Tony Mumfrey v. CVS Pharmacy, Inc.

Opinion

REVISED June 25, 2013

IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT United States Court of Appeals Fifth Circuit

FILED No. 12-40419 June 10, 2013

Lyle W. Cayce Clerk TONY MUMFREY,

Plaintiff–Appellant v.

CVS PHARMACY, INC. ET AL.,

Defendants–Appellees

Appeal from the United States District Court for the Eastern District of Texas

Before REAVLEY, PRADO, and ELROD, Circuit Judges. EDWARD C. PRADO, Circuit Judge: Plaintiff–Appellant Tony Mumfrey (“Mumfrey”) appeals the district court’s denial of his motion to remand his retaliation suit against his former employer, CVS Pharmacy, Inc. (“CVS”). He also appeals the district court’s conclusions of law, arguing that it erroneously applied controlling law to conclude that Mumfrey did not prove retaliatory termination. We affirm. I. FACTUAL AND PROCEDURAL BACKGROUND A. Factual Background Mumfrey worked as a CVS pharmacist from 2004 until his termination in February 2009. Pharmacy Supervisor Gary LeBlanc was Mumfrey’s direct No. 12-40419

supervisor from mid-2006 until the end of 2008, when LeBlanc transferred to California. A district manager, Michael Cooney, became Mumfrey’s supervisor in January 2009. The store manager, Doug Jenny,1 had no supervisory responsibilities over Mumfrey. Starting in October 2008, Mumfrey began exhibiting multiple performance issues. Mumfrey had received oral warnings in the past for violating the company’s prohibition on sharing one’s cash register number and password with others. In October, LeBlanc learned Mumfrey had again given out his cash register number and password to two individuals in the pharmacy. LeBlanc issued Mumfrey two “coaching and counseling” forms, one for each violation.2 LeBlanc also issued Mumfrey a third “coaching and counseling” form based on a customer complaint filed with CVS on October 17, 2008. The customer stated that Mumfrey had given her his personal phone number. She said he called and e-mailed her and that the contact was unwanted. Mumfrey would not disclose to LeBlanc how he received her contact information, other than to say it was from an outside source. Cooney was present when Mumfrey received each of the three October 2008 “coaching and counseling” forms. Mumfrey’s performance issues continued. In November and December 2008, multiple pharmacy technicians complained to LeBlanc about Mumfrey’s behavior. On December 4, 2008, LeBlanc issued Mumfrey another “coaching and counseling” form for creating a hostile working environment, exhibiting poor customer service, and yelling and speaking rudely to a technician. Although LeBlanc had been informally addressing Mumfrey’s professionalism for years,

1 Jenny’s name was misspelled as “Jenney” in the Original Petition and the misspelling has persisted throughout the litigation. 2 LeBlanc had also received oral warnings in the past for sharing his register password, but had complied with the rule following the warning.

2 No. 12-40419

he issued the warning on this occasion to document the incident. He wrote on the form that it was Mumfrey’s final warning for that issue. Mumfrey’s work troubles continued. CVS requires a pharmacist to complete a mis-fill report within twenty-four hours of mis-filling a prescription. A pharmacist mis-fills by giving the customer a different medication, strength, or quantity than prescribed. On December 16, 2008, Mumfrey mis-filled a prescription by not providing the strength and form that the physician directed. Mumfrey did not complete the mis-fill report, despite being reminded to. He then initiated a medical leave of absence from December 17, 2008 to January 26, 2009.3 LeBlanc did not think it was appropriate to discipline Mumfrey during his leave of absence. Because LeBlanc was transferred to California during Mumfrey’s leave, Cooney became Mumfrey’s direct supervisor upon his return. Cooney finally issued a “coaching and counseling” for Mumfrey’s failure to fill out the mis-fill report before his medical leave of absence. In the “coaching and counseling” session, Cooney told Mumfrey to consider the session a final warning on behavior and performance, and warned Mumfrey that any future issues could lead to termination. On February 2, 2009, Mumfrey e-mailed CVS Human Resources Business Partner Todd Hine to complain about Hine’s handling of Mumfrey’s return-to- work authorization. The authorization stated that Mumfrey be allowed to “sit as needed.” Hine had asked Mumfrey for clarification on what that meant. Mumfrey replied by sending Hine the dictionary definitions of the three words. On February 4, 2009, Mumfrey e-mailed and faxed a complaint to CVS’s Ethics Line, stating he believed that he was receiving “coaching and counseling”

3 Mumfrey had received an accommodation to use a special chair at work to alleviate back pain. He initially pleaded that he was retaliated against for requesting an accommodation as well, but has abandoned that claim on appeal.

3 No. 12-40419

sessions for trivial reasons in retaliation for requesting health accommodations. On February 5, 2009, Mumfrey alleged retaliation to the EEOC. On February 16, 2009, a customer complained that when he brought his Category II prescription4 to CVS, Mumfrey told him that he would have to verify the prescription with the doctor and would not be able to fill it that day. Based on the customer complaint, Cooney initiated an investigation by calling the customer. Cooney advised the customer to send him the complaint via e-mail. The customer’s e-mail stated that on Sunday, February 15, 2009, he had driven to Beaumont from a family event in Fort Polk, Louisiana. He said Mumfrey told him to come back the next day because Mumfrey would have to verify the prescription with the prescribing doctor because the doctor was from out of town. According to the customer’s complaint, when he asked Mumfrey to look at the customer’s past prescriptions in the CVS system to verify that he received the same medicine and quantity from the same doctor every month, Mumfrey told him that he did not care and that he could come back the next day. Mumfrey did not attempt to call the doctor to verify the prescription. The customer returned the next day and received the medicine without further incident. Cooney investigated the complaint. He concluded that a prescription from an out-of-town doctor was not a reason to not immediately fill it. Neither was the fact that the prescription’s release date was written for five days after the date the prescription was signed. CVS’s computers indicated that the customer had previously filled a prescription for the same medicine at the same store from the same doctor. The previous fill was more than thirty days prior and had been for thirty days’ worth of medicine. Mumfrey had claimed that insurance rejected the prescription and that the customer had brought the prescription to other pharmacies and been turned away. However, Cooney called the customer’s

4 Category II prescriptions are more strictly regulated and are kept in a locked safe.

4 No. 12-40419

insurance company, which told him that no prescription claim had been made by Mumfrey on February 15, 2009, and that no other pharmacy had recently presented the same prescription to the insurance company. After watching a video of Mumfrey’s interaction with the customer, Cooney invited Human Resources Director Jimmy Griffin to interview Mumfrey with him. Because Mumfrey had previously filed complaints against Cooney, Cooney intended for Griffin to make the ultimate determination. Griffin was not satisfied with Mumfrey’s explanation of events. At the meeting, Mumfrey protested that another pharmacy technician had mis-filled a prescription and not been disciplined. Griffin, concerned that Mumfrey had not attempted to verify the prescription, decided to terminate Mumfrey. B. Procedural Background Mumfrey filed suit on October 14, 2009 against CVS for the following Texas Labor Code claims: (1) retaliation for seeking an accommodation based on his good faith belief that he had a disability, and (2) unlawful retaliatory termination after Mumfrey filed an EEOC charge of discrimination.5 He also filed suit against Mike Cooney (“Cooney”), Gary LeBlanc (“LeBlanc”), and Douglas Jenny (“Jenny”) (collectively “the individual defendants”). His suit against the individual defendants claimed tortious interference with business relationships, defamation, and constructive fraud.6 Mumfrey’s Original Petition did not specify an amount in controversy and pleaded only general categories of damages. Mumfrey’s Original Petition requested the following damages: “compensatory or pecuniary damages for past lost wages up to the time of trial, and future pecuniary losses; past, present and

5 Mumfrey does not appeal the district court’s decision on his claim of retaliation for requesting a workplace accommodation. 6 Mumfrey abandoned his claims for defamation and constructive fraud against the individual defendants.

5 No. 12-40419

future emotional pain, suffering, inconvenience, mental anguish, loss of enjoyment of life; and other past and future nonpecuniary losses. . . . Plaintiff requests an additional amount in attorneys’ fees, to be set by the court, pursuant to statute.” CVS moved the state district court to require Mumfrey to specify the maximum amount of damages sought. The state court granted the motion, and in response, Mumfrey filed a First Amended Petition claiming $3,575,000 in damages. Within thirty days of requesting and receiving the First Amended Petition, CVS removed to federal court. CVS, acknowledging that complete diversity did not exist on the face of the complaint, asserted that removal was nonetheless proper because the individual defendants from Texas were improperly joined, and that without them, complete diversity existed. Mumfrey filed a motion to remand, arguing that CVS’s removal was untimely and that the individual defendants were not improperly joined. The district court issued an order finding CVS’s removal timely and the individual defendant’s improperly joined such that the district court had jurisdiction. Mumfrey filed a motion to reconsider, which the district court denied. Several months after a three-day bench trial, the court entered Findings of Fact and Conclusions of Law finding that Mumfrey did not prove retaliation. Mumfrey timely appealed. II. JURISDICTION A. Standard of Review This court reviews a district court’s denial of a motion to remand for lack of subject matter jurisdiction de novo. Allen v. R & H Oil & Gas Co., 63 F.3d 1326, 1336 (5th Cir. 1995). B. Timeliness of Removal A defendant can remove to federal court any civil action brought in state court over which the district court would also have had original jurisdiction. 28

6 No. 12-40419

U.S.C. § 1441(a). A federal district court has subject matter jurisdiction over a state claim when the amount in controversy is met and there is complete diversity of citizenship between the parties. See 28 U.S.C. § 1332(a). The amount in controversy required by § 1332(a) is currently $75,000. Id. Any ambiguities are construed against removal and in favor of remand to state court. Manguno v. Prudential Prop. & Cas. Ins. Co., 276 F.3d 720, 723 (5th Cir. 2002) (citing Acuna v. Brown & Root, Inc., 200 F.3d 335, 339 (5th Cir. 1988)). The party seeking to remove bears the burden of showing that federal jurisdiction exists and that removal was proper. Id. (citations omitted). The timing of removal is controlled by 28 U.S.C. § 1446(b) as follows:7 The notice of removal of a civil action or proceeding shall be filed within thirty days after the receipt by the defendant, through service or otherwise, of a copy of the initial pleading setting forth the claim for relief upon which such action or proceeding is based, or within thirty days after the service of summons upon the defendant if such initial pleading has then been filed in court and is not required to be served on the defendant, whatever period is shorter. . . .

If the case stated by the initial pleading is not removable, a notice of removal may be filed within thirty days after receipt by the defendant, through service or otherwise, of a copy of an amended pleading, motion, order or other paper from which it may first be ascertained that the case is one which is or has become removable, except that a case may not be removed . . . more than 1 year after commencement of the action.

28 U.S.C. § 1446(b). Restated, if the initial pleading sets forth a claim that triggers the removal clock, the defendant must file notice of removal within

7 28 U.S.C. § 1446 was amended in 2011. This Court relies on the version that was applicable at the time Mumfrey filed his Original Petition. The statute as amended is textually identical; only paragraph lettering was changed.

7 No. 12-40419

thirty days of receiving it.8 If the initial pleading did not trigger the thirty-day removal clock, a notice of removal must be filed within thirty days of the defendant’s receipt of a document from which it may ascertain that the case is, or has become, removable. Id. In any event, removal may not occur more than one year after commencement of the action. Id. At the time Mumfrey initiated his suit, Texas prohibited plaintiffs from pleading a specific amount of unliquidated damages. Tex. R. Civ. P. 47(b) (“An original pleading . . . shall contain . . . in all claims for unliquidated damages only the statement that the damages sought are within the jurisdictional limits of the court.”) (emphasis added);9 see also Unauthorized Practice of Law Comm’n v. Am. Home Assur. Co., 261 S.W.3d 24, 40 (Tex. 2008) (“Texas procedure does not permit a plaintiff claiming unliquidated damages . . . to state a dollar figure in his petition.”). Under the rule at the time of Mumfrey’s petition, there were two main ways that a plaintiff disputed a case’s removal. The first type of dispute arose when the defendant removed a case within thirty days of receiving the initial pleadings—before the amount in controversy was clearly established—and the plaintiff moved to remand, objecting that the amount in controversy had not been met. In that situation, the defendant had to prove by a preponderance of the evidence that the amount-in-controversy requirement was satisfied. De Aguilar v. Boeing Co., 47 F.3d 1404, 1409 (5th Cir. 1995)

8 If the defendant seeks to remove on the basis of an initial pleading where the jurisdictional amount is not established, the removing defendant must demonstrate by a preponderance of the evidence that the jurisdictional amount is satisfied. De Aguilar v. Boeing Co., 47 F.3d 1404, 1409 (5th Cir. 1995) (quoting De Aguilar v. Boeing Co., 11 F.3d 55, 58 (5th Cir. 1993)). In so determining, the court can rely on “summary judgment-type” evidence to ascertain the amount in controversy. St. Paul Reins. Co., Ltd. v. Greenberg, 134 F.3d 1250, 1253 (5th Cir. 1998). This situation is not before us. 9 Texas Rule of Civil Procedure 47(b) was amended in an order dated February 12, 2013, effective March 1, 2013. It now states: “An original pleading . . . shall contain . . . a statement that the damages sought are within the jurisdictional limits of the court.” Neither party briefs whether this amendment allows a plaintiff to plead a specific damages amount, and we need not consider the issue in this case.

8 No. 12-40419

(quoting De Aguilar v. Boeing Co., 11 F.3d 55, 58 (5th Cir. 1993)). Almost all removal cases in the circuit fall into this first category—“amount disputes.” The second situation—“a timeliness dispute”—occurs when the defendant did not remove within thirty days after receipt of the initial pleadings under § 1446(b)’s first paragraph, but instead removed under the second paragraph, within thirty days of receiving some amended pleading or “other paper.” In a timeliness dispute, the plaintiff tries to avoid removal by arguing that it was clear from the initial pleadings that the case was removable such that the defendant has missed the deadline and is forever barred from removing. Distinguishing between the two removal disputes is critical because different standards apply to each. It is the second scenario—“a timeliness dispute”—we are presented with. Because CVS filed its notice of removal more than thirty days after receiving Mumfrey’s Original Petition,10 but within thirty days of receiving the First Amended Petition, the central issue here is whether Mumfrey’s Original Petition triggered the thirty-day time period for removal, even though it did not include a specific amount of liquidated damages. This situation arose because Texas, at the time the suit arose, did not permit plaintiffs to plead a specific amount of liquidated damages. Mumfrey argues that his initial pleadings were removable because his list of damages was so extensive it was clear his claims satisfied the jurisdictional amount. Specifically, he pleaded for lost wages and CVS, as his employer, knew his salary. CVS argues that it timely removed because Mumfrey’s Original Petition does not include any specific allegation that his claimed damages exceeded $75,000. In essence, CVS contends it should not have to rely on its subjective knowledge of Mumfrey’s salary from outside the pleading.

10 28 U.S.C. § 1446(b) refers to initial pleadings broadly. Here, the particular initial pleading is an Original Petition.

9 No. 12-40419

In our seminal case on timeliness disputes, Chapman v. Powermatic, Inc., 969 F.2d 160 (5th Cir. 1992), this Court held that the thirty-day removal period under the first paragraph is triggered only where the initial pleading “affirmatively reveals on its face that the plaintiff is seeking damages in excess of the minimum jurisdictional amount of the federal court.” Id. at 163 (emphasis added). The Chapman court specifically declined to adopt a rule which would expect defendants to “ascertain[] from the circumstance[s] and the initial pleading that the [plaintiff] was seeking damages in excess of the minimum jurisdictional amount.” Id. The Chapman court had several practical ramifications in mind when it announced this standard. By rejecting a so-called due-diligence standard, it sought to promote efficiency by preventing courts from expending copious time determining what a defendant should have known or have been able to ascertain at the time of the initial pleading. Id. Morever, the Chapman court wanted to avoid encouraging defendants to remove cases prematurely for fear of accidentally letting the thirty-day window to federal court close when it is unclear that the initial pleading satisfies the amount in controversy. Id. Ultimately, Chapman lays out a “bright line rule requiring the plaintiff, if he wishes the thirty-day time period to run from the defendant’s receipt of the initial pleading, to place in the initial pleading a specific allegation that damages are in excess of the federal jurisdictional amount.” Id. (emphasis added). Such a statement would provide notice to defendants that the removal clock had been triggered, but would not run afoul of state laws that prohibit pleading unliquidated damage amounts. A later case reaffirmed the Chapman standard. In Bosky v. Kroger Tex., LP, the parties agreed that the complaint was insufficient to trigger the thirty- day removal clock. 288 F.3d 208, 210 (5th Cir. 2002). Though timeliness was

10 No. 12-40419

thus not at issue, the court attempted to clarify “the question of timeliness of removal.”11 Id. at 209. Bosky first reiterated Chapman’s holding: the first paragraph’s thirty-day limit is triggered “only when that pleading affirmatively reveals on its face that the plaintiff is seeking damages in excess of the minimum jurisdictional amount of the federal court.” 288 F.3d at 210 (quoting Chapman, 969 F.2d at 163). Bosky also reiterated Chapman’s bright line rule “requiring the plaintiff, if he wishes the thirty-day time period to run from the defendant’s receipt of the initial pleading, to place in the initial pleading a specific allegation that damages are in excess of the federal jurisdictional amount.” Id. In a line that has become the source of significant confusion, Bosky then went on: “[w]e have since held that specific damage estimates that are less than the minimum jurisdictional amount, when combined with other unspecified damage claims, can provide sufficient notice that an action is removable so as to trigger the time limit for filing a notice of removal.” Id. at 210 (emphasis added) (citing Marcel v. Pool Co., 5 F.3d 81, 82–85 (5th Cir. 1993), and DeAguilar, 47 F.3d 1408–12). However, both cases Bosky cites for this assertion are amount dispute cases—where the time limit was not triggered. Thus, Bosky’s above statement—“we have since held” that complaints stating unspecified damage amounts trigger the time limit—is incorrect. In Marcel, the defendant filed a notice of removal less than a month after the state court petition was filed. 5 F.3d at 82. The plaintiff sought to establish that his claim was below the amount in controversy, such that remand was required. Id. Thus, the court decided that the defendant was permitted to remove—not that the thirty-day

11 Because the court explained the standards for triggering the thirty-day removal period abstractly without reference to the particular facts before it, its discussion is dicta. Id. at 209–11.

11 No. 12-40419

removal clock had been triggered, as Bosky implied. See id. at 83. In De Aguilar, the plaintiffs filed in state court and, pursuant to Tex. R. Civ. P. 47, did not plead damages of a specific dollar amount. 47 F.3d at 1406. The defendants removed to federal court. Id. The issue was whether the plaintiffs, to avoid federal court, could specifically allege that their damages would not exceed the jurisdictional amount. Id. at 1409–10. Thus, despite Bosky’s statement, neither case considered what triggers the thirty-day window. Notably, there seems to be no Fifth Circuit case since Chapman that calls into question its bright line rule for timeliness disputes.12 Thus, the rule remains that the thirty-day clock is not triggered unless the initial pleading “affirmatively reveals on its face” that the plaintiff’s sought damages exceeding the jurisdictional amount. Chapman, 969 F.2d at 163; see also Rios v. City of Del Rio, Tex., 444 F.3d 417, 425 n.8 (5th Cir. 2006) (“The rule in this circuit is that where two previous holdings or lines of precedent conflict the earlier opinion controls . . . .”). Mumfrey relies exclusively on inapposite “amount dispute” cases, and fails to cite Chapman. His “amount dispute” cases held that removal was appropriate when the jurisdictional amount was “facially apparent,” despite the absence of a specific allegation that damages were in excess of the federal jurisdictional amount. See, e.g., Gebbia v. Wal–Mart Stores, Inc., 233 F.3d 880, 882–83 (5th Cir. 2000); Luckett v. Delta Airlines, Inc., 171 F.3d 295, 298 (5th Cir. 1999) (stating that, from “the face of the complaint,” the district court’s denial of remand was not error). However, the “facially apparent” inquiry is relevant only to “amount dispute” cases.

12 See, e.g., Leffall v. Dall. Indep. Sch. Dist., 28 F.3d 521, 525 (5th Cir. 1994) (emphasizing Chapman’s bright line rule that the removal clock runs only when the “pleading affirmatively reveals on its face that the plaintiff is seeking damages in excess of the minimum jurisdictional amount of the federal court”) (citation omitted).

12 No. 12-40419

Here, the parties agree that the initial complaint did not contain a “specific allegation that damages are in excess of the federal jurisdictional amount.” The removal clock was thus not triggered until CVS received a copy of an “amended pleading, motion, order, or other paper from which” it was first ascertainable that the case was removable, i.e., when Mumfrey filed his First Amended Petition seeking $3,575,000 in damages.13 C. Improper Joinder14 A district court’s improper joinder decision is subject to de novo review. McDonal v. Abbott Labs., 408 F.3d 177, 182 (5th Cir. 2005) (citing Great Plains Trust Co. v. Morgan Stanley Dean Witter & Co., 313 F.3d 305, 311 (5th Cir. 2002)). “[Improper] joinder can be established in two ways: (1) actual fraud in the pleading of jurisdictional facts, or (2) inability of the plaintiff to establish a cause of action against the non-diverse party in state court.” McKee v. Kansas City S. Ry. Co., 358 F.3d 329, 333 (5th Cir. 2004) (quoting Travis v. Irby, 326 F.3d 644, 647 (5th Cir. 2003)). CVS did not contend that the pleadings contained actual fraud; thus, only the second prong was before the district court. “[T]he test for [improper] joinder is whether the defendant has demonstrated that there is no possibility of recovery by the plaintiff against an in-state defendant, [restated,] there is no reasonable basis for the district court to predict that the plaintiff might be able to recover against an in-state defendant.” In re 1994 Exxon Chem. Fire, 558 F.3d 378, 385 (5th Cir. 2009)

13 We note that CVS likely could have removed immediately after Mumfrey filed his Original Petition, giving rise to an amount dispute case if Mumfrey had sought to remand by arguing that the amount-in-controversy was not satisfied. But even if CVS could have removed immediately, Mumfrey’s Original Petition did not start the clock such that CVS was required to remove if it wanted to. 14 The Fifth Circuit adopted the terminology “improper joinder,” Smallwood v. Ill. Cent. R.R. Co., 385 F.3d at 571 n.1, instead of the terminology “fraudulent joinder,” which is “a term of art” used in other circuits to describe the doctrine that ignores a lack of complete diversity where the plaintiff joins a nondiverse defendant to avoid federal jurisdiction. E.g., Morris v. Princess Cruises, Inc., 236 F.3d 1061, 1067 (9th Cir. 2001).

13 No. 12-40419

(some alteration in original) (emphasis added) (quoting Smallwood v. Ill. Cent. R.R. Co., 385 F.3d 568, 573 (5th Cir. 2004) (en banc)). Although the district court erred by improperly placing the burden on the plaintiff, the error is harmless because CVS demonstrated that Mumfrey had no possibility of recovering against the individual defendants, and that thus, the individual defendants were improperly joined. This Court’s en banc opinion in Smallwood sets out a procedure for determining whether a nondiverse defendant was improperly joined. First, the court should focus on the complaint: “Ordinarily, if a plaintiff can survive a Rule 12(b)(6) challenge, there is no improper joinder.” Id. at 573. However, where a complaint states a claim that satisfies 12(b)(6), but has “misstated or omitted discrete facts that would determine the propriety of joinder . . . the district court may, in its discretion, pierce the pleadings and conduct a summary inquiry.” Id. (citing Badon v. RJR Nabisco, Inc., 236 F.3d 282, 389 n.10 (5th Cir. 2000)). The purpose of the inquiry is limited to identifying “the presence of discrete and undisputed facts that would preclude plaintiff’s recovery against the in-state defendant.” Id. at 573–74. Ultimately, the defendant bears the burden: “the test for fraudulent joinder is whether the defendant has demonstrated that there is no possibility of recovery . . . .” Id. at 573 (emphasis added). Here, the district court, following Smallwood, evaluated the complaint and determined that Mumfrey failed to state a claim against the individual defendants. It then conducted a summary inquiry. Cf. Smallwood, 385 F.3d at 573. However, in doing so, the trial court placed the burden on the plaintiff, requiring him to show that he could possibly recover against the individual defendants notwithstanding his failure to state a claim: Having examined Defendant’s response and found Plaintiff’s pleadings insufficient, the Court allowed Plaintiff the opportunity to reply and offer facts and argument in support of his theories against the named individuals. Plaintiff’s reply contained

14 No. 12-40419

additional facts pertinent to his theories of discrimination and retaliation committed by C.V.S. but did not offer any additional support for his allegations of tortious interference with prospective business relationships, constructive fraud, or defamation.

Under Smallwood, however, not only is the initial burden on the defendant to show the complaint fails to state a claim, but if the court elects “in its discretion[] [to] pierce the pleadings and conduct a summary inquiry,” the burden remains with the defendant. Id. at 573–75 (“Our insistence that a removing defendant demonstrate that the joinder was improper . . .”; “[t]o justify removal on improper joinder grounds, [the defendant] was required to prove that the joinder of [non-diverse defendant] was improper.”). Moreover, the district court analyzed the wrong claim. The district court concluded that Mumfrey had insufficient evidence to support a claim for tortious interference with prospective business relationships. But, Mumfrey pleaded only tortious interference with business relationships, namely his employment contract with CVS. The two torts require different elements. See, e.g., Wal–Mart Stores, Inc. v. Sturges, 52 S.W.3d 711, 727 (Tex. 2001). Nonetheless, these errors are harmless because CVS has carried its burden and established that Mumfrey had no possibility of recovery against the individual defendants. In In re Exxon, this court held that individual defendants were improperly joined. 558 F.3d at 386–87. The court, in determining whether the individual defendants could be held liable under applicable state law, looked to a 1973 state case that articulated the four factors necessary for establishing individual liability. Id. at 385–86. The court applied the case’s test to the individual defendants, and determined they could not be individually liable under state law. Id. at 386. (“Restated, the conduct at issue did not involve personal fault giving rise to liability under Louisiana Law.”). The court held that there was “nothing in the record” to satisfy the third and fourth factors. Id. at 386 (“The evidence does not show that Paul had personal knowledge of the

15 No. 12-40419

condition of the valves . . . [and] Paul did not personally perform the valve swap.”). Thus, the individual defendants were improperly joined because the defendant demonstrated that there is no reasonable basis for the district court to predict that the plaintiff could recover against the in-state defendants. Id. at 386–87. Similarly, CVS has established that Mumfrey cannot satisfy the second element of tortious interference. Under Texas law, the elements of tortious interference with contract are: (1) the existence of a contract, (2) willful and intentional interference, (3) interference that proximately caused damage, and (4) actual damage or loss. Powell Indus., Inc. v. Allen, 985 S.W.2d 455, 456 (Tex. 1998) (citing ACS Investors, Inc. v. McLaughlin, 943 S.W.2d 426, 430 (Tex. 1997)). “When the defendant is both a corporate agent and the third party who allegedly induces the corporation’s breach, the second element is particularly important.” Id. at 456–57. To maintain a tortious interference suit against a corporate agent or representative, a plaintiff must show that the agent acted willfully and intentionally to serve the agent’s personal interests at the corporation’s expense. Id. at 457. Even an agent’s mixed motives—benefitting himself and the corporation—are insufficient. Id. The Texas Supreme Court has provided guidance on determining when an corporation’s agent is acting against the corporation’s interests: If a corporation does not complain about it’s agents actions, then the agent cannot be held to have acted contrary to the corporation’s interests. Id. (citing Morgan Stanley & Co. v. Tex. Oil Co., 985 S.W.2d 178, 181–82 (Tex. 1997)). Mumfrey alleged the following acts on the part of the individual defendants: LeBlanc retaliated by telling Mumfrey he was going to “rip the chair out” of the pharmacy; LeBlanc wrote Mumfrey up and subjected him to long disciplinary meetings as retaliation for his having sought an accommodation; he was generally harassed by the individual defendants for seeking the

16 No. 12-40419

accommodation; he was subjected to snide and profane comments regarding his use of the chair; he was written up frequently; he was passed over for promotions; he was denied a managerial position; and his pay did not progress. Mumfrey’s pleadings do not allege that the individual defendants were acting to serve their own personal interests. Cf. In re Exxon, 558 F.3d at 386–87 (finding individual defendants improperly joined where the record could not support two required elements of the plaintiff’s state law claim). Mumfrey even admits in his brief that the “individual [d]efendants . . . were acting in the scope of their employment at the time of the retaliatory acts.” Further, CVS never complained or disciplined the individual defendants for their behavior, such that the individual defendants cannot be held to have acted contrary to CVS’s interests. Because CVS demonstrated that Mumfrey has no reasonable possibility for recovery against the individual defendants under Texas law, they were improperly joined. Therefore, there was complete diversity between the parties such that the district court had jurisdiction. III. DETERMINATION ON THE MERITS A. Standard of Review “The standard of review for a bench trial is well established: findings of fact are reviewed for clear error and legal issues are reviewed de novo.” Bd. of Trs. New Orleans Emp’rs Int’l Longshoremen’s Ass’n v. Gabriel, Roeder, Smith & Co., 529 F.3d 506, 509 (5th Cir. 2008) (internal quotation marks omitted) (citing Water Craft Mgmt. LLC v. Mercury Marine, 457 F.3d 484, 488 (5th Cir. 2006)). “A finding is clearly erroneous if it is without substantial evidence to support it, the court misinterpreted the effect of the evidence, or this court is convinced that the findings are against the preponderance of credible testimony.” Id. B. Discussion After the bench trial, the district court concluded that Mumfrey had not

17 No. 12-40419

proved retaliation by a preponderance of the evidence. Seemingly conceding the district court’s findings of fact, Mumfrey argues on appeal that the district court erred in its conclusions of law. As mentioned above, the district court’s legal conclusions are reviewed de novo. Id. Mumfrey did not meet his burden of establish, by a preponderance of the evidence, that CVS retaliated against him. Retaliation claims are evaluated under a burden-shifting framework. Medina v. Ramsey Steel Co., 238 F.3d 674, 684 (5th Cir. 2001) (citing Sherrod v. Am. Airlines, Inc., 132 F.3d 1112, 1122 (5th Cir. 1998)). Under the three-part framework, a plaintiff must first establish a prima facie case of retaliation. Id. If he can do so, the burden shifts to the defendant to produce a legitimate, non- retaliatory justification for its actions. Id. Finally, in the third stage of the burden-shifting framework, the plaintiff must establish that the defendant’s proffered reason is pretext for intentional retaliation. Id. Texas law prohibits retaliation by employers against employees who, among other things, file a complaint. Tex. Labor Code Ann. § 21.055 (West 2006). To ultimately prevail, the plaintiff must establish that without his protected activity, the employer would not have reached the same adverse employment decision. Wal–Mart Stores, Inc. v. Lane, 31 S.W.3d 282, 295 (Tex. App.— Corpus Christi 2000, pet. denied); Pineda v. United Parcel Serv., Inc., 360 F.3d 483, 487 (5th Cir. 2004). In essence, Mumfrey had to establish that but for filing his complaints, he would not have been fired when he was. CVS seems to concede the district court’s determination that Mumfrey established a prima facie case of retaliation. CVS asserted that it terminated Mumfrey for legitimate non-retaliatory reasons—namely refusing to verify a prescription while on final review. Mumfrey argues that CVS’s stated reason is pretext for retaliation based on Mumfrey’s recently filed EEOC complaint. Mumfrey relies on three main arguments to establish pretext. First, he argues that the final prescription incident was not a legitimate reason to terminate his

18 No. 12-40419

employment. Second, he argues that he was treated differently than other employees in similar situations, implying that his disparate treatment is evidence that CVS’s termination justification was pretext. Third, he relies on the temporal proximity between the filing of his EEOC complaint and his termination. The court addresses each in turn. First, Mumfrey argues it was appropriate for him, as a licensed professional, to exercise his best judgment in determining whether to immediately fill what he considered to be a suspicious prescription before he could verify it with the authorizing physician. Mumfrey presented evidence to the district court to support his position that he was justified in not filling the prescription immediately. CVS investigated the incident and presented the district court its evidence as well. After considering both, the district court made findings of fact. Mumfrey did not attempt to claim the prescription on the customer’s insurance, despite his assertion that he did so. Mumfrey told the customer he needed to verify the prescription with the doctor, but made no calls to the doctor and was not scheduled to work the next day. After the customer urged Mumfrey to confirm that he had filled the exact same prescription at that CVS before, he refused to do so. A videotape of the encounter confirms these omissions. Mumfrey argues that a fact-finder can infer pretext if the employer’s stated reason is false or not worthy of credence. The district court’s conclusions of law against Mumfrey affirm that, as the fact finder, the district court did not find CVS’s stated reasons false or not worthy of credence. We see no clear error in the district court’s findings of fact.15 This final failure, in addition to his previous final warnings for rudeness and failing to fill out required reports, is hardly evidence of a non-legitimate termination justification and does not

15 Again, Mumfrey also presumably conceded the court’s findings of fact by not challenging any specifically in his argument on appeal.

19 No. 12-40419

support to Mumfrey’s allegations of pretext. Further, the district court’s factual findings rely on both documentary and testimonial evidence. Based on the evidence, the district court found that Mumfrey’s failure to at least attempt to verify the prescription motivated CVS’s decision. This finding suggests that the parties’ dispute about whether Mumfrey’s stated suspicions were reasonable is beside the point. Even if Mumfrey had established that he was justified in his suspicion of the prescription, it would be immaterial; he was fired instead for not making an attempt to verify it. Thus, it is immaterial that Mumfrey and CVS employees had different opinions about whether the prescription, as presented, was suspicious. Mumfrey’s second argument for establishing pretext is his assertion that two other employees, Rodriguez and LeBlanc, were treated differently by CVS. This argument is without merit. For Mumfrey to establish disparate treatment, the situation and conduct of the employees must be “nearly identical.” AutoZone, Inc. v. Reyes, 272 S.W.3d 588, 594 (Tex. 2008). Employees with different responsibilities or disciplinary records are not “nearly identical.” Id.; see also Okoye v. Univ. of Tex. Hous. Health Sci. Ctr., 245 F.3d 507, 514–15 (5th Cir. 2001). As a pharmacy technician, Rodriguez had different responsibilities and thus was not similarly situated to Mumfrey.16 LeBlanc was not similarly situated because he had not failed to fill a prescription while on final warning, as Mumfrey had. Further, unlike Mumfrey, LeBlanc did not continue to give out his cash register password to other employees after being counseled not to, as Mumfrey did. Perhaps most obviously, LeBlanc and Mumfrey had different disciplinary records on file. Mumfrey argues that because he claimed retaliation, not discrimination, the district court relied on inapplicable authority to conclude that Mumfrey was

16 To the extent that Mumfrey’s disparate treatment argument is based on Rodriguez not being disciplined for mis-filling, it is baseless. Mumfrey was never disciplined for mis- filling, but for failing to file a report afterward, as required.

20 No. 12-40419

not “similarly situated.” However, courts use the same standard in both discrimination and retaliation cases. See, e.g., Bryant, 413 F.3d at 478 (“Disparate treatment of similarly situated employees is one way to demonstrate unlawful discrimination and retaliation.”); Rios v. Rossotti, 252 F.3d 375, 380 (5th Cir. 2001) (“The framework for analyzing a retaliation claim is the same as that used in the employment discrimination context.”). His second argument is thus without merit and has no weight when considering the ultimate merits of his pretext argument. Mumfrey’s third argument is that the temporal proximity between his complaints and his termination establishes that the termination was pretextual. The district court relied on clear Fifth Circuit precedent to hold that Mumfrey’s proximity evidence is insufficient to prove retaliation. See Strong v. Univ. Healthcare Sys., L.L.C., 482 F.3d 802, 808 (5th Cir. 2007) (“[W]e affirmatively reject the notion that temporal proximity standing alone can be sufficient proof of but for causation.”). Mumfrey takes issue with the district court’s application of Strong to his facts: while he admits Strong is good law, he contends he has produced other evidence of pretext to support his proximity evidence. Strong stands for the proposition that proximity plus other grounds can establish pretext. Here, even granting his proximity argument, neither Mumfrey’s termination justification argument, nor his disparate treatment arguments are sufficient to complete the equation. Additionally, although the timing appears suspect at first glance, when taken together with Mumfrey’s three October 2008 “conferences and counselings,” his December 2008 Final Warning, his January 2009 Final Warning, and his final prescription incident, the more likely cause for his termination was the cascade of recent discipline episodes rather than one specific infraction. Mumfrey’s proximity evidence, without other persuasive evidence of pretext is insufficient to demonstrate by a preponderance of the evidence that

21 No. 12-40419

CVS retaliated against Mumfrey. Therefore, Mumfrey’s retaliation claim fails. IV. CONCLUSION For the foregoing reasons, we AFFIRM.