants have presented sufficient evidence to shift the burden of proof back to Plaintiffs.
Plaintiffs must now show Defendants' proffered legitimate business justification is pretextual. First, Plaintiffs proffered expert evidence indicating that 17 of 22 employees who took CNC leaves of 12 weeks or more and had SPE scores on record for the quarter before leave began and the quarter before they returned had their SPE scores decrease. Martin Aff. [Docket No. 68] Ex. D.
Second, Plaintiffs present evidence suggesting Defendants were aware extended CNC leaves may negatively affect employees' SPE scores and their resulting performance ratings. Vosdingh claims there were multiple time when she spoke to her manager, Mike Nielsen about being penalized under the SPE system for taking maternity leave. Vosdingh Aff. [Docket No. 84] ¶ 11. In her notes from a February 4, 2002 case management conference to discuss Vosdingh's performance, Kathleen Burke-Scheffler wrote Vosdingh "continually questions how a person out on CNC can be compared to employees who are on job the full 8 quarters." 2/4/02 Case Management Notes. At deposition, Mark Roberts, the SPE administrator for Qwest Dex, was asked whether this comment "triggered any further investigation." Roberts Dep. at 120. He responded "No, but it probably started to raise a flag" and "at some point, it became apparent that we needed to look at that. I don't know when that was." Id. at 120-21.
Nicholls also complained about the impact of CNC leave on her SPE score. At her first quarter 2001 review, Nicholls told her manager, Sandy Juettner, that she believed it was unfair for the SPE system to incorporate her CNC leave time into her overall score and resulting performance rating. Nicholls Dep. II (Recher Aff. [Docket No. 88] Ex. M) at 133. Juettner agreed it was not fair and suggested Nicholls look for a new job. Id. In her 2001 annual performance review, signed by her manager, Mary Cocchiarella, Nicholls wrote:
I want to emphasize that I am trending upwards to a satisfactory rating and am driving to be outstanding which I know I am! I still have one year of no data or numbers from my maternity leave in the year 2000. This does affect my total score, because some components of the SPE are averages and I have zeros for one year. Until the year falls off and I am judged only for the time I was here will this be a true reflection of my results.
12/31/01 Annual Review at 6. Cocchiarella later told Nicholls' that her maternity leave was a "bump in the road." Nicholls Dep. at 154-55. Cocchiarella made a similar comment to Mark Stanton, a union representative. Stanton Aff. ¶ 13 ("Ms. Cocchiarella stated Ms. Nicholls `hit a bump in the road when she missed a year' due to her maternity leave").
Stanton also voiced his concerns that the SPE system was unfair because it assigned zeroes to components of the system when employees were out on leave. Id. ¶ 7. He was told this was the way the system was intended to work. Id. He also voiced these concerns at Vosdingh and Nicholls' case management discussions but felt his objections were ignored. Id. ¶¶ 9, 10, 12, 15. In addition, Judy Scott, a sales director, made multiple inquiries concerning whether the SPE system was correctly incorporating the absence of data for employees on extended leaves, including CNC leaves, with respect to the internet sales component. Scott Dep. at 60-68, 74-75. She was told that the system was operating in the manner intended and there were no plans to change it. Id. Finally, subsequent to Plaintiffs' respective termination and resignation, Defendants adopted a normalization process that recalculated SPE scores to exclude extended leave time, including CNC leave.
After considering all of the proffered evidence, a reasonable jury could find that Defendants impermissibly retaliated against Plaintiffs in violation of the MPLA. As a result, Plaintiffs' MPLA claim survives summary judgment.
D. Title VII and MHRA Claims Regarding Unlawful Discrimination on the Basis of Sex
Plaintiffs claim Defendants' conduct impermissibly discriminated against them on the basis of sex. Both Title VII and the MHRA prohibit employers from discriminating because of an employee's sex. 42 U.S.C. § 2000e, et seq.; Minn. Stat. § 363A.08. Both Title VII and MHRA claims are analyzed under the same rubric. Mems v. City of St. Paul, 224 F.3d 735, 738 (8th Cir. 2000) (citations omitted). Sex discrimination may be established under either a disparate impact or a disparate treatment theory.
1. Disparate Impact
To establish a disparate impact claim, a plaintiff must prove, as a threshold matter, that the challenged employment practice, while facially neutral, has a disparate impact on certain employees "because of their membership in a protected group." EEOC v. McDonnell Douglas Corp., 191 F.3d 948, 950 (8th Cir. 1999) (citing Watson v. Fort Worth Bank and Trust, 487 U.S. 977, 994 (1988) (plurality opinion)). To present a prima facie case, a plaintiff must (1) identify the particular, facially neutral employment practice that she is challenging; (2) demonstrate a disparate impact on the protected group to which she belongs; and (3) prove that the particular challenged practice caused the disparate impact. See McDonnell Douglas, 191 F.3d at 950. "Statistical evidence may be probative where it reveals a disparity so great that it cannot be accounted for by chance, or, to state in other words, the statistical disparities must be sufficiently substantial that they raise an inference of causation." Waisome v. The Port Authority of New York, 948 F.2d 1370, 1375 (2d Cir. 1991). The Supreme Court has recognized a standard deviation greater than "two or three" between the expected value and the observed number raises an inference of discrimination. Castaneda v. Partida, 430 U.S. 482, 497 n. 17 (1977). Other evidence of discrimination, coupled with statistical evidence, can also establish a prima facie case.See Waisome, 948 F.2d at 1375; Mitchell v. Board of Trustees of Pickens County, 599 F.2d 582, 586 n. 7 (4th Cir. 1979).
If a plaintiff establishes a prima facie case, the burden shifts to the employer to demonstrate that the challenged practice is job related and consistent with business necessity.See Watson, 487 U.S. at 994. Finally, "[i]f the employer successfully establishes a business justification, the plaintiff may still prevail by demonstrating that a comparably effective alternative practice would produce a significantly smaller adverse impact on the protected class." Evers v. Alliant Techsystems, Inc., 241 F.3d 948, 954 (8th Cir. 2001) (citingWatson, 487 U.S. at 998).
Plaintiffs have identified the component of the SPE system that assigns zeroes to categories of data while employees are on company approved leave as the facially neutral employment practice. Since this absence of data continues to impact employees' overall SPE scores for the next 24 months, Plaintiffs contend the SPE system penalizes any employee, male or female, who takes an extended, company approved leave, regardless of whether it is disability leave, CNC leave, pregnancy leave, or other extended absence. Plaintiffs present statistical evidence to support this assertion. 52 leaves of absence of 12 weeks or more were taken by 48 employees from the third quarter of 1999 until January 1, 2004; 23 for child care purposes, 25 for illness or medical reasons and 4 were for other reasons. Martin Aff. Exs. B, C. Of the 52 leaves of absences, 43 leaves (83%) resulted in SPE scores that declined relative to the median from the time the employee went on leave to the time they returned to work. Martin Aff. Exs. B, C. This 83% decrease is more than four standard deviations from what would be expected if leave was a neutral event under the SPE. Martin Aff. Exs. B, C. This percentage increases as the leaves become longer: 31 of 33 employees taking leaves of 20 weeks or more and 21 of 21 employees taking leaves of six months or more saw their overall SPE scores decrease. Martin Aff. Exs. B, C.
Plaintiffs also aver that the magnitude of the decline in SPE scores is significant and becomes more severe as the duration of the leaves increase. Of the 52 extended leaves of absence, 36 leaves (69%) of more than 12 weeks resulted in percentage declines of 10% or more relative to the universe median, while 24 leaves (46%) resulted in a decrease of 20% or more. Martin Aff. Ex. D. The data responsible for these declines in SPE scores continues to exert downward pressure on employees' overall SPE scores for the next 24 months. It is also uncontested that these SPE scores translate directly to employees' performance ratings, provided they have worked a sufficient number of days per quarter.
Plaintiffs also offer anecdotal evidence, largely describedsupra, to demonstrate the SPE system penalized employees who took extended, company-approved leave. See also Roberts Dep. at 94-95 ("there could be some declining pressure, some pressure because of everybody else being able to continue to move up; that is true"), 68-69, 167-68; Cocchiarella Dep. at 56 ("if you are gone for a very long period of time and not having any data input, it can give you a lower score"), 50-62.
Plaintiffs' statistical evidence also suggests the SPE system's inclusion of the lack of data generated by employees on extended, company-approved leaves had a disparately negative impact on women. 36 of the 52 employees (69%) who took leaves of absence of 12 weeks or more were women, while only 48% of the sales consultant workforce were women. Martin Aff. Ex. C. These figures are more than 2 standard deviations from the result expected under a gender blind system. Id. Furthermore, 81% of employees taking more than 20 weeks of leave were women, a figure more than 3.5 standard deviations from a gender neutral model. Id. In addition, Plaintiffs again rely on the anecdotal comments to bolster their claim that the SPE system had a disparate impact on female sales consultants.
Defendants contest the validity of Plaintiffs' statistical evidence. They argue that the proffered statistics only show that employees' overall SPE scores decline over an extended leave period, but do not show this decline results in a lower performance rating or disciplinary action. Furthermore, they note 17 of the 52 leave takers included in Plaintiffs' study elected not to return to work and argue this data should not be considered. Sprenkle Aff. [Docket No. 77] ¶ 3. Defendants present their own statistical evidence, which they claim indicates no disparate impact occurred as a result of the SPE system. McCall Aff. (Kiovsky Aff. [Docket No. 42] Ex. K). The conflicting statistical evidence represents a material fact dispute best resolved at trial. At this juncture of the lawsuit, the Court is satisfied that Plaintiffs have presented sufficient evidence to present a prima facie disparate impact claim.
Under the disparate impact framework, once an employee has established a prima facie case, it is the employer's burden to demonstrate a legitimate, business-related justification for the challenged practice. See Watson, 487 U.S. at 994. If the employer meets this burden, the employee may then show that "a comparably effective alternative practice would produce a significantly smaller adverse impact on the protected class."Evers, 241 F.3d at 954 (citing Watson, 487 U.S. at 998). Assuming, arguendo, Defendants can demonstrate that the SPE system is a legitimate, job related means of evaluating sales consultants' performances and disciplining underachieving employees, Plaintiffs can sustain their burden of overcoming such a showing. Defendants have since adopted a normalization process that excludes company approved leave from SPE system consideration. As a result, Defendants' actions indicate the feasibility of implementing a comparably effective performance system that does not consider company approved leave time and thereby has a less adverse impact on women.
For the aforementioned reasons, Plaintiffs' disparate impact claim under Title VII survives summary judgment.
2. Disparate Treatment
In the absence of sufficient direct evidence of discrimination, Title VII disparate treatment claims are analyzed under the burden shifting analysis established in McDonnell Douglas. See Kiel, 169 F.3d at 1134-36. To demonstrate a prima facie case of disparate treatment on the basis of sex, a plaintiff must prove: (1) she belonged to a protected class; (2) she was qualified to perform her job; (3) she suffered an adverse employment action; and (4) non-members of her class were treated differently. See Bredding v. Arthur J. Gallagher and Co., 164 F.3d 1151, 1156 (8th Cir. 1999). The fourth element of a discrimination claim has been described as requiring some "evidence that would give rise to an inference of unlawful discrimination." Putnam Search Term End v. Unity Health Sys., 348 F.3d 732, 735-36 (8th Cir. 2003).
To pursue their disparate treatment claims, Plaintiffs must first show membership in a protected class. Plaintiffs argue both pregnancy and sex are protected classes to which they belong. Defendants claim Plaintiffs' pregnancy-related leaves were too far removed temporally from their adverse employment action to establish membership in a pregnancy protected class. Defendants also contend Plaintiffs' sex discrimination claim is based on allegedly discriminatory conduct incurred as a result of taking CNC leave. CNC leave, by definition, is a leave provided to employees, male or female, to permit them to care for their children following birth or adoption. Defendants rely on the Eighth Circuit's holding that parenting is not a protected class and cannot serve as the basis for a Title VII discrimination claim. Piantanida v. Wyman Center, Inc., 116 F.3d 340, 342 (8th Cir. 1997).
The Pregnancy Discrimination Act amended Title VII's prohibition on sex discrimination to include discrimination on the basis of:
pregnancy, childbirth, or related medical conditions; and women affected by pregnancy, childbirth, or related medical conditions shall be treated the same for all employment-related purposes, including receipt of benefits under fringe benefit programs, as other persons not so affected but similar in their ability or inability to work.42 U.S.C. § 2000e(k). To establish she is a member of this protected class, a plaintiff "must do more than show she was, past tense, pregnant." Brinkman v. State Dept. of Corrections, 863 F. Supp. 1479, 1486 (D. Kan. 1994). She must show that her medical conditions at or near the time of her adverse employment action were related to her pregnancy.See id.
More than 19 months elapsed between the birth of Vosdingh's child on May 10, 2000 and her termination in February 2002. Nicholls gave birth to her first child on January 18, 2000, nearly two years before her February 2002 demotion. Neither Plaintiff has presented any evidence that medical conditions related to pregnancy or childbirth affected her performance after returning to work. Furthermore, there is no indication that the limited time Plaintiffs missed work before giving birth to their respective children (approximately one month in both cases) impacted their SPE scores sufficiently to constitute a causal factor in their ultimate termination many months later. See Sura v. Stearns Bank, N.A., 2002 U.S. Dist. LEXIS 25376, * 16 (D. Minn. Dec. 18, 2002) (finding a time of three and a half months between birth of plaintiff's child and the adverse employment action was too attenuated to support inference of pregnancy discrimination). Finally, although Nicholls was pregnant with her second child at the time she was demoted, the gravamen of her Complaint is that the SPE system penalized her for taking CNC leave following the birth of her first son. She contends this downward pressure resulted in the unsatisfactory performance ratings that led to her demotion. She does not allege demotion as a result of telling Cocchiarella that she was pregnant for a second time. As a result, Plaintiffs can not establish they are members of the protected class for pregnancy discrimination.
Title VII and the MHRA also prohibit discrimination on the basis of an individual's sex. 42 U.S.C. § 2000e-2(a)(1). Defendants correctly note discriminatory treatment based solely on the status of being a parent is not prohibited by sex discrimination laws. Piantanida, 116 F.3d at 342. "However, where an employer's objection to an employee's parental duties is actually a veiled assertion that mothers, because they are women, are insufficiently devoted to work, or that work and motherhood are incompatible, such treatment is gender based and is properly addressed under Title VII." Plaetzer v. Borton Automotive, Inc., 2004 U.S. Dist. LEXIS 19095, *17 (D. Minn. Aug. 13, 2004) (citing Back v. Hastings, 365 F.3d 107, 121 (2d Cir. 2003)). Plaintiffs argue they can establish a "sex plus" discrimination claim, i.e., based on their gender plus status as parents. See Back, 365 F.3d at 118-19. Viewing the facts in the light most favorable to the non-moving party, the Court finds Plaintiffs have presented sufficient evidence to establish they belong to a protected class.
To establish the second element of a prima facie disparate treatment case, Plaintiffs must show they were qualified for their positions. Prior to going on CNC leave in April 2000, Vosdingh had never received an unsatisfactory rating in her telephone sales consultant position. She had been recognized as a top performer in the St. Paul telephone universe and won the President's Circle of Excellence for her performance from 1998-1999. In the first quarter of 2000, the last full quarter before Vosdingh went on leave, she received a satisfactory rating. As a result, Vosdingh has made a prima facie showing that she was qualified to perform her job.
Whether Nicholls was qualified to perform her job as a premise sales consultant requires more analysis. Nicholls had six consecutive quarters of satisfactory performance as a St. Paul premise sales consultant before transferring to the Minneapolis premise office. In St. Paul, she twice won the Vice President's Circle of Excellence Award. Upon moving to Minneapolis, however, Ms. Nicholls was rated unsatisfactory for two consecutive quarters before she began a CNC leave in mid-December 1999. Although Nicholls received two unsatisfactory reviews before beginning her leave, her prior satisfactory ratings, coupled with Defendants' policy of terminating only employees who have four quarters of unsatisfactory ratings, is evidence Nicholls was qualified for her position. The stated purpose of the SPE system is to mitigate short-term market inconsistencies over a 24 month period. Since company policy would have precluded Defendants from taking adverse employment action against Nicholls prior to taking CNC leave, the Court cannot say she was not qualified for the position.
Defendants argue the number of successive quarters in which Vosdingh and Nicholls failed to achieve a satisfactory performance rating indicate that they were not qualified for their positions. Plaintiffs contend, however, their performance would not have been unsatisfactory if the SPE system did not impermissibly consider the lack of data generated while they were on CNC leave. An employer cannot implement a discriminatory performance system and then justify its actions by relying on the tainted system. See Stacks v. Southwestern Bell Yellow Pages, Inc. 27 F.3d 1316, 1325 (8th Cir. 1994). Whether Vosdingh and Nicholls were qualified at the time they suffered their respective adverse employment actions depends on the impact of the SPE system and raises a question of fact.
The Court has previously found Plaintiffs suffered adverse employment actions to satisfy the third prong of the prima facie case for disparate treatment.
To establish the final element of a prima facie case, Plaintiffs must show they were treated differently than nonmembers of their class. As previously noted, satisfying this element requires some "evidence that would give rise to an inference of unlawful discrimination." Putnam, 348 F.3d at 735-36.
In Nev. Dep't of Human Res. v. Hibbs, the Supreme Court recognized statements that suggest "women's family duties trump those of the workplace" are "gender stereotype[s]" sufficient to establish a prima facie case of discriminatory intent. 538 U.S. 721, 731 n. 5 (2003). When Nicholls told Cocchiarella she was pregnant for the second time, her manager responded by asking "what are you going to do about your job?" Nicholls Dep. at 32-34. Cocchiarella also stated "It's hard to come back after having a child" and "It's hard to do this job with two kids, look what happened to Andrea." Id. When Vosdingh briefly returned from CNC leave in October 2000, her manager at the time, Lynn Hanf, allegedly made derogatory comments about her need for a place to express breast milk, her choice to return to work, her status as an unwed mother and her need to go to church. Vosdingh Aff. [Docket 84] ¶ 5. After discussing Vosdingh's decision to nurse, Vosdingh claims Hanf commented "I know how hard this is for you, isn't there any way you can stay at home?" Vosdingh also claims she had a number of conversations with managers about being promoted to a premise position, but these conversations ceased once she announced her pregnancy. Vosdingh Dep. at 77-79.
Despite the SPE system's stated goal that a "case management call will insure consistency across all universes," Plaintiffs claim males and non-pregnant female sales consultants were frequently given extra chances after receiving four unsatisfactory ratings to correct their performances. See SPE System at 28; Ex. 131 (Recher Aff. [Docket No. 60]); Roberts Dep. at 105-17; Cocchiarella Dep. at 187-93 (12 instances of males being allowed an additional quarter after receiving four unsatisfactory ratings). Additionally, Plaintiffs claim Defendants did not take into account "extenuating circumstances," including "[t]ime away from the job" as required by the SPE System agreement. SPE System at 28. Nicholls also claims Defendants did not consider she was trending upward and was close to achieving a satisfactory rating. The Court is satisfied that this evidence, when viewed in the light most favorable to Plaintiffs, is sufficient to give rise to an inference of unlawful discrimination.
The burden now shifts to Defendants to articulate some legitimate, non-discriminatory reason for the adverse employment actions taken against Plaintiffs. Defendants have consistently claimed Vosdingh was terminated and Nicholls was demoted based on four consecutive quarters of unsatisfactory performance ratings. Defendants also claim Vosdingh suffered from sub-par work flow and lack of follow-through on coaching plans. See 2/4/02 Case Management Notes; April 2, 2002 Meeting Notes. Nicholls had two additional unsatisfactory ratings prior to her first CNC leave. Defendants allege that, despite the company's informal policy of terminating employees with four unsatisfactory ratings in eight quarters, they provided Nicholls with an opportunity to return to a position in which she had been successful. For the purposes of this motion, Defendants have carried their burden of demonstrating non-discriminatory reasons for the adverse employment actions.
The burden then returns to Plaintiffs to show the employers' stated reason was a mere pretext for discrimination. Plaintiffs again rely on their expert statistical analysis coupled with statements made by Defendants purportedly evidencing their intent to discriminate. See Hazelwood School District v. United States, 433 U.S. 299, 307-10 (1977) (endorsing the use of statistical evidence in Title VII disparate treatment case). Again, Plaintiffs' statistical evidence indicates that a statistically significant number of sales consultants who took company-approved leaves of 12 weeks or more experienced a decline in their overall SPE scores relative to the universe median. The extent of the decline was significant and its severity correlated to the length of the leave. Furthermore, the policy had a demonstrable disparate impact on women as 81% of employees taking leaves of 20 weeks or more were female. Women also took 22 of the 23 CNC leaves.
The requisite intent to discriminate against the protected class can be inferred if Plaintiffs can establish Defendants were aware of the policy's disparate impact but persisted in the conduct. See Equal Opportunity Commission v. Inland Marine Ind., 729 F.2d 1229, 1235 (9th Cir. 1984). Cocchiarella stated that Nicholls' maternity leave was a "bump in the road" and testified at her deposition, "if you are gone for a very long period of time and not having any data input, [an extended leave] can give you a lower score." Nicholls Dep. at 54-55; Stanton Aff. ¶ 13; Cocchiarella Dep. at 56. After seeing the impact of extended leaves, including CNC leaves, on sales consultant's SPE scores, Scott inquired multiple times as to whether the SPE system was correctly including the absence of data. Scott Dep. at 60-68, 74-75. She was told that the system was operating in the manner intended and there were no plans to change it. Id. Nicholls and Vosdingh also repeatedly complained to their respective managers about the perceived unfairness of having the SPE system include data from CNC leaves into their overall scores. See Vosdingh Aff. ¶ 11; 2/4/02 Case Management Notes; Nicholls Dep. II at 133; 12/31/01 Annual Review at 6. Roberts, the SPE administrator for Qwest Dex, also testified "at some point, it became apparent that we needed to look at [the impact of extended leave on SPE scores]. I don't know when that was." Id. at 120-21. Ultimately, they claim Defendants' legitimate, non-discriminatory reason — namely, that Plaintiffs suffered adverse employment actions as a result of sustained unsatisfactory performance ratings — is pretextual because these ratings were based on an SPE system tainted by discriminatory conduct. When viewed in the light most favorable to the non-moving party, the Court finds Plaintiffs have presented sufficient evidence to survive summary judgment on their disparate treatment claim.
E. Title VII and MHRA Retaliation Claims
Finally, Plaintiffs claim Defendants retaliated against them in violation of Title VII and the MHRA. 42 U.S.C. § 2000e, et seq.; Minn. Stat. § 363A.15. Claims of retaliation are also analyzed under the McDonnell Douglas burden shifting framework. To establish a prima facie case of retaliation under Title VII, a plaintiff must show: (1) she engaged in statutorily protected activity; (2) the employer subjected plaintiff to an adverse employment action; and (3) a causal connection existed between the two. Kipp v. Missouri Highway Transp. Comm'n, 280 F.3d 893, 896 (8th Cir. 2002). The same standards apply to retaliation claims brought under the MHRA. Buettner v. Easter Arch Coal Sales, 216 F.3d 707, 714 (8th Cir. 2000).
Plaintiffs do not specify the statutorily protected activity they engaged in which led to the alleged retaliation. As previously noted, Vosdingh and Nicholls repeatedly complained to their respective managers about the manner in which the SPE system incorporated their CNC leave. See Vosdingh Aff. ¶ 11; 2/4/02 Case Management Notes; Nicholls Dep. II at 133; 12/31/01 Annual Review at 6. Although Plaintiffs pursued the initial stages of grievances and filed complaints with the EEOC, these actions did not occur until after Vosdingh and Nicholls suffered their respective adverse employment actions. Assuming,arguendo, their complaints to management about the manner in which the SPE system incorporated CNC leaves into their overall scores is protected activity, Plaintiffs have failed to show a causal connection between the protected activity and the adverse employment actions.
Plaintiffs have offered no evidence to suggest that an employer's "retaliatory motive played a part in the adverse employment action." Kipp, 280 F.3d at 897 (citation omitted). The Eighth Circuit has repeatedly held a temporal connection alone is generally insufficient to establish causation. See, e.g., id. The Eighth Circuit has found an interval of two months between the protected activity and the adverse employment action is too long to establish a temporal connection. Id. Plaintiffs' complaints to their managers occurred intermittently from the time they returned from CNC leave until their respective termination and demotion. Such a period is sufficiently long to dilute any inference of causation. The evidence suggests Plaintiffs suffered adverse employment actions because they received SPE scores that correlated with successive unsatisfactory ratings, rather than because they complained about the way the SPE score was calculated. As a result, Plaintiffs have not established a prima facie case of retaliation under Title VII or the MHRA and these claims are dismissed.
For the aforementioned reasons, Plaintiffs' interference and retaliation claims under the FMLA (Count I) and retaliation claims under Title VII and the MHRA are dismissed (Counts V and VI). Plaintiffs' MPLA claim (Count II), and disparate impact and disparate treatment claims under Title VII (Count III) and the MHRA (Count IV) survive summary judgment.
IV. CONCLUSION
Based upon the foregoing, and all the files, records, and proceedings herein, IT IS HEREBY ORDERED that:1. Plaintiffs' Motion for Partial Summary Judgment [Docket No. 44] is DENIED;
2. Defendants' Motion for Summary Judgment [Docket No. 40] is GRANTED as to Counts I, V and VI and DENIED as to Counts II, III and IV.