Williams v. Jani-King of Philadelphia Inc, No. 15-2049 (3d Cir. 2016)

Annotate this Case
Justia Opinion Summary

Jani-King, the world’s largest commercial cleaning franchisor, classifies its franchisees as independent contractors. Its cleaning contracts are between Jani-King and the customer; the franchisee is not a party, but may elect to provide or not provide services under a contract. Jani-King exercises a significant amount of control over how franchisees operate and controls billing and accounting. Two Jani-King franchisees assert that they are misclassified and should be treated as employees. On behalf of a class of Jani-King franchisees in the Philadelphia area (approximately 300 franchisees), they sought unpaid wages under the Pennsylvania Wage Payment and Collection Law (WPCL), 43 Pa. Stat. 260.1–260.12. The Third Circuit affirmed certification of the class under Federal Rule of Civil Procedure 23(f). The misclassification claim can be made on a class-wide basis through common evidence, primarily the franchise agreement and manuals. Under Pennsylvania law, no special treatment is accorded to the franchise relationship. A franchisee may be an employee or an independent contractor depending on the nature of the franchise system controls.

Download PDF
PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ______ No. 15-2049 ______ DARRYL WILLIAMS; HOWARD BROOKS v. JANI-KING OF PHILADELPHIA INC.; JANI-KING INC.; JANI-KING INTERNATIONAL INC., Appellants ______ On Appeal from the United States District Court for the Eastern District of Pennsylvania (E.D. Pa. No. 2-09-cv-01738) District Judge: Honorable R. Barclay Surrick ______ Argued January 20, 2016 Before: FISHER, CHAGARES and COWEN, Circuit Judges. (Filed: September 21, 2016) Kerry L. Bundy, Esq. Eileen M. Hunter, Esq. Aaron D. Vanoort, Esq. [ARGUED] Faegre Baker Daniels 90 South 7th Street 2200 Wells Fargo Center Minneapolis, MN 55402 Counsel for Appellants Jonathan N. Solish, Esq. Bryan Cave 120 Broadway, Suite 300 Santa Monica, CA 90401 Counsel for Amicus Appellant International Franchise Association David J. Cohen, Esq. 604 Spruce Street Philadelphia, PA 19106 Shannon Liss-Riordan, Esq. [ARGUED] Adelaide Pagano, Esq. Lichten & Liss-Riordan 729 Boylston Street, Suite 2000 Boston, MA 02116 Counsel for Appellees ______ OPINION OF THE COURT ______ 2 FISHER, Circuit Judge. Disputes about whether workers are properly classified as employees or independent contractors are a classic and reoccurring issue in American law. This case presents such a dispute. Jani-King,1 the world’s largest commercial cleaning franchisor, classifies its franchisees as independent contractors. Two Jani-King franchisees, Darryl Williams and Howard Brooks, assert that they are misclassified and should be treated as employees. On behalf of a class of Jani-King franchisees in the Philadelphia area, Brooks and Williams seek unpaid wages under the Pennsylvania Wage Payment and Collection Law (WPCL), 43 Pa. Stat. §§ 260.1–260.12. The District Court granted the Plaintiffs’ motion for class certification. In this interlocutory appeal under Federal Rule of Civil Procedure 23(f ), we consider whether the misclassification claim can be made on a class-wide basis through common evidence, primarily the franchise agreement and manuals. We hold that the claims in this case are susceptible to class-wide determination and that the District Court did not abuse its discretion by certifying the class. 1. The defendants are three corporate entities: Jani-King of Philadelphia, Inc.; Jani-King, Inc.; and Jani-King International, Inc. The parties refer to these related corporations as “Jani-King,” and we adopt that convention in this opinion. 3 I A Jani-King franchisees provide janitorial and other cleaning services for offices, restaurants, warehouses, and other commercial properties. Jani-King licenses its trademarks, goodwill, and cleaning system to its franchisees and provides franchisees with administrative, billing, and advertising support. To obtain a franchise, an individual must pass a background check, pay Jani-King an initial franchise fee of between $14,625 and $142,750, and sign the Jani-King franchise agreement. (App. 108–10.) Jani-King requires new franchisees to meet several requirements before starting operations. A new franchisee must attend a 13-day training course and pass a test about Jani-King’s safety and training manual, which is more than 450 pages long. A franchisee must also purchase cleaning equipment and insurance, both of which are offered directly by Jani-King, although a franchisee may select alternative sources. A franchisee must secure any needed licenses and permits. Jani-King guarantees new franchisees a certain dollar amount in cleaning contracts for a set period. For a larger initial investment, Jani-King offers contracts with higher value and guarantees them for a longer period. Jani-King is responsible for obtaining new customers, and someone from Jani-King’s sales office will meet with prospective customers to determine their cleaning needs and give them a quote. If a customer agrees to a cleaning services contract, the contract is between the customer and Jani-King. The franchisee is not a party. Jani-King asks franchisees whether they want to provide services under the contract. A franchisee may accept or reject the contract. 4 Jani-King exercises a significant amount of control over how franchisees operate. The Jani-King policies and procedures manual dictates when and how frequently franchisees must communicate with customers, how franchisees must dress when meeting customers, and what uniforms must be worn while performing cleaning work. (App. 9.) Franchisees must be able to respond to any messages within four hours at all times and must notify JaniKing in advance of any vacations and delegate all business decision-making authority to someone else while on vacation. (App. 9.) Any marketing materials must be approved by the Jani-King regional office, and franchisees are not permitted to advertise their individual phone numbers or have individual websites. (App. 11.) Franchisees must make a monthly report to Jani-King of all services and supplies invoiced and must keep accurate books and records, which Jani-King may audit. (App. 9.) Jani-King requires franchisees to maintain sufficient working capital. (App. 9.) Jani-King requires franchisees to perform cleaning services adequately and may inspect the premises serviced by the franchisee at any time. (App. 9–10.) Customer complaints must be handled in a prescribed manner and within a certain time frame. Jani-King has numerous tools to ensure franchisees adhere to its requirements. Jani-King may charge a $50 complaint fee for failure to adequately address customer concerns. If the complaint is serious enough, the Jani-King regional office will address the problem and bill the franchisee for any response work it must do. (App. 10.) JaniKing may require franchisees to take remedial training. (App. 7.) Jani-King may reassign customer accounts for inadequate performance or failure to adhere to policies and procedures. (App. 10.) Ultimately, Jani-King may suspend any franchisee for failure to comply with its procedures and policies. 5 Jani-King invoices customers and controls billing and accounting. Jani-King subtracts all fees from the gross revenue and pays the remainder to the franchisee. The fees include a 10 percent royalty fee, a 2.5 percent accounting fee (although this fee is reduced for particularly large franchises), a 1.5 percent technology licensing fee, and a 1 percent advertising fee. (App. 111–12.) In addition, once the initial guaranteed business period expires, franchisees pay a finder’s fee to Jani-King for all new customer accounts. Franchisees may solicit new business within certain parameters, but any new cleaning contract is between the customer and Jani-King, which then has sole control over the contract. (App. 7, 11.) Franchisees have control over certain aspects of their business. While some franchisees do all cleaning work personally, others hire employees. Jani-King requires franchisees to keep certain employee documents and records, but franchisees otherwise have total control over hiring and firing employees. There are approximately 300 Jani-King franchises in the Philadelphia area. The named plaintiffs, Darryl Williams and Howard Brooks, purchased Jani-King franchises in the Philadelphia area. Williams’s and Brooks’s franchises were small. Brooks never hired any employees and performed cleaning services for his franchise himself, with occasional help from his wife or friends. (App. 553.) Williams services one Jani-King account and performs the cleaning himself, although he paid an employee to help him for two months at one point. (App. 653.) Franchisees have a wide range of business sizes—some have large businesses and many employees, and some have small businesses and no employees. For example, franchisee Charles Jones has 27 employees, including five supervisors, 6 monthly gross revenue of $43,497.39 in February 2013, and total gross revenue since 2009 of $1.28 million. At the opposite end of the spectrum, Kadri Memedoski has no employees, $4,556.44 in monthly revenue for February 2013, and $500,000 total gross revenue since 2006. Jani-King presented other examples of franchisees with many employees, including Sulejuman Smanovski with 35 and Althea Lanier with 16. (App. 701–21.) B Williams and Brooks filed suit on behalf of a class of about 300 Jani-King franchisees in the Philadelphia area in state court in 2009, asserting that Jani-King violated the WPCL. Jani-King removed the case to the District Court for the Eastern District of Pennsylvania and moved to dismiss. For reasons unclear from the record, it took the District Court three years to rule on the motion to dismiss, which it eventually granted in part and denied in part in December 2012. The Plaintiffs filed a motion for class certification, which the District Court granted. Jani-King petitioned our Court for leave to appeal under Federal Rule of Civil Procedure 23(f ), and we granted permission for this appeal. The District Court stayed the case pending the outcome of this appeal. 2 2. Brooks was substituted for one of the original plaintiffs in 2013. 7 II3 Class certification is appropriate when the prerequisites of Federal Rule of Civil Procedure 23 are met.4 The party 3. The District Court had jurisdiction under 28 U.S.C. § 1332(d)(2), and we have jurisdiction under 28 U.S.C. § 1292(e) and Federal Rule of Civil Procedure 23(f ). 4. Federal Rule of Civil Procedure 23(a) sets forth four prerequisites for a class action: “(1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) the representative parties will fairly and adequately protect the interests of the class.” Rule 23(b) organizes class actions into three types, of which only one is relevant in this case: (3) the court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. The matters pertinent to these findings include: 8 seeking certification must prove, by a preponderance of the evidence, that each of the four conditions of Rule 23(a)— numerosity, commonality, typicality, and adequacy—is met and that at least one of the provisions of Rule 23(b) is satisfied. Comcast Corp. v. Behrend, 133 S. Ct. 1426, 1432 (2013). We review a district court’s class certification order for abuse of discretion. In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 312 (3d Cir. 2008). A district court abuses its discretion if its decision “rests upon a clearly erroneous finding of fact, an errant conclusion of law or an improper application of law to fact.” Id. (quoting Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 165 (3d Cir. 2001)). (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties managing a class action. 9 in The District Court found that each of the Rule 23(a) factors was met. Jani-King did not dispute numerosity before the District Court. It did argue to the District Court that the Plaintiffs failed to establish typicality and adequacy. The District Court explained at length why typicality and adequacy were satisfied (App. 12–20), and Jani-King did not challenge these rulings on appeal. Nor did Jani-King challenge the District Court’s conclusion that the superiority requirement of Rule 23(b)(3) was met. The issue on appeal is whether the Plaintiffs’ claims are capable of class-wide resolution. This involves the commonality requirement of Rule 23(a)(2) and the predominance requirement of Rule 23(b)(3). Commonality requires that there be common issues of law or fact. Fed. R. Civ. P. 23(a)(2). To meet the predominance requirement, these common issues of law or fact must predominate over issues affecting individual class members. Comcast, 133 S. Ct. at 1432. This is a challenging standard to meet that requires the district court to undertake a “rigorous analysis.” Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011); see Comcast, 133 S. Ct. at 1432 (“If anything, Rule 23(b)(3)’s predominance criterion is even more demanding than Rule 23(a).”). Predominance turns on the “nature of the evidence” and whether “proof of the essential elements of the cause of action requires individual treatment.” Hydrogen Peroxide, 552 F.3d at 311 (internal quotation marks omitted). Because of this consideration, addressing predominance “[f]requently . . . entail[s] some overlap with the merits of the plaintiff’s underlying claim.” Wal-Mart, 131 S. Ct. at 2551. 10 III A We begin our analysis by considering the evidence needed to prove the Plaintiffs’ misclassification claim under the WPCL. The WPCL requires employers, among other things, to pay to employees wages and agreed-upon fringe benefits in a regularly scheduled manner and by lawful money or check and to make only lawful deductions from employees’ pay. 43 Pa. Stat. §§ 260.3, 260.4. The WPCL gives employees the right to institute a civil action to recover wages owed under the statute. Id. § 260.9a(a). The dispute in this case is whether the Plaintiffs are employees of Jani-King eligible to bring a suit under the WPCL or independent contractors not covered by the WPCL. The WPCL does not define “employee.” Morin v. Brassington, 871 A.2d 844, 849 (Pa. Super. Ct. 2005). In interpreting the meaning of employee under the WPCL, Pennsylvania courts have looked to similar statutes such as the Pennsylvania Unemployment Compensation Act and the Pennsylvania Workers’ Compensation Act. Id. at 849–50 (applying the Workers’ Compensation Act and related case law to the WPCL); Frank Burns, Inc. v. Interdigital Commc’ns Corp., 704 A.2d 678, 680 (Pa. Super Ct. 1997) (applying the definitions in the Unemployment Compensation Act and Workers’ Compensation act to the WPCL).5 5. In the absence of a definitive pronouncement by the Pennsylvania Supreme Court about the definition of “employee” under the WPCL, we must predict how that court would decide the issue. In doing so, “we must look to decisions of state intermediate appellate courts, 11 Pennsylvania courts interpreting the Unemployment Compensation Act and the Workers’ Compensation Act have applied a multifactor test to determine whether a worker is an employee or independent contractor. Courts interpreting the WPCL have adopted the same multifactor test. The factors include: the control of the manner that work is to be done; responsibility for result only; terms of agreement between the parties; the nature of the work or occupation; the skill required for performance; whether one employed is engaged in a distinct occupation or business; which party supplies the tools; whether payment is by the time or by the job; whether the work is part of the regular business of the employer, and the right to terminate the employment at any time. of federal courts interpreting that state’s law, and of other state supreme courts that have addressed the issue, as well as to analogous decisions, considered dicta, scholarly works, and any other reliable data tending convincingly to show how the highest court in the state would decide the issue at hand.” Spence v. ESAB Grp., Inc., 623 F.3d 212, 216–17 (3d Cir. 2010) (internal quotation marks omitted). We find the method of analysis used by the Pennsylvania Superior Court in Morin and Frank Burns—comparison to similar statutes—persuasive. 12 Morin, 871 A.2d at 850 (quoting Lynch v. Workmen’s Comp. Appeal Bd., 554 A.2d 159, 160 (Pa. Commw. Ct. 1989)). The applicability of this multifactor test to the WPCL is bolstered by the use of the same factors by the Pennsylvania Supreme Court to distinguish between employees and independent contractors in the context of vicarious liability. Hammermill Paper Co. v. Rust Eng’g Co., 243 A.2d 389, 392 (Pa. 1968). Although no factor is dispositive, the “paramount” factor is the right to control the manner in which the work is accomplished. Morin, 871 A.2d at 850. The Pennsylvania Supreme Court has also noted, in the context of workers’ compensation, that “control over the work to be completed and the manner in which it is to be performed are the primary factors in determining employee status.” Universal Am-Can, Ltd. v. Workers’ Comp. Appeal Bd., 762 A.2d 328, 333 (Pa. 2000) (calling these factors the “dominant considerations”); see Lynch, 554 A.2d at 160 (“[T]he right to control is the most persuasive indication of [employee or independent contractor status].”) Jani-King argues that actual control, not the right to control, is the key factor in the test. Jani-King criticizes a number of Pennsylvania lower-court decisions, including Lynch, and argues that they are mistaken or rely on inapposite borrowed-employee cases. But the Pennsylvania Supreme Court has held that the right to control is more significant than actual control. Universal Am-Can, 762 A.2d at 333 (“[I]t is the existence of the right to control that is significant, irrespective of whether the control is actually exercised.”). And we have also recognized the importance of the right to control. Drexel v. Union Prescription Ctrs., Inc., 582 F.2d 781, 785 (3d Cir. 1978) (“Actual control of the manner of work is not essential; rather, it is the right to control which is determinative.”). According to Jani-King, Universal Am-Can, 13 an employee–independent contractor case, relied on borrowed-employee cases for this conclusion. But to the extent the Pennsylvania case law is premised on a “mistake,” that mistake is longstanding and accepted by the Pennsylvania Supreme Court. It is not our place to “correct” such a mistake. Pennsylvania courts, including the Pennsylvania Supreme Court, apply the multifactor test for distinguishing between employee and independent contractor status in many different contexts. Based on this authority, we predict that the Pennsylvania Supreme Court would employ this test in the context of the WPCL. We also predict that the right to control, rather than actual control, is the most important of the factors. B We turn next to the more difficult question of whether, using the multifactor test, the employment status question can be resolved in this case through evidence common to the class. The common evidence identified by the Plaintiffs and the District Court are the Jani-King franchise agreement, policies manual, and training manual, and representative testimony about those documents. The District Court concluded that the Plaintiffs’ claims could be proven through this common evidence and that, therefore, the Plaintiffs met the predominance requirement. The District Court accurately summarized the controls placed on franchisees by the franchise agreement and manuals: Plaintiffs have pointed to specific provisions in the Franchise Agreement, the Policies Manual, and the Training Manual (collectively “Jani-King Documents”) to show that Jani-King 14 has the ability to control the manner in which franchisees perform their day-today tasks. All the proposed class members are bound by the Jani-King Documents, which include mandates regarding how often the franchisee must communicate with customers, how franchisees must address customer complaints, where franchisees can solicit business, what franchisees must wear, what types of records the franchisee must keep, how the franchisee can advertise, how far in advance franchisees must inform the franchisor of vacations, and how quickly the franchisee must be able to be reached. In addition, Jani-King controls the franchisees’ work assignments, has the right to inspect the franchisees work, and has the ability to change the policies and procedures as it sees fit. (App. 29–30.) The Jani-King franchise agreement, policies manual, and training manual are common to the class—they apply to the franchisee who has no employees and services a low-value contract and to the franchisee with dozens of employees and many cleaning contracts. These documents describe the level of Jani-King’s right to control its franchisees. They also address many of the secondary factors considered in Pennsylvania decisions—the terms of agreement, the nature of the work, the skill required, who supplies the tools, whether payment is by time or by job, and the right to 15 terminate at any time. See Morin, 871 A.2d at 850. We find no legal error in the District Court’s conclusion that these documents “could be read” to give Jani-King the right to control its franchisees. (App. 30.) We go no further toward resolving the merits of the WPCL claim. Although the court must undertake a rigorous analysis at the certification stage and consider some meritsrelated issues, the class certification stage is not the place for a decision on the merits. Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184, 1194–95 (2013) (“Rule 23 grants courts no license to engage in free-ranging merits inquiries at the certification stage. Merits questions may be considered to the extent—but only to the extent—that they are relevant to determining whether the Rule 23 prerequisites for class certification are satisfied.”). Jani-King asks us to weigh in on the merits of the Plaintiffs’ claim now. Jani-King’s first issue on appeal is: “Do the system controls inherent in a franchise relationship make a franchisor the employer of its franchise owners under Pennsylvania’s multi-factor employment test?” (Jani-King Br. 1.) This appeal is not the proper place for us to answer this question, and we decline to do so. It is enough for us to determine that some franchise agreements may contain sufficient controls to render the relationship one of employment and that the common documents in this case contain the types of evidence that courts and juries use to make that determination under Pennsylvania law. Jani-King may ultimately be correct that the franchise agreement and manual do not contain sufficient controls over the day-to-day work of its franchisees to make them employees under Pennsylvania law, and we express no opinion on that matter here. Either way, it is possible to make 16 the determination on a class-wide basis. If Jani-King is correct and the documents cannot, as a matter of fact, establish that the franchisees are employees, Jani-King will prevail class wide. C Jani-King makes two primary arguments for why the franchise agreement and manual are insufficient to resolve the WPCL claim. One, Jani-King argues that, as a matter of law, written agreements alone are insufficient to determine employment status. And, two, Jani-King argues that franchises are inherently different from other types of business relationships and that franchise system controls should be categorically excluded from consideration when determining whether an employment relationship exists. Neither of these arguments persuades us that certifying the class was an abuse of discretion. 1 Jani-King asserts that the test for employee status under Pennsylvania law is not susceptible to proof through common evidence. This is so, according to Jani-King, because “the entire employment relationship must be examined in determining whether it is an employment relationship.” C E Credits OnLine v. Unemployment Comp. Bd. of Review, 946 A.2d 1162, 1168 (Pa. Commw. Ct. 2008). Jani-King notes that “courts have repeatedly held that examining the written terms of an agreement alone is not sufficient to determine employment status.” (Jani-King Br. 19.) Therefore, Jani-King argues, the district court erred as a matter of law “[b]y relying on the terms of Jani-King’s common franchise documents, to the exclusion of individual evidence of the actual relationships between Jani-King and” franchisees. (Id. at 20.) 17 We are not convinced that the terms of a written agreement alone are never sufficient to determine employment status. Jani-King cited Urbano v. STAT Courier, Inc., 878 A.2d 58 (Pa. Super. Ct. 2005), a class action WPCL case in which the plaintiff delivery drivers argued they were misclassified. In that case, the Superior Court stated that, “[w]hile [an] agreement [is] relevant when identifying whether an employee/employer relationship exists, it is just one of the criteria to be utilized.” Id. at 62. The trial court had granted judgment on the pleadings, finding that the plaintiffs were independent contractors as a matter of law because the driver agreement repeatedly identified drivers as independent contractors. But the Superior Court reversed because there were disputed facts that the alleged employer exercised control to a greater degree than provided by the agreement. Id. In this case, there is no assertion that the agreement and manuals do not control. We are equally unpersuaded by two other cases relied upon by Jani-King, Jones v. Century Oil U.S.A., Inc., 957 F.2d 84 (3d Cir. 1992), and Kurbatov v. Department of Labor & Industry, 29 A.3d 66 (Pa. Super. Ct. 2011). Jani-King cited Jones for its statement that “it is the actual practice between the parties that is crucial.” Jones, 957 F.2d at 87 (quoting George v. Nemeth, 233 A.2d 231, 233 (Pa. 1967)). Kurbatov similarly held that it is the “actual working relationship between worker and employer” that is determinative. 29 A.3d at 70. But in both these cases, the courts were responding to an argument by the defendant that the worker was an independent contractor because the agreement identified the worker as an independent contractor. Jones, 957 F.2d at 87; Kurbatov, 29 A.3d at 70. These courts were contrasting the “actual practice” and “actual working relationship” against the labels used in an agreement. It is these labels that are not 18 determinative. Jones, 957 F.2d at 86 (“[T]he determining factor is not the way in which plaintiffs or defendant regards this relationship but ‘what it really was under the facts and applicable rules of law.’” (quoting Feller v. New Amsterdam Cas. Co., 70 A.2d 299, 302 (Pa. 1950))). The provisions of an agreement may be evidence of what the actual practice or working relationship is. In Kurbatov, for example, the Superior Court affirmed an order of the Pennsylvania Department of Labor and Industry, which had determined that an employment relationship existed. The Department of Labor and Industry relied on the agreement, even though the employer’s testimony conflicted with the agreement. The Department of Labor and Industry was entitled to weigh the evidence and rely on the documentary evidence to the exclusion of testimony. Kurbatov, 29 A.3d at 72. Compare this to Jones, where we vacated a directed verdict because there was testimony that the right to control was “not fully reflected in the written agreements.” Jones, 957 F.2d at 88–89 & n.3. The three decisions cited by Jani-King contrast with Green v. Independent Oil Co., 201 A.2d 207 (Pa. 1964). Green was a vicarious liability case: The defendant company argued it was not responsible for the negligence of a worker whom the company alleged was an independent contractor. The “sole evidence of the relationship” between the worker and the company was the agreement. Id. at 211. Since the terms of the agreement were not in dispute, the determination of the relationship was a question of law for the court. Id. “An examination of the . . . agreement in its entirety indicate[d] clearly and convincingly that the relationship . . . was that of independent contractee-contractor, not employer-employee.” Id. Under certain circumstances, therefore, an employment 19 relationship—or the lack of one—can be determined by examining the documents alone. Because documentary evidence can be sufficient to resolve the multifactor employment status test, it was not an error of law for the District Court to rely on this evidence in ruling on class certification. 2 Jani-King and the amicus International Franchise Association stress that franchising is an important and beneficial way of conducting business that is fundamentally different from other situations involving misclassification claims. They both assert that an adverse decision “directly threatens the viability of franchising in Pennsylvania.” (JaniKing Br. 2; see Amicus Br. 23.) Jani-King argues that the system controls inherent in franchising should be irrelevant when considering whether an alleged employer has the right to exercise day-to-day control. Some case law lends credence to this position. For example, we have noted that “[s]ome degree of control by the franchisor over the franchisee would appear to be inherent in the franchise relationship and may even be mandated by federal [trademark] law.” Drexel, 582 F.2d at 786 (citation omitted). In Myszkowski v. Penn Stroud Hotel, Inc., 634 A.2d 622, 627 (Pa. Super. Ct. 1993), for example, the Pennsylvania Superior Court held that the franchise system controls imposed by Best Western “addresse[d] the result of the work and not the manner in which it is conducted.” Jani-King also cites cases from other jurisdictions, including a California case involving Jani-King, Juarez v. Jani-King of California, Inc., 273 F.R.D. 571 (N.D. Cal. 2011). In that case, the district court found “it likely that under California law, a franchisee must show that the 20 franchisor exercised ‘control beyond that necessary to protect and maintain its interest in its trademark, trade name and goodwill’ to establish a prima facie case of an employeremployee relationship.” Id. at 583 (quoting Cislaw v. Southland Corp., 6 Cal. Rptr. 2d 386, 394 (Ct. App. 1992)). Once the district court “set[] aside the policies required to protect Jani-King’s service mark and goodwill, [it found] very little—if any—common evidence tending to prove an employer-employee relationship between Jani-King and its franchisees.” Id. In the instant case, the District Court discounted Juarez because Pennsylvania law “does not distinguish between controls put in place to protect a franchise’s goodwill and intellectual property and controls for other purposes.” (App. 29.) We have found no Pennsylvania case holding otherwise. The District Court did not err by discounting Juarez. The Pennsylvania cases cited by Jani-King do not stand for the proposition that franchise system controls are somehow categorically excluded from consideration in the employee–independent contractor analysis. As we succinctly stated in Drexel, the mere existence of a franchise relationship does not necessarily trigger a master-servant relationship, nor does it automatically insulate the parties from such a relationship. Whether the control retained by the franchisor is also sufficient to establish a masterservant relationship depends in each case upon the nature and extent of such control as defined in the franchise agreement or by the actual practice of the parties. 21 Drexel, 582 F.2d at 786. Under Pennsylvania law, no special treatment is accorded to the franchise relationship. A franchisee may be an employee or an independent contractor depending on the nature of the franchise system controls. IV The Plaintiffs’ WPCL misclassification claim can be resolved by the evidence that is common to the class. We find no clearly erroneous finding of fact or errant conclusion of law in the District Court’s judgment. We therefore conclude that the District Court did not abuse its discretion by certifying the class. We will affirm the District Court’s class certification order. 22 Williams, et al. v. Jani-King of Philadelphia, Inc., et al., No. 15-2049, dissenting. COWEN, Circuit Judge. Franchising constitutes “a bedrock of the American economy.” Queen City Pizza, Inc. v. Domino’s Pizza, Inc., 124 F.3d 430, 433 (3d Cir. 1997). Yet the majority’s opinion threatens the viability of this basic economic bedrock. I do not believe that such a result is consistent with either basic class action principles, the nature and importance of the franchisor-franchise relationship, or prior franchising case law. I predict that the Pennsylvania Supreme Court would ultimately hold that controls necessary to protect a franchisor’s trademark, trade name, and goodwill—in short, “franchise system controls”—are insufficient by themselves to establish the existence of an employer-employee relationship between the franchisor and its franchisees. In this case, the Plaintiffs’ purported common evidence merely sets forth various franchise system controls. Because of the absence of common evidence tending to prove that the franchisees are employees of the franchisor, the District Court abused its discretion by certifying a class of Jani-King franchisees. I therefore must respectfully dissent. Under Federal Rule of Civil Procedure 23(b)(3), a class action may be maintained if “the questions of law or fact common to class members predominate over any questions affecting only individual members.” The majority acknowledges that “[t]his is a challenging standard to meet that requires the district court to undertake ‘a rigorous analysis.’” (Maj. Op. at 10 (quoting Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551 (2011)).) In fact, predominance 1 depends on the nature of the evidence and whether proof of the elements of the cause of action requires individual treatment. See, e.g., In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 311-12 (3d Cir. 2008). “Because of this consideration, addressing predominance ‘[f]requently . . . entail[s] some overlap with the merits of the plaintiff’s claim.’ Wal-Mart, 132 S. Ct. at 2551.” (Id. at 11.) In short, we must “examine the elements of plaintiffs’ claim ‘through the prism’ of Rule 23 to determine whether the District Court properly certified the class.” Hydrogen Peroxide, 552 F.3d at 311 (quoting Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 181 (3d Cir. 2001)). The district court should envision the form a representative trial of the claim will take and then conduct a rigorous assessment of the available evidence as well as the methods that the plaintiff proposes to use to prove his or her claim. See, e.g., id. at 311-12; Sherman v. Am. Exp., Inc., Civil Action No. 09-575, 2012 WL 748400, at *8 (E.D. Pa. Mar. 8, 2012). However, the majority purportedly refuses to answer the basic question raised by Jani-King in its appeal—whether “franchise system controls” make a franchisor the employer of its franchisees under Pennsylvania’s multi-factor employment test. Given the rigorous obligations imposed by Rule 23, I do not see how we could avoid addressing this basic question. The predominance inquiry turns on what Pennsylvania law requires as evidence of employment status. In other words, could Plaintiffs really show that Jani-King’s franchisees are, in fact, its employees under Pennsylvania law using the Jani-King franchise agreement and other evidence common to the entire putative class? If, as a matter of state law, controls that are necessary to protect the franchisor’s own trademark, trade name, and goodwill are, by themselves, 2 not sufficient to make the franchisor the employer of its own franchisees—and the common evidence in the record merely implicates such “franchise system controls,” the Court then must determine that Plaintiffs cannot possibly show that ‘each element is capable of proof through evidence that is common to the class,” Hydrogen Peroxide, 552 F.3d at 311. In any event, the majority, despite its disclaimers, effectively answers this basic question in the affirmative. After all, it addresses at some length case law considering the employee-independent contractor distinction under Pennsylvania law, including two franchising decisions: Drexel v. Union Prescription Centers, Inc., 582 F.2d 781 (3d Cir. 1978), and Myszkowski v. Penn Stroud Hotel, Inc., 634 A.2d 622 (Pa. Super. Ct. 1993). It then upheld the District Court’s ruling insofar as it “discounted” a class certification opinion involving Jani-King (Maj. Op. at 21)—which held that “it is likely that under California law, a franchisee must show that the franchisor exercised ‘control beyond that necessary to protect and maintain its interest in its trademark, trade name and goodwill’ to establish a prima facie case of an employer-employee relationship,” Juarez v. Jani-King of Cal., Inc., 273 F.R.D. 571, 583 (N.D. Cal. 2011) (quoting Cislaw v. Southland Corp., 6 Cal. Rptr. 2d 386, 394 (Ct. App. 1992)). Rejecting Juarez, the majority instead held that, “[u]nder Pennsylvania law, no special treatment is accorded to the franchise relationship” and that “[a] franchisee may be an employee or an independent contractor depending on the nature of the franchise system controls.” (Id. at 22.) Obviously, this opinion will be cited as precedent (or, at the very least, as dicta) for the proposition that, despite the nature and importance of the franchisor-franchisee relationship, franchise system controls may by themselves give rise to an 3 employer-employee relationship under a variety of different statutory schemes (i.e., the WPCL, the Pennsylvania Unemployment Compensation Act, and the Pennsylvania Workers Compensation Act) as well as the doctrine of vicarious liability. I further note that Jani-King now has the burden of defending against a class action and that this opinion will most likely lead to additional class action litigation against other franchisors. It appears that the Pennsylvania Supreme Court has not expressly answered the specific question of whether or not franchise system controls are sufficient by themselves to establish the existence of an employer-employee relationship between the franchisor and the franchisees. Given the nature and importance of the franchisor-franchisee relationship, existing Pennsylvania case law like Drexel and Myszkowski, and franchising cases from other jurisdictions, I predict that the state supreme court would answer this question in the negative.1 1 Like the majority, I look, in the absence of state supreme court precedent directly on point, to “‘decisions of state intermediate appellate courts, of federal courts interpreting that state’s law, and of other state supreme courts that have addressed the issue,’ as well as to ‘analogous decisions, considered dicta, scholarly works, and any other reliable data tending convincingly to show how the highest court in the state would decide the issue at hand.’” Spence v. ESAB Grp., Inc., 623 F.3d 212, 216-17 (3d Cir. 2010) (quoting Norfolk So. Ry. v. Basell USA Inc., 512 F.3d 86, 92 (3d Cir. 2008)). 4 Franchise system controls constitute an essential aspect of the franchising mechanism. The Pennsylvania Supreme Court has recognized that “‘the franchise has evolved into an elaborate agreement by which the franchisee undertakes to conduct a business or sell a product or service in accordance with methods and procedures prescribed by the franchisor, and the franchisor undertakes to assist the franchisee through advertising, promotion and other advisory services.’” Atl. Richfield Co. v. Razumic, 390 A.2d 736, 740 (Pa. 1978) (quoting Piercing Pagoda, Inc. v. Hoffner, 351 A.2d 207, 211 (Pa. 1976)). The franchisor’s basic product is the “franchise itself.” Piercing Pagoda, 351 A.2d at 211 (citations omitted). Product uniformity and quality control attract customers and are critical to the success of both the franchisees as well as the franchisor. See, e.g., Queen City Pizza, 124 F.3d at 433; Piercing Pagoda, 124 F.3d at 211. In fact, a “trademark owner risks losing his rights” by failing to maintain adequate control over “the quality of his licensees’ products.” Ungar v. Dunkin’ Donuts of Am., Inc., 531 F.2d 1211, 1216 n.6 (3d Cir. 1976) (citing 15 U.S.C. § 1127). This franchisor-franchisee relationship thereby offers several important advantages to franchisors and franchisees alike. For example, the franchisee may benefit from existing inventory, training, and directions for market development, and, in addition, the creation and maintenance of goodwill through strict system controls should help to bring in more customers for the franchisee. See, e.g., id. at 1222-23; Piercing Pagoda, 351 A.2d at 211-12. Given these characteristics, it is not surprising that the franchising sector represents a major component of the economy. Nationally, this sector employs millions of people, 5 has payrolls in the billions, and generate trillions of dollars in total sales. See, e.g.¸ Patterson v. Domino’s Pizza, LLC, 333 P.3d 723, 733 (Cal. 2014). Amicus International Franchise Association estimates that, in 2007, Pennsylvania had 29,514 franchise establishments, with a total payroll of $10.7 billion and more than $82.4 billion in output. As the majority admits, we stated in Drexel that “‘[s]ome degree of control by the franchisor over the franchisee would appear to be inherent in the franchise relationship and may even be mandated by federal [trademark] law.’” (Maj. Op. at 20 (quoting Drexel, 582 F.2d at 786).) While the Drexel Court then ruled that the district court erred by granting summary judgment in favor of a drug store franchisor, we did so because, among other things, the agreement at issue was so broadly drafted “as to render uncertain the precise nature and scope of [the franchisor’s] rights vis-à-vis its franchisee.” Drexel, 582 F.2d at 788. We even assumed that the franchisor’s right to control various facets of the franchisee’s operations—ranging from the appearance and contents of the franchisee’s store to the color of its delivery trucks—implicated the franchisor’s interest in the result of the work as opposed to the manner in which the franchisee operated. Id. After all, “‘[t]he hallmark of an employee-employer relationship is that the employer not only controls the result of the work but has the right to direct the manner in which the work shall be accomplished; the hallmark of an independent contractee-contractor relationship is that the person engaged in the work has the exclusive control of the manner of performing it, being responsible only for the result.’” Id. at 785 (quoting Green v. Indep. Oil Co., 201 A.2d 207, 210 (Pa. 1964)). 6 More recently, the Pennsylvania Superior Court determined that summary judgment was properly granted in favor of a hotel marketing organization sued by an individual who had been sexually assaulted by a third party in one of its affiliated hotels. Myszkowski, 634 A.2d at 623-30. According to Myszkowski, “the owners of [the hotel] managed the day-to-day operations of the business and made all the decisions incidental to this operation.” Id. at 626-27. Expressly distinguishing Drexel on the grounds that the franchise agreement was so broadly written as to give the drug store franchisor the power to impose virtually anything it desired, id. at 628 n.6, the state appellate court concluded that “the fact that [the organization] sets certain standards in order to maintain a uniform quality of inn service only addresses the result of the work and not the manner in which it is conducted.” Id. at 627; see also, e.g., Smith v. Exxon Corp., 647 A.2d 577, 582 (Pa. Super. Ct. 1994) (“As in Myszkowski, here, the standards were implemented to maintain a uniform quality of service.”). In reaching this conclusion, it thereby took into account the nature of the franchisor-franchisee relationship, indicating that a franchise represents “‘a uniform system of inn service—that carried with it an obligation to maintain certain standards prescribed by the seller.’” Myszkowski, 634 A.2d at 627-28 (quoting Schear v, Motel Mgmt. Corp. of Am., 487 A.2d 1240, 1249 (Md. 1985)); see also, e.g., id. at 628 (“‘The general purpose of the contract is the maintenance of uniform service within, and public good toward, the Ramada Inn system.’” (quoting Hayman v. Ramada Inn, 357 S.E.2d 394, 397 (N.C. Ct. App. 1987))). In fact, Myszkowski relied on a similar ruling by the Pennsylvania Supreme Court. In Green v. Independent Oil 7 Co., 201 A.2d 207 (Pa. 1964), an oil company was named as a defendant in litigation arising out of a deadly explosion at one of its dealer stations, id. at 208-09. According to the state supreme court, the trial court erred in submitting the question of whether there was an employer-employee relationship between the oil company and the dealer to the jury. Id. at 210-11. As the Pennsylvania Superior Court aptly explained, several considerations were “[s]ignificant” to the outcome in Green: (1) the agreement between the parties specifically disclaimed the existence of any agency relationship; (2) all profits went to the dealer; (3) the sales tax permits and electric bills were in the dealer’s name; (4) the dealer hired, fired, and paid his own employees; (5) all monies were kept by the dealer in the dealer’s personal bank account; and (6) the dealer purchased the oil company’s products. Myszkowski, 634 A.2d at 626 (citing Green, 201 A.2d at 210) (footnote omitted). Accordingly, “[s]ome case law lend credence” to the line of reasoning offered by Jani King (and the amicus). (Maj. Op. at 20.) As I have already noted, the Juarez court recently held that, under California law, a franchisee must show that the franchisor (Jani-King itself) exercised “‘control beyond that necessary to protect and maintain its interest in its trademark, trade name and goodwill.’” Juarez, 273 F.R.D. at 583 (quoting Cislaw, 6 Cal. Rptr. 2d at 394). “As such, the Court can safely exclude from the employee-employer relationship analysis facts that merely show the common hallmarks of a franchise.’” Id. Subsequently, the California Supreme Court essentially adopted the Juarez approach. It concluded that a franchisor will be held vicariously liable “only if it has retained or assumed a general right of control over factors such as hiring, direction, discipline, discharge, 8 and relevant day-to-day aspects of the workplace behavior of the franchisee’s employees.” Patterson, 333 P.3d at 739 (footnote omitted). The Maine Supreme Judicial Court and the Wisconsin Supreme Court have reached similar conclusions. See Rainey v. Langen, 998 A.2d 342, 347 (Me. 2010) (“[C]ourts commonly distinguish between control over a franchisee’s day-to-day operations and ‘controls designed primarily to insure “uniformity and the standardization of products and services.”’” (citation omitted))); Kerl v. Dennis Rasmussen, Inc., 682 N.W.2d 328, 332 (Wis. 2004) (concluding that “the marketing, quality and operational standards commonly found in franchise agreements are insufficient to demonstrate the existence of a master/servant relationship for all purposes or as a general matter”). Under these circumstances, the District Court was wrong to conclude that Pennsylvania law “does not distinguish between controls put in place to protect a franchise’s goodwill and intellectual property and controls for other purposes.” Myers v. Jani-King of Philadelphia, Inc., Civil Action No. 09-1738, 2015 WL 1055700, at *14 (E.D. Pa. Mar. 11, 2015). Neither the District Court nor the majority cites to any case law expressly refusing to draw such a distinction. On the contrary, the existing Pennsylvania case law weighs in favor of distinguishing controls “inherent in the franchise relationship,” Drexel, 582 F.2d at 786 (citation omitted). After all, such a distinction is consistent with the fundamental and well-established notion that, while an employer has the right to direct the manner of performance, an independent contractee merely has the right to control the result of the work. See, e.g., Rainey, 998 A.2d at 349 (“The traditional test allows a franchisor to regulate the uniformity and the standardization of products and services without 9 risking the imposition of vicarious liability. If a franchisor takes further measures to reserve control over a franchisee’s performance of its day-to-day operations, however, the franchisor is no longer merely protecting its mark, and imposing vicarious liability may be appropriate.’” (citing, inter alia, Drexel, 582 F.2d at 786)); Myszkowski, 634 A.2d at 627 (explaining that fact hotel marketing organization set certain standards to maintain uniform service quality implicated “the result of the work and not the manner in which it is conducted”). I further note that California—like Pennsylvania—treats the right to control as the most important factor in a multi-factor employment inquiry.2 See, e.g., Juarez, 273 F.R.D. at 581. I ultimately conclude that Pennsylvania’s highest court would not allow “the very thing that defines [franchising]— 2 I do question whether, in these circumstances, the existence of an employment relationship can really be determined based solely on documentary evidence. After all, Pennsylvania law mandates a multi-factor test requiring the courts to examine “the entire employment relationship.” C E Credits v. Unemployment Comp. Bd. of Review, 946 A.2d 1162, 1168 (Pa. Commw. Ct. 2008) (citing Beacon Flag Car Co. v. Unemployment Comp. Bd. of Review, 910 A.2d 103, 108 (Pa. Commw. Ct. 2006)). “[N]o one factor is dispositive of one’s status and . . . each case must be determined on its own facts.” Zimmerman v. Commonwealth, 522 A.2d 43, 45 (Pa. 1987). Even if it is typically the alleged employee who offers evidence of actual control or performance, I fail to see why such evidence may not also weigh in favor of the putative employer in certain circumstances under what is a rather open-ended inquiry. 10 the ‘uniformity of product and control of its quality and distribution’”—to be used to put at risk this critical and generally beneficial sector of our economy (Appellants’ Reply Brief at 1-2 (quoting Atl. Richfield, 390 A.2d at 740)). See, e.g., Patterson, 333 P.3d at 726 (“Analysis of the franchise relationship for vicarious liability purposes must accommodate these contemporary realities. The imposition and enforcement of a uniform marketing and operational plan cannot automatically saddle the franchisor with responsibility for employees of the franchisee who injure each other on the job.”). In turn, the approach I set out here does not grant franchisors any sort of immunity from either possible vicarious liability or generally applicable employment laws like the WPCL. Unlike Jani-King, I do not believe that franchise system controls are simply irrelevant to the employment inquiry. Instead, such controls are insufficient by themselves to establish the existence of an employeremployee relationship. The inquiry thereby must go beyond the mere use of labels like “franchisor” and “franchisee” and assess whether the controls at issue exceed what is necessary to protect a franchisor’s trademark, trade name, or goodwill. See, e.g., Drexel, 582 F.2d at 786 (“[T]he mere existence of a franchise relationship does not necessarily trigger a masterservant relationship, nor does it automatically insulate the parties from such a relationship. Whether the control retained by the franchisor is also sufficient to establish a masterservant relationship depends in each case upon the nature and extent of such control as defined in the franchise agreement or by the actual practice of the parties.” (citations omitted)). Applying this approach, the next step is to consider the purported common evidence offered by Plaintiffs and identify the various policies that constitute controls instituted as 11 necessary to protect the franchisor’s own trademark, trade name, and goodwill. Jani-King provides an especially thorough—and persuasive—explanation for why “each of the contractual provisions and policies identified by the district court is simply an example of a common franchise system control, not a manifestation of the type of day-to-day supervisory control that indicates an employment relationship.” (Appellants’ Brief at 38.) To give just one example, it appropriately notes that, while the District Court pointed to Jani-King’s right to dictate what the franchisees must wear on the job, such uniform requirements are generally considered necessary to protect the goodwill associated with the franchise brand. See, e.g., Kerl, 682 N.W.2d at 341 (“The agreement specifies standards regarding containers, uniforms, paper goods, and other packaging supplies.”). Admittedly, Jani-King does appear to possess a great deal of power, especially with respect to customer negotiations, account assignments, and billing. Specifically, the franchisor negotiates and contracts with a prospective customer and then offers this contract to the respective franchisee (which is not a contractual party itself). Jani-King likewise invoices customers and controls billing and accounting. In contrast, it was the hotel in Myszkowski (and not the marketing organization) that set its own prices. Myszkowski, 534 A.2d at 627. Nevertheless, “[a] franchisee may accept or reject the contract” (Maj. Op. at 5), and they “may solicit new business within certain parameters” (id. at 6). See, e.g., Applied Measurement Prof’ls, Inc. v. Unemployment Comp. Bd. of Review, 844 A.2d 632, 636 (Pa. Commw. Ct. 2004) (finding that worker was independent contractor because, inter alia, she was “free to accept or reject an assignment”). While Jani-King sets certain quality standards, it appears that the franchisor does not specify the 12 specific methods the franchisees must utilize to perform the cleaning work. Finally, although Jani-King requires its franchisees to keep certain employment records, “franchisees otherwise have total control over hiring and firing employees” (id. at 7). See, e.g., Myszkowski, 634 A.2d at 627 (noting that hotel hired, fired, paid, and supervised its own employees). “Once [we] set[ ] aside the policies required to protect Jani-King’s service mark and goodwill, [there appears to be] very little—if any—common evidence tending to prove an employer-employee relationship between Jani-King and its franchisees.” Juarez, 372 F.R.D. at 583. In fact, the record is replete with individual evidence—franchisees vary widely in size ranging from small operations consisting of merely the individual franchisee to million-dollar enterprises with multiple employees. Accordingly, the District Court clearly committed reversible error by finding that common questions of law and fact predominate over individual ones. For the foregoing reasons, I would vacate the class certification order and remand for further proceedings on an individual (non-class action) basis. 13