Ralph Johnson v. The National Collegiate Athletic Association

108 F.4th 163

• Court: Court of Appeals for the Third Circuit
• Decided: July 11, 2024
• Docket: 22-1223
• Status: Published

Opinion

PRECEDENTIAL

UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT _______________

No. 22-1223 _______________

RALPH TREY JOHNSON; CLAUDIA RUIZ; JACOB WILLEBEEK-LEMAIR; NICHOLAS LABELLA; ALEXA COOKE, STEPHANIE KERKELES; RHESA FOSTER; ESTEBAN SUAREZ; ZACHARY HARRIS; LAURA HAMILTON; MATTHEW SCHMIDT; LIAM WALSH; GINA SNYDER; TAMARA SCHOEN, Individually and on Behalf of All Persons Similarly Situated

v.

NATIONAL COLLEGIATE ATHLETIC ASSOCIATION, also known as the NCAA, and the Following NCAA Division I Member Schools as Representatives of a Defendant Class of All Private and Semi-Public NCAA Division I Member Schedules; DREXEL UNIVERSITY; LAFAYETTE COLLEGE; VILLANOVA UNIVERSITY; UNIVERSITY OF PENNSYLVANIA; CORNELL UNIVERSITY; SACRED HEART UNIVERSITY; FORDHAM UNIVERSITY; UNIVERSITY OF OREGON; TULANE UNIVERSITY; UNIVERSITY OF ARIZONA; PURDUE UNIVERSITY; DUKE UNIVERSITY; MARIST COLLEGE

National Collegiate Athletic Association; Cornell University; Fordham University; Lafayette College; Sacred Heart University; Villanova University, Appellants _______________

On Appeal from the United States District Court for the Eastern District of Pennsylvania (D.C. No. 2-19-cv-05230) District Judge: Honorable John R. Padova _______________

Argued: February 15, 2023

Before: RESTREPO, PORTER, and McKEE, Circuit Judges

(Filed: July 11, 2024)

Steven B. Katz [ARGUED] CONSTANGY BROOKS SMITH & PROPHETE 2029 Century Park East, Ste. 1100 Los Angeles, CA 90067

John E. MacDonald CONSTANGY BROOKS SMITH & PROPHETE 3120 Princeton Pike, Ste. 301 Lawrenceville, NJ 08648

Donald S. Prophete CONSTANGY BROOKS SMITH & PROPHETE 2600 Grand Boulevard, Ste. 750 Kansas City, MO 64108 Counsel for Appellants

Allan Bloom

2 Adam L. Deming Mark D. Harris Adam Lupion Paul Salvatore PROSKAUER ROSE 11 Times Square, 17th Floor New York, NY 10019

John E. Roberts PROSKAUER ROSE One International Place Boston, MA 02110 Counsel for Amicus Appellants Southeastern Conference; American Council on Education; American Association of Community Colleges; American Association of State Colleges and Universities; Association of American Universities; Association of Catholic Colleges and Universities; Association of Governing Boards of Universities and Colleges; Association of Jesuit Colleges and Universities; Association of Public and Land-Grant Universities; College and University Professional Association for Human Resources; Council for Christian Colleges and Universities; National Association of College and University Business Officers; National Association of Independent Colleges and Universities; and Southern Association of Colleges and Schools Commission on Colleges

Erik R. Zimmerman ROBINSON BRADSHAW & HINSON 1450 Raleigh Road, Ste. 100 Chapel Hill, NC 27517 Counsel for Amicus Appellant Southeastern Conference

3 Benjamin F. Johns SHUB & JOHNS 200 Barr Harbor Drive Four Tower Bridge, Ste. 400 West Conshohocken, PA 19428 Counsel for Amicus Appellant Professor Michael H. Leroy

Renan Varghese Michael J. Willemin [ARGUED] WIGDOR 85 Fifth Avenue, 5th Floor New York, NY 10003

Paul L. McDonald 1800 John F. Kennedy Boulevard, Ste. 300 Philadelphia, PA 19103 Counsel for Appellees

_______________

OPINION OF THE COURT _______________

RESTREPO, Circuit Judge. Do efforts that provide tangible benefits to identifiable institutions deserve compensation? In most instances, they do. And yet athletes at our most competitive colleges and universities are told that their “amateur” status renders them ineligible for payment. The issue raised by this interlocutory

4 appeal is not whether the athletes before us are actually owed the protections of the Fair Labor Standards Act (FLSA), but rather, whether college athletes, by nature of their so-called amateur status, are precluded from ever bringing an FLSA claim. Our answer to this question is no.

This case originated in 2019 when athletes at several National Collegiate Athletic Association (NCAA) Division I (D-I) member schools filed a complaint asserting violations of the FLSA and various state wage laws. The plaintiffs argued that they were entitled to federal minimum wage compensation for the time they spent representing their schools. The NCAA and member schools moved to dismiss pursuant to Federal Rule of Civil Procedure Rule 12(b)(6), asserting that the athletes—as “amateurs”—are not, and historically have never been, considered employees of their respective schools or the NCAA. The District Court determined that the athletes had sufficiently pleaded facts that, under a multifactor balancing test, might allow them to be classified as employees under the FLSA and denied the motion to dismiss. The NCAA and member schools appealed.

For the reasons stated below, we will affirm in part the District Court’s decision denying Appellants’ motion to dismiss. But because the District Court erred by applying the test from Glatt v. Fox Searchlight Pictures, Inc., 811 F.3d 528 (2d Cir. 2016), to determine whether college athletes can be employees under the FLSA, we will vacate and remand for application of an economic realities analysis grounded in common-law agency principles.

5 I. BACKGROUND

Appellees contend that they are entitled to minimum wage under the FLSA for time spent on their sport-related activities.1 Appellants are thirteen colleges and universities that are members of the NCAA. The NCAA regulates intercollegiate sports and has jurisdiction over approximately 1,100 schools and some 500,000 athletes. The NCAA has multi-year, multi- billion-dollar contracts with ESPN, CBS, and Turner Sports to broadcast athletic competitions between D-I schools, and it distributes shares of those broadcasting fees to its member institutions. In addition to shares of broadcasting fees, D-I schools receive fees from multi-year, multi-million-dollar agreements with television and radio networks that they have entered, either individually or as part of an NCAA conference, to broadcast their athletic competitions. To understand how collegiate sport generates these revenues, a brief historical survey is instructive.2

A. College Athletics in Historical Context

American intercollegiate athletics began when a group of Yale students formed a boat club in 1843; undergraduates at

1 Additionally, Appellees raise claims under Connecticut, Pennsylvania, and New York state wage laws. 2 Given the significance of the question presented, we provide an abridged history of intercollegiate athletics. We do so purely for context, and our disposition does not rely upon this historical background.

6 Harvard followed suit the next year.3 In 1852, the two clubs staged our nation’s first intercollegiate athletic competition (The Race) on a lake in New Hampshire.4 From that first contest, the spectacle of college sports has grown steadily to become a multi-billion-dollar industry.5

Put simply, athletic victories have provided many colleges with the institutional visibility needed to facilitate tremendous growth.6 Indeed, although tension continues to exist between the demands of traditional education and athletics, even early college presidents came to see athletes as effective avatars for

3 Guy Lewis, The Beginning of Organized Collegiate Sport, 22 AM. Q. 222, 222, 224 (1970). 4 Id. at 224. For an in-depth history of The Race, see generally THOMAS C. MENDENHALL, THE HARVARD–YALE BOAT RACE, 1852–1924 (1993). 5 Doug J. Chung, The Dynamic Advertising Effect of Collegiate Athletics, 32 MARKETING SCI. 679, 681 (Sept.–Oct. 2013). 6 RONALD A. SMITH, SPORTS AND FREEDOM, at vii (1988). In 1800, there were twenty-five colleges in the United States. Today there are around four thousand. See Josh Moody, A Guide to the Changing Number of U.S. Universities, U.S. NEWS & WORLD REPORT (Apr. 27, 2021, 9:30 AM), https://www.usnews.com/education/best- colleges/articles/how-many-universities-are-in-the-us-and- why-that-number-is-changing.

7 their institutions.7 By 1875, intercollegiate regattas had become feature items in nationally distributed magazines and front-page material for leading newspapers.8 Both “students and the public began to regard victory as a measure of an institution’s prestige.”9 One student of the era explained that the contests were “sacredly connected with the glory of Alma Mater herself.”10

Such glory was especially valuable to lesser-known institutions. Take the 1871 regatta between Harvard, Brown, and the “Farmer Boys” of the Massachusetts Agricultural College of Amherst for example. An unexpected victory over widely-favored Harvard made the little-known, eight-year-old land grant college now known as UMass Amherst a nationally recognized institution overnight.11 More importantly, it inspired hope among other lesser-known colleges that they too might do the same.12

7 See, e.g., CLIFFORD PUTNEY, MUSCULAR CHRISTIANITY: MANHOOD AND SPORTS IN PROTESTANT AMERICA, 1880– 1920, at 39 (2001) (describing Harvard President Charles W. Eliot’s appreciation for “athletic and curricular reforms”). 8 Lewis, supra note 3, at 227–28 (citations omitted). 9 Id. 10 Id. 11 SMITH, supra note 6, at 43. 12 Id.

8 This phenomenon would later become known as the “Flutie Effect” following a 1984 football game between Boston College and the University of Miami. With six seconds on the clock and Miami up by four, Boston College’s quarterback, Doug Flutie, completed an astounding forty-eight-yard Hail Mary touchdown pass to win the game. Over the next two years, applications to Boston College jumped thirty percent.13 Successful football and basketball programs have more recently driven notoriety and applications to, among other institutions, Georgetown, Northwestern, Boise State, Texas Christian University, Butler, Gonzaga, Virginia Commonwealth University, Texas A&M, Florida Gulf Coast, Lehigh, and Wichita State.14

13 Harvard Business School Working Knowledge, The Flutie Effect: How Athletic Success Boosts College Applications, FORBES (Apr. 29, 2013, 9:48 AM), https://www.forbes.com/sites/hbsworkingknowledge/2013/04 /29/the-flutie-effect-how-athletic-success-boosts-college- applications/?sh=2e892b8e6e96. 14 See, e.g., Nicholas Fram & T. Ward Frampton, A Union of Amateurs: A Legal Blueprint to Reshape Big-Time College Athletics, 60 Buff. L. Rev. 1003, 1019 (2012); Michael McCann, The Flutie Effect: How UMBC Can Benefit From a Historic NCAA Tournament Upset, SPORTS ILLUSTRATED (Mar. 17, 2018), https://www.si.com/college/2018/03/17/umb c-virginia-upset-doug-flutie-jairus-lyles; Hayley Glatter, The March Madness Application Bump, THE ATLANTIC (Mar. 16, 2017), https://www.theatlantic.com/education/archive/2017/0 3/the-march-madness-application-bump/519846/; Polly Mos- endz, Dashiell Bennett & Lance Lambert, March Madness Cinderella Stories Send Applications Soaring, BLOOMBERG

9 Indeed, Professor Doug J. Chung describes athletic programs as higher education’s primary form of mass media advertising. In one study, Professor Chung found that raising a football team from mediocrity to national status caused, on average, a 17.7 percent increase in the number of applications to the team’s institution.15 Increased applications then contribute to a positive feedback loop producing more revenue, greater selectivity in admissions, improved alumni engagement, greater fundraising, and better faculty recruiting, all of which can catapult regional universities into national prominence in a way that would otherwise be impossible.16

Profit, after all, has always played a role in intercollegiate college athletics. The Race—the very first intercollegiate competition—was neither proposed nor organized by the students of Yale or Harvard, but by James Elkins, the superintendent of the Boston, Concord, and Montreal Railroad.17 Mr. Elkins had hoped that staging a regatta on Lake Winnipesaukee would increase ridership on his rail line, raise the value of his nearby real estate holdings, and bring tourists to the quiet, lakeside resort. Unsurprisingly, the teams were treated to a lavish vacation, and the winners received “a

(Mar. 13, 2017, 11:00 AM), https://www.bloomberg.com/new s/features/2017-03-13/march-madness-more-students-apply- to-schools-that-break-brackets. 15 Chung, supra note 5, at 681. 16 McCann, supra note 14. 17 SMITH, supra note 6, at 3–4, 27–29.

10 handsome pair of black, silver-tipped, walnut oars.”18 Such commercialization was the norm in early athletic competitions.19

The same is true today. First and foremost, the colleges themselves stand to profit substantially from television contracts, licensing fees, and ticket, concessions, and merchandise sales that their athletic programs generate. Some estimate that college athletes generate roughly $3 billion in annual revenue for their schools, conferences, and the NCAA.20 And at least 38 NCAA member colleges currently gross more than $100 million annually in sports revenue.21 The athletic department of the University of Texas, for example, reported $271 million in revenue for 2023, more than the highest-earning National Hockey League team.22 In 2020, 63

18 Id. at 27–29. 19 See, e.g., id. at 29–34, 42–51. 20 Liz Clarke, The NCAA Coined the Term ‘Student- athlete’ in the 1950s. Its Time Might Be Up, WASH. POST (Oct. 28, 2021, 9:00 AM), https://www.washingtonpost.com/sports/ 2021/10/27/ncaa-student-athlete-1950s/. 21 Marc Edelman, Why Congress Would Be Crazy to Grant the NCAA an Antitrust Exemption, FORBES (May 6, 2020, 9:50 AM), https://www.forbes.com/sites/marcedelman/2020/05/06 /why-congress-would-be-crazy-to-grant-the-ncaa-an-antitrust- exemption/?sh=6cdddce070a9. 22 Steve Berkowitz & Kirk Bohls, Texas Reported Athletic Department Revenue of $271 Million in 2023, a Record for NCAA Schools, USA TODAY (Jan. 17, 2024, 5:19 PM),

11 other NCAA member colleges earned more than $25 million from their football programs.23

The colleges are not alone in profiting. One study reported that 45 million Americans planned to wager a combined $3.1 billion on the 2022 NCAA March Madness basketball tournament.24 And the NCAA itself, first founded to help regulate dangerous playing conditions, has grown into a financial behemoth with revenues often exceeding $1 billion annually.25 ESPN, for example, recently announced that it will

https://www.usatoday.com/story/sports/ncaaf/big12/2024/01/ 17/texas-athletic-department-271-million- revenue/72255138007/; Mike Ozanian & Justin Teitelbaum, NHL Team Values 2022: New York Rangers on Top at $2.2 Billion, FORBES (Dec. 14, 2022, 6:23 AM), https://www.forbes.com/sites/mikeozanian/2022/12/14/nhl- team-values-2022-new-york-rangers-on-top-at-22- billion/?sh=7fe345287deb. 23 Brad Crawford, Ranking College Football’s Richest, Poorest Programs, 247 SPORTS (Dec. 29, 2020, 7:45 AM), https://247sports.com/longformarticle/college-football- richest-poorest-programs-alabama-crimson-tide-texas- longhorns-ohio-state-buckeyes-157982941/. 24 45 Million Americans to Wager $3.1B on March Madness, AM. GAMING ASS’N (Mar. 13, 2022), https://www.americangaming.org/resources/march-madness- 2022/. 25 Associated Press, NCAA Earns $1.15 Billion in 2021 as Revenue Returns to Normal, ESPN (Feb. 2, 2022, 3:53 PM), https://www.espn.com/college-

12 pay the NCAA $115 million each year for exclusive broadcast rights to 40 leagues’ championship games.26 March Madness and the College Football Playoff each have their own television deals valued at $8.8 billion and $5.64 billion, respectively.27

By far the most obvious beneficiaries of college sports are a select few administrators, athletic directors, and coaches.28 The recently retired Alabama football coach, Nick Saban, earned over $11.4 million in his last year leading the Crimson Tide, making him the highest-paid coach in college sports29

sports/story/_/id/33201991/ncaa-earns-115-billion-2021- revenue-returns-normal. 26 Ben Straus, ESPN, NCAA Strike $920 Million Deal to Keep 40 Championships on Network, WASH. POST (Jan. 4, 2024, 11:45 AM), https://www.washingtonpost.com/sports/20 24/01/04/ncaa-tv-deal-espn/. 27 Id. 28 Marc Edelman, Note, Reevaluating Amateurism Standards in Men’s College Basketball, 35 U. MICH. J.L. REFORM 861, 874 (2002) (noting that former NCAA Executive Director Cedric Dempsey earned $647,000 [$1.096 million adjusted] each year during the described period and that many coaches made more that $1 million [$1.694 million adjusted] each year in total compensation). 29 College Football Head Coach Salaries, USA TODAY (Oct. 2, 2023, 7:18 PM), https://sports.usatoday.com/ncaa/sal aries/football/coach.

13 and the eighth-highest-paid football coach in America.30 In fact, seventeen of the 37 highest-paid coaches in any sport in the United States make their living in college football or basketball.31 By contrast, university professors and administrators make far less. The University of Virginia, for example, pays $900,000 each year to its president32 and $600,000 to its law school dean33 while its basketball coach receives $5.2 million.34 This is not abnormal; in forty states,

30 Reem Abdalazem & Jeffrey May, Who Are the Highest and Lowest Paid Coaches in the NFL?, AS (Jan. 19, 2024, 10:10 AM), https://en.as.com/nfl/who-are-the-highest-and- lowest-paid-nfl-coaches-n-2/. 31 Kurt Badenhausen, Highest-Paid Coaches 2023: Belichick, Payton, Popovich Pocket $16m+, SPORTICO (Nov. 28, 2023, 12:02 AM), https://www.sportico.com/personalities /people/2023/highest-paid-coaches-2023-belichick-payton- popovich-1234747782/; Highest Paid Coaches in U.S. Sports, SPORTICO (Nov. 28, 2023), https://www.sportico.com/feature/ highest-paid-coaches-american-sports-1234747983/. 32 Sarah Larimer, U-Va.’s New President Will Earn $750,000 a Year, WASH. POST (Sept. 17, 2021, 4:16 PM), https://www.washingtonpost.com/local/education/u-vas-new- president-will-earn-750000-a-year/2017/09/20/2a5cfdce- 9e2f-11e7-9083-fbfddf6804c2_story.html. 33 U.Va. Faculty & Staff Salaries 2022, CAVALIER DAILY (Aug. 2022), https://www.cavalierdaily.com/page/faculty- salary-2022. 34 Abigail Johnson Hess, UVA Basketball Coach Turned Down a Raise After National Championship Win: ‘I Have

14 the highest-paid public employee is a D-I coach.35 Ohio State president E. Gordon Gee summarized this upside-down world when, asked whether he would consider firing his embattled football coach, he quipped, “I’m just hoping the coach doesn’t dismiss me.”36

B. Amateurism and the “Student-Athlete” in College Athletics

We have opted against using a term both parties employ liberally in briefing: “student-athlete.” Like “band-aid” or

More Than I Need’, CNBC (Sept. 17, 2019, 10:21 AM), https://www.cnbc.com/2019/09/17/why-uva-basketball- coach-tony-bennett-turned-down-a-raise.html. Public university presidents in 1986 slightly out-earned head football coaches; now coaches earn almost four times as much as university presidents. Fram & Frampton, supra note 14, at 1020. 35 David Evans, Complete List Of The Highest-Paid State Employees: College Coaches Dominate Top Earning Public Sector Workers, SPORTS DAILY (July 20, 2023), https://thesportsdaily.com/news/complete-list-of-the-highest- paid-state-employees-college-coaches-dominate-top-earning- public-sector-workers/. In our circuit, two out of three states’ highest-paid employees are college football coaches: James Franklin at Penn State, who earns $8.7 million per year, and Greg Schiano at Rutgers, who earns $4 million each year. Id. 36 Fram & Frampton, supra note 14, at 1020 (citing Albert R. Hunt, Athletics Overrun the Ivory Tower, INT’L HERALD TRIBUNE 2 (Dec. 12, 2011)).

15 “laundromat,” “student-athlete” is essentially a brand name that has become synonymous with its product.37 As scholars have noted, the term is an NCAA marketing invention designed to “conjure the nobility of amateurism,” assert “the precedence of scholarship over athletic[s],” and “obfuscate the nature of the legal relationship at the heart of a growing commercial enterprise.”38 Context makes this vividly apparent.39

The NCAA arose from the public outcry over the dangers of early college football. In 1904 alone, at least twenty players died, not on battlefields, but on football fields.40 The next year, urged on by President Theodore Roosevelt, a group of colleges chartered the non-profit organization that would become the NCAA to establish common safety guidelines in college athletics. In doing so, the NCAA also promoted an ethos of strict amateurism that forbade all forms of payment, including athletic scholarships. Yet for the first fifty years of its existence, this ethos was openly defied: most member schools admitted

37 Mark Abadi, Taser, Xerox, Popsicle, and 31 More Brands-Turned-Household Names, BUSINESS INSIDER (June 3, 2018, 10:41 AM), https://www.businessinsider.com/google- taser-xerox-brand-names-generic-words-2018-5. 38 Fram & Frampton, supra note 14, at 1015. 39 See generally RONALD A. SMITH, THE MYTH OF THE AMATEUR: A HISTORY OF COLLEGE ATHLETIC SCHOLARSHIPS (2021). 40 Fram & Frampton, supra note 14, at 1013.

16 to offering under-the-table compensation to star players.41 By the 1950s, even the ban on athletic scholarships—a central tenant of the original, British-inspired amateur ideal—lacked so much as a pretense of enforcement. In reality, such payments were already quite commonplace.42 Conceding defeat to this fact, the NCAA elected in 1956 to bring some forms of compensation (including athletic scholarships) aboveground in the hope that it could better regulate the market.43 But the NCAA also foresaw the explosion of college athletics and hoped to both facilitate and capitalize on that growth. Athletics scholarships proved to be an ideal mechanism for promoting order and retaining economic control.44

In response, courts began to question the economic realities of college athletics. Two state appellate court cases in particular took direct aim at the professed amateur status of athletes at D-I schools.45 Those courts saw that college sports

41 Id. 42 Id. at 1013–14. 43 See SMITH, supra note 39, at 127. 44 Fram & Frampton, supra note 14, at 1014. 45 See Van Horn v. Indus. Accident Comm’n, 33 Cal. Rptr. 169, 172 (Ct. App. 1963) (holding decedent scholarship-athlete eligible for benefits, as he “participated in the college football program under a contract of employment with the college”); Univ. of Denver v. Nemeth, 257 P.2d 423, 426 (Colo. 1953) (holding that an athlete employed by the university to “discharge certain duties, not a part of his education program,” i.e., to compete, “is no different than the employee who is

17 had become a big business, and that athletes thus operated in the dual capacity of both student and employee.46 These cases stoked fears in NCAA leaders that college athletes might someday receive statutory employment protections.47 The NCAA’s answer was the term “student-athlete,” which it imposed as the exclusive label for its players.48 As the historian Taylor Branch notes:

The term student-athlete was deliberately ambiguous. College players were not students at play (which might understate their athletic obligations), nor were they just athletes in college (which might imply they were professionals). That they were high-performance athletes meant they could be forgiven for not meeting the academic standards of their peers; that they were students meant they did not have to be compensated, ever, for anything more than the cost of their studies. Student-athlete became the NCAA’s signature term, repeated constantly in and out of courtrooms.49

taking no course of instruction” so far as state workers’ compensation laws are concerned). 46 Fram & Frampton, supra note 14, at 1014. 47 Id. at 1015. 48 Id. 49 Taylor Branch, How the Myth of the NCAA “Student- Athlete” Was Born, DEADSPIN (Feb. 20, 2014),

18 The NCAA’s strategy has worked for some time, supported in part through dicta from the Supreme Court’s decision in NCAA v. Board of Regents of the University of Oklahoma, stating that “[t]he NCAA plays a critical role in the maintenance of a revered tradition of amateurism in college sports.” 468 U.S. 85, 120 (1984). With Board of Regents in hand, the NCAA and its member colleges have largely succeeded in persuading courts to grant the concept of amateurism the force of law.50 As one federal district judge

https://deadspin.com/how-the-myth-of-the-ncaa-student- athlete-was-born-1524282374. 50 See, e.g., Berger v. NCAA, 843 F.3d 285, 291 (7th Cir. 2016) (“[The] long-standing tradition [of amateurism] defines the economic reality of the relationship between student athletes and their schools.” (citing Board of Regents, 468 U.S. at 120)); Agnew v. NCAA, 683 F.3d 328, 344 (7th Cir. 2012) (“For the purposes of college sports, and in the name of amateurism, we consider players who receive nothing more than educational costs in return for their services to be ‘unpaid athletes.’”); Bassett v. NCAA, 528 F.3d 426, 433 (6th Cir. 2008) (“It also violates the spirit of amateur athletics by providing remuneration to athletes in exchange for their commitments to play for the violator’s football program.”); Smith v. NCAA, 139 F.3d 180, 187 (3d Cir. 1998) (“We agree with these courts that, in general, the NCAA’s eligibility rules allow for the survival of the product, amateur sports, and allow for an even playing field.”), vacated, 525 U.S. 459 (1999); McCormack v. NCAA, 845 F.2d 1338, 1345 (5th Cir. 1988) (“That the NCAA has not distilled amateurism to its purest form does not mean its attempts to maintain a mixture containing some amateur elements are unreasonable.”).

19 wrote, “[e]ven in the increasingly commercial modern world, there is still validity to the Athenian concept of a complete education derived from fostering full growth of both mind and body.” Gaines v. NCAA, 746 F. Supp. 740, 744 (M.D. Tenn. 1990). The Court of Appeals for the Seventh Circuit similarly found the idea that college athletes are “selling their services” and that universities are “purchasers of labor” to be a “surprisingly cynical view of college athletics.” Banks v. NCAA, 977 F.2d 1081, 1091–92 (7th Cir. 1992). College football players, the Court reasoned, are not market participants because they are “student-athletes.” See id. at 1090 (“We consider college football players as student-athletes simultaneously pursuing academic degrees that will prepare them to enter the employment market in non-athletic occupations.” (citing Justice v. NCAA, 577 F. Supp. 356, 382 (D. Ariz. 1983)).

Until recently, NCAA rules barred athlete compensation beyond “tuition and fees, room and board, books and other expenses related to attendance.”51 But the Supreme Court’s unanimous decision in NCAA v. Alston disrupted the status quo by holding that that Board of Regents did not create a binding precedent “reflexively” supporting the organization’s compensation rules. 141 S. Ct. 2141, 2158 (2021). The NCAA responded by changing its rules to allow athletes to profit from their name, image, and likeness (NIL) with direct endorsement

51 NCAA, DIVISION I MANUAL 209 (2021), https://web3.ncaa.org/lsdbi/reports/getReport/90008.

20 deals.52 Historically, the NCAA and the colleges had been the only entities permitted to do so.53

Justice Kavanaugh, in an oft-cited concurrence, noted that the NCAA’s remaining rules restricting non-education-related compensation raised serious antitrust questions as well. See Alston, 141 S. Ct. at 2166–67 (Kavanaugh, J., concurring). Rebuking the NCAA’s argument that maintaining compensation restrictions is necessary to distinguish college athletics from professional athletics, Justice Kavanaugh wrote that “[b]usinesses like the NCAA cannot avoid the consequences of price-fixing labor by incorporating price- fixed labor into the definition of [its] product.” Id. at 2168. Although Justice Kavanaugh did suggest that the NCAA could protect itself from future judicial scrutiny by permitting collective bargaining, id. at 2168, he also flatly concluded that “[n]owhere else in America can businesses get away with agreeing not to pay their workers a fair market rate on the theory that their product is defined by not paying their workers a fair market rate. . . . The NCAA is not above the law,”54 id. at 2169.

52 Alan Blinder, College Athletes May Earn Money From Their Fame, N.C.A.A. Rules, N.Y. TIMES (Sept. 29, 2021), https://www.nytimes.com/2021/06/30/sports/ncaabasketball/n caa-nil-rules.html. 53 Id.; Fram & Frampton, supra note 14, at 1019. 54 Incidentally, the NCAA’s former president Myles Brand highlighted the weakness of the organization’s defense over a decade ago—though perhaps inadvertently—in an interview for Sports Illustrated:

21 Appellants raise similarly circular arguments. But as the Supreme Court recently suggested, such rationales no longer hold the weight they once did. See Alston, 141 S. Ct. at 2158. The National Labor Relations Board (NLRB), likely in response to Alston, is for the first time taking the position that college athletes are employees for purposes of the National Labor Relations Act (NLRA).55 In the midst of these changes, our Court is the first to consider whether college athletes may also be employees under the ambit of the related FLSA.

[Brand] They can’t be paid. [Q] Why? [Brand] Because they’re amateurs. [Q] What makes them amateurs? [Brand] Well, they can’t be paid. [Q] Why not? [Brand] Because they’re amateurs. [Q] Who decided they are amateurs? [Brand] We did. [Q] Why? [Brand] Because we don’t pay them.

Michael Rosenberg, Change Is Long Overdue: College Football Players Should Be Paid, SPORTS ILLUSTRATED (Aug. 26, 2010), https://www.si.com/more-sports/2010/08/26/pay- college. 55 Office of Public Affairs, NLRB General Counsel Jennifer Abruzzo Issues Memo on Employee Status of Players at Academic Institutions, NLRB (Sept. 29. 2021), https://www.nlrb.gov/news-outreach/news-story/nlrb-general- counsel-jennifer-abruzzo-issues-memo-on-employee-status- of.

22 C. The Athletes at Bar

The plaintiffs allege that although the NCAA and its member schools profit from their efforts, the NCAA’s bylaws prohibit member schools from offering wages and forbid students from accepting them. To enforce these rules, the bylaws prescribe sanctions for violating schools and students, including suspension or termination of athletes, suspension of coaching staff, and disqualification of teams from competitions. The NCAA and defendant schools argue that, although athletes do not earn wages, the benefits of participation include payment in other forms, such as increased discipline, a stronger work ethic, improved strategic thinking, time management, leadership, and goal setting skills, and a greater ability to work collaboratively.

The athletes allege that the soft skills the Appellants point to are inadequate compensation for their services and that they were subject to extensive training and performance requirements that regularly interfered with their learning. As just one example, the plaintiffs allege that they were forced to schedule classes around their athletic commitments, limiting their range of learning options. During the football season at Villanova University, for example, Mr. Johnson was allegedly required to spend weekdays from 5:45 AM to 11:30 AM practicing or engaging in other activities related to athletics. This commitment locked him out of hundreds of available classes, including prerequisites for certain academic degrees. In addition to Mr. Johnson’s personal experiences, the athletes cite to studies showing that NCAA requirements frequently prevent athletes from pursuing their preferred majors.

23 In their First Amended Complaint,56 the athletes asserted claims under the FLSA for the NCAA’s and member colleges’ failure to pay them a minimum wage and sought relief in the form of unpaid wages, an equal amount in liquidated damages, and attorneys’ fees. Some athletes also asserted state-specific failure-to-pay claims. Finally, the athletes asserted unjust enrichment claims. The defendant schools moved to dismiss the First Amended Complaint pursuant to Federal Rule of Civil Procedure 12(b)(6), arguing that athletes cannot be employees as a matter of law and therefore had failed to state a claim.

On August 25, 2021, the District Court rejected this argument. In the absence of controlling authority providing a specific multifactor test to evaluate whether athletes can be considered “employees” under the FLSA, the District Court applied the Court of Appeals for the Second Circuit’s multifactor test from Glatt, 811 F.3d 528, where the Court considered whether unpaid interns must be deemed employees under the FLSA and therefore compensated for their work. The District Court determined that Glatt required it to assess the “economic reality” of the relationship by identifying whether the athletes or the NCAA and schools were the primary beneficiary of the relationship. After balancing and considering the seven Glatt factors, the District Court concluded that the athletes had plausibly pleaded that they may

56 Appellees filed their initial complaint on November 6, 2019, and subsequently filed two amended complaints on December 12, 2019, and September 23, 2021, respectively. See App. at 43 n.1 (stating Second Amended Complaint, filed after the District Court issued the order and memorandum on the motion to dismiss, did not alter any of the claims).

24 be employees and denied the motion to dismiss. Subsequently, the District Court granted the Appellants’ motion under 28 U.S.C. § 1292(b) to certify an interlocutory appeal from the denial of their motion to dismiss. The question certified for appeal was: “Whether NCAA Division [I athletes] can be employees of the colleges and universities they attend for purposes of the Fair Labor Standards Act solely by virtue of their participation in interscholastic athletics.” App. at 35.

II. DISCUSSION57

57 We have jurisdiction to review interlocutory appeals certified under 28 U.S.C. § 1292(b). When reviewing an interlocutory appeal, we exercise de novo review over the certified question. In re Enter. Rent-A-Car Wage & Hour Emp. Pracs. Litig., 683 F.3d 462, 467 (3d Cir. 2012). Our review is not limited to the certified question and may include consideration of “any issue fairly included within the certified order.” Barbato v. Greystone All., LLC, 916 F.3d 260, 264 (3d Cir. 2019) (citing Yamaha Motor Corp., U.S.A. v. Calhoun, 516 U.S. 199, 205 (1996)). We have even gone so far as to declare that, because it is the order that is appealable in an interlocutory appeal, “we may address any issue necessary to decide the appeal before us.” Ivy Club v. Edwards, 943 F.2d 270, 275 (3d Cir. 1991). This “plainly includes the threshold question” of whether the athletes before us have established a prima facie case. Egervary v. Young, 366 F.3d 238, 245 (3d Cir. 2004).

In reviewing a motion to dismiss, we “review any legal determinations anew and presume that a complaint’s factual allegations are true.” Pennsylvania v. Navient Corp., 967 F.3d 273, 283 n.7 (3d Cir. 2020) (reviewing de novo, following § 1292(b) certification, a district court’s denial of a motion to

25 The FLSA protects “the rights of those who toil, of those who sacrifice a full measure of their freedom and talents to the use and profit of others.” Tenn. Coal, Iron & R.R. Co. v. Muscoda Local No. 123, 321 U.S. 590, 597 (1944). Accordingly, it gives specific, non-waivable minimum protections to individuals to ensure that each covered employee receives “[a] fair day’s pay for a fair day’s work,” Overnight Motor Transp. Co. v. Missel, 316 U.S. 572, 578 (1942) (quoting 81 Cong. Rec. 4983 (1937) (statement of President Franklin Roosevelt)), and is protected from “the evil of ‘overwork’ as well as ‘underpay,’” id.

Consistent with the FLSA’s “remedial and humanitarian” purpose, Tenn. Coal, 321 U.S. at 597, Congress adopted definitions of “employee” and “employer” that brought a broad swath of workers under the statute’s coverage, including even “those who would decline its protections.” Tony & Susan Alamo Found. v. Sec’y of Labor, 471 U.S. 290, 302 (1985). Accordingly, “the term ‘employee’ means any individual employed by an employer,” 29 U.S.C. § 203(e)(1), a definition that has been described as “the broadest . . . that has ever been included in any one act.” United States v. Rosenwasser, 323 U.S. 360, 363 n.3 (1945) (quoting 81 Cong. Rec. 7657 (1937) (statement of Sen. Hugo Black)). Similarly open-ended, an “employer” is “any person acting directly or indirectly in the interest of an employer in relation to an employee,” 29 U.S.C. § 203(d), and to “employ” is “to suffer or permit to work,” id. § 203(g). These “statutory definitions regarding employment status are necessarily broad to effectuate the remedial purposes

dismiss for failure to state a claim). The presumption of truth does not extend to legal conclusions. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009).

26 of the Act.” Martin v. Selker Bros., 949 F.2d 1286, 1293 (3d Cir. 1991).

The “striking breadth” of these definitions brings within the FLSA’s ambit workers “who might not qualify as [employees] under a strict application of traditional agency law principles” or under other federal statutes, Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 326 (1992), and these definitions have long been held to apply notwithstanding any “prior custom or contract . . . not to compensate employees for certain portions of their work.” Tenn. Coal, 321 U.S. at 602. Accordingly, to determine employment under the Act, the Supreme Court has instructed that we “look to the economic realities of the relationship.” Martin, 949 F.2d at 1293 (citing Rutherford Food Corp. v. McComb, 331 U.S. 722, 730 (1947)).

Under this framework, the employer-employee “relationship does not depend on . . . isolated factors but rather upon the circumstances of the whole activity.” McComb, 331 U.S. at 730; see also Haybarger v. Lawrence Cnty. Adult Prob. & Parole, 667 F.3d 408, 418 (3d Cir. 2012) (holding that “whether a person functions as an employer depends on the totality of the circumstances rather than on ‘technical concepts of the employment relationship’”) (citation omitted). Limitations articulated by the Supreme Court include that independent contractors are not employees under the FLSA, see McComb, 331 U.S. at 729, and “[a]n individual who, ‘without promise or expectation of compensation, but solely for his personal purpose or pleasure, worked in activities carried on by other persons either for their pleasure or profit,’ is outside the sweep of the Act,” Tony & Susan Alamo Found., 471 U.S. at 295 (quoting Walling v. Portland Terminal Co., 330 U.S. 148, 152 (1947)).

27 Importantly, in determining that the Alamo “volunteers” were actually employees because they expected “in-kind” compensation for services performed, Tony & Susan Alamo Found., 471 U.S. at 301, the Court distinguished their situation from that of a group of trainees in Walling. Id. at 299–301. In Walling, the trainees participated in a week-long course, during which they performed some work under close supervision without receiving or expecting remuneration beyond the possibility of future employment. 330 U.S. at 149–150. But the Court held that the trainees did not qualify as “employees” under the FLSA, as their work did not confer an “immediate advantage” to the purported employer. Id. at 153. Instead, as the Court in Alamo explained, the trainees in Walling were akin to “students in a school,” whose activities are driven by the educational benefits. Tony & Susan Alamo Found., 471 U.S. at 300. By contrast, the Alamo “volunteers” engaged in work over extended periods, sometimes years, and received “in-kind benefits” like food, clothing, shelter, and other benefits as compensation. Id. at 292. These benefits were “wages in another form.” Id. at 301. Even though the Alamo “volunteers” claimed they expected no compensation, the Court explained that a compensation agreement can be either “express” or “implied,” and “[i]f an exception to the Act were carved out for employees willing to testify that they performed work “voluntarily,” employers might be able to use superior bargaining power to coerce employees to make such assertions, or to waive their protections under the Act.” Id. at 301–02.

Since McComb, we and other courts of appeal have adopted multifactor tests to analyze, based on the circumstances of the whole relationship between the parties, whether individuals are employees or independent contractors, whether entities are

28 joint employers, and whether individuals are employees or interns. See, e.g., Donovan v. DialAmerica Mktg., Inc., 757 F.2d 1376, 1382 (3d Cir. 1985) (adopting a six-factor test developed to determine whether an individual is an employee or an independent contractor); In re Enterprise Rent-A-Car, 683 F.3d at 469 (adopting a four-factor test to determine whether entities are joint employers); Glatt, 811 F.3d at 536– 37 (adopting a non-exhaustive set of seven factors to determine whether an individual is a student intern or an employee). Here, we confront circumstances unlike those previously addressed, but core principles that traditionally define an employee- employer relationship are no less applicable.

A. Determining the Employment Status of College Athletes

In looking to “the economic realities of the relationship” between college athletes and their schools or the NCAA, Martin, 949 F.2d at 1293, we begin by noting that athletes in the collegiate context are sui generis. After all, merely playing sports, even at the college level, cannot always be considered commercial work integral to the employer’s business in the same way that the activities performed by independent contractors or interns are assumed to be in previously mentioned multifactor tests. See, e.g., Donovan, 757 F.2d at 1379–83 (undisputed that home-based researchers who distributed their and other researchers’ confirmed phone numbers to a telemarketing firm performed work); Glatt, 811 F.3d at 531 (undisputed that “[p]laintiffs worked as unpaid interns” on a Fox Searchlight-distributed film or at the Fox corporate offices (emphasis added)). The Supreme Court has acknowledged this possibility, explaining that the FLSA does not cover a person who, “without promise or expectation of

29 compensation, but solely for his personal purpose or pleasure” performs “activities carried on by other persons either for their pleasure or profit.” Walling, 330 U.S. at 152. The Department of Labor (DOL) makes the same distinction.58 But just as intuitively, with professional athletes as the clearest indicators, playing sports can certainly constitute compensable work. Any test to determine college athlete employee status under the FLSA must therefore be able to identify athletes whose play is also work.

For its part, the FLSA does not define “work.” The Supreme Court “broadly” interprets it in the FLSA context and initially defined it as “physical or mental exertion (whether burdensome or not) controlled or required by the employer and pursued necessarily and primarily for the [employer’s] benefit.” Tenn. Coal, 321 U.S. at 598. The Court has since

58 See U.S. DEP’T OF LABOR, FIELD OPERATIONS HANDBOOK § 10b03(e) (2016) (activity of college students participating in interscholastic athletics primarily for their own benefit as part of the educational opportunities provided to the students by the school is not “work”). Appellants argue that Department of Labor guidance entitles them to a complete statutory defense against Appellees’ claims. We will not address this argument as “an affirmative defense may not be used to dismiss a plaintiff’s complaint under Rule 12(b)(6).” In re Adams Golf, Inc. Sec. Litig., 381 F.3d 267, 277 (3d Cir. 2004). Although there is an exception to this rule when the affirmative defense appears on the face of the complaint, see Jones v. Bock, 549 U.S. 199, 215 (2007), that exception is not applicable here, as the Appellees’ reliance on DOL guidance is not established on the face of the Complaint.

30 clarified that “exertion” is “not in fact necessary for an activity to constitute ‘work’” because “an employer . . . may hire [someone] to do nothing.” IBP, Inc. v. Alvarez, 546 U.S. 21, 25 (2005) (quotation omitted). Accordingly, for an activity to constitute “work” it need only be controlled by an employer and pursued necessarily and primarily for that employer’s benefit. See De Asencio v. Tyson Foods, Inc., 500 F.3d 361, 371 (3d Cir. 2007) (explaining that courts should “preclude[] the consideration of cumbersomeness or difficulty on the question of whether activities are ‘work’”). A putative employee, meanwhile, is expected to receive either express or implied “in-kind” compensation for services rendered. Tony & Susan Alamo Found., 471 U.S. at 301.

Read together, these cases largely mirror common-law agency principles others have used to help decide cases involving similar purported employer-employee relationships. Chief among them is the NLRB’s decision in Trustees of Columbia University in the City of New York, 364 NLRB 1080, 1081 (2016), where the Board applied a common-law agency test (also known as the “right-of-control” test) to answer the threshold question of whether graduate students who perform services at a university in connection with their studies are statutory employees within the meaning of Section 2(3) of the NLRA. Id. at 1081. That test asks whether the individual, in return for payment, performs services under the control of another person, or under a person with the right to control such services. Id. at 1081–82, 1094. By reverting to common-law agency principles, the Board notably rejected a Glatt-like primary beneficiary analysis. Id. at 1101 n.49. Instead, the Board held that student teaching and research assistants are employees under the NLRA if they meet the Act’s broad definition of “employee,” which encompasses individuals who

31 meet the common law test for employment. Trs. of Columbia Univ., 364 NLRB at 1083 (“Where student assistants have an employment relationship with their university under the common law test—which they do here—this relationship is sufficient to establish that the student assistant is a Section 2(3) employee for all statutory purposes.”).

We recognize that the NLRA and FLSA have distinct policy goals, but their shared history often inspires courts to draw interchangeably from each statute’s caselaw to answer fundamental questions related to the equitable regulation of the American workplace.59 See, e.g., Kasten v. Saint-Gobain Performance Plastics Corp., 563 U.S. 1, 13 (2011) (“The similar enforcement needs of [the NLRA] argue for an interpretation of the word ‘complaint’ [in the FLSA] that would provide ‘broad rather than narrow protection to the employee.’” (quoting NLRB v. Scrivener, 405 U.S. 117, 122 (1972))); Maryland v. Wirtz, 392 U.S. 183, 192 (1968) (relying on NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1 (1937), to explain that Congress had commerce power to expand the FLSA); In re Enterprise Rent-A-Car, 683 F.3d at 468 (adopting the FLSA definition of “joint employer” from an earlier NLRA

59 Moreover, although graduate-student employees are often exempt from FLSA coverage (and notably, college athletes are not), see 29 U.S.C. § 213(a)(1), the graduate student may be the closest equivalent we have. Like the athlete, the graduate student is enrolled in college or university. Like the college athlete, the graduate student’s tuition is often covered by the institution as a recruitment incentive. Also similar to the college athlete, the graduate student sometimes performs work for the institution that involves little educational value or direct connection to a course of study.

32 case); Brock v. Richardson, 812 F.2d 121, 124 (3d Cir. 1987) (“NLRA cases are often considered of assistance in interpreting the Fair Labor Standards Act.”) (citation omitted).

Significantly, the NLRA and FLSA both use broad definitions of “employee” and “employer” to delineate statutory coverage. See 29 U.S.C. §§ 152(2)–(3), 203(d)–(e). The “striking breadth” of the FLSA’s definitions, after all, brings within the Act’s purview workers “who might not qualify as [employees] under a strict application of traditional agency law principles.” Darden, 503 U.S. at 326. It necessarily follows that determining an employer-employee relationship under the FLSA includes, but is not limited to, a strict application of traditional agency law principles. The NLRA, meanwhile, does not explicitly define the terms. But it is well established that “when Congress uses the term ‘employee’ in a statute that does not define the term, courts interpreting the statute ‘must infer, unless the statute otherwise dictates, that Congress means to incorporate the established meaning”’ of the term, with reference to “common-law agency doctrine.” NLRB v. Town & Country Elec., 516 U.S. 85, 94 (1995) (quoting Darden, 503 U.S. at 322–323); see also Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 739–40 (1989) (“when Congress has used the term ‘employee’ without defining it, we have concluded that Congress intended to describe the conventional master-servant relationship as understood by common-law agency doctrine”) (citations omitted). Put otherwise, common-law agency doctrine, a doctrine largely symmetrical to governing FLSA caselaw, is also a helpful analytical tool in evaluating college athletes’

33 purported employer-employee relationships in either the NLRA or the FLSA context.60

We do not reproach the District Court for being drawn to Glatt. But while we agree with our sister circuit that “an employment relationship is not created when the tangible and intangible benefits provided to [a] [worker] are greater than [a] [worker]’s contribution to the employer’s operation,” Glatt, 811 F.3d at 535; see Tenn. Coal, 321 U.S. at 598, Glatt’s overall utility with respect to college athletes is undercut by its accurate presumption that unpaid interns all perform work for their employers.

Indeed, the facts that animate Glatt are not sufficiently analogous to the case at bar because the work performed during properly designed unpaid internships “can greatly benefit interns,” as “the intern enters into the relationship with the expectation of receiving educational or vocational benefits that are not necessarily expected with all forms of employment.” 811 F.3d at 535–36. Meanwhile, the educational and vocational benefits of college athletics cited by Appellants as alternative forms of remuneration (increased discipline, a stronger work ethic, improved strategic thinking, time management, leadership, and goal setting skills, and a greater ability to work collaboratively) are all exactly the kinds of skills one would typically acquire in a work environment. Additionally, the Glatt test has limited relevance to athletes because it compares

60 Indeed, decision-makers in the NLRA context are already using it. See, e.g., Trs. of Dartmouth Coll. & Serv. Emps. Int’l Union, Local 560, N.L.R.B. No. 01-RC-325633, at 18 (Feb. 5, 2024).

34 the benefits that an intern might receive at an internship with the training received at the intern’s formal education program. Id. at 537–38. In comparison, interscholastic athletics are not part of any academic curriculum. Here, the plaintiffs go as far as alleging that the sports played are actually detrimental to their academic performance because athletic performance provides no academic benefits, they are frequently precluded from enrolling in hundreds of courses that conflict with their athletic obligations, and they are unable to declare their preferred majors.

We therefore hold that college athletes may be employees under the FLSA when they (a) perform services for another party, (b) “necessarily and primarily for the [other party’s] benefit,” Tenn. Coal, 321 U.S. at 598, (c) under that party’s control or right of control, id., and (d) in return for “express” or “implied” compensation or “in-kind benefits,” Tony & Susan Alamo Found., 471 U.S. at 301 (quotation omitted).61 If

61 Although FLSA cases involving patients at rehabilitation centers present entirely different factual circumstances, courts of appeal already weigh these same factors to determine whether a patient is performing work. See, eg., Klick v. Cenikor Found., 79 F.4th 433, 441–42 (5th Cir. 2023) (rejecting the Ninth Circuit Court of Appeal’s decision in Williams v. Strickland, 87 F.3d 1064 (9th Cir. 1996), and holding that the district court did not abuse its discretion in concluding that patients in a rehab program that worked at outside businesses were employees because they were promised in-kind benefits in the form of housing, food, medical care, and clothing); Fochtman v. Hendren Plastics, Inc., 47 F.4th 638, 646 (8th Cir. 2022) (holding that the patient-worker was not an employee because his circumstances were more akin to those described

35 so, the athlete in question may plainly fall within the meaning of “employee” as defined in 29 U.S.C. § 203(e)(1). Ultimately, the touchstone remains whether the cumulative circumstances of the relationship between the athlete and college or NCAA reveal an economic reality that is that of an employee- employer.

B. The “Frayed Tradition” of Amateurism is No Shield to FLSA Claims

Appellants argue that the history and tradition of amateurism is sufficient not only to remove college athletes from the general population of people whose FLSA employment status is routinely determined through the application of multifactor tests, but also compels dismissal of

in Portland Terminal than those in Susan Alamo); Vaughn v. Phoenix House N.Y., Inc., 957 F.3d 141, 144–45 (2d Cir. 2020) (holding that residential drug treatment patient was not an employee of the facility, even though he received “food, a place to live, therapy, vocational training, and jobs that kept him busy and off drugs” because he was the “primary beneficiary” of the relationship and not the facility); Acosta v. Cathedral Buffet, Inc., 887 F.3d 761, 767 (6th Cir. 2018) (holding that church volunteers “were not economically dependent upon” the alleged employer given that “[t]he volunteers neither expected nor received any wages or in-kind benefits in exchange for their service.”); Williams, 87 F.3d at 1064 (holding that a participant in six-month rehabilitation program that included work was not an employee of the Salvation Army because the relationship did not contemplate “in-kind benefits” in exchange for work).

36 this suit. We disagree. Although the Supreme Court remarked in Board of Regents dicta that “[t]he NCAA plays a critical role in the maintenance of a revered tradition of amateurism in college sports,” 468 U.S. at 120, it has since unanimously clarified that Board of Regents did not expressly approve of every NCAA limit on athlete compensation or foreclose “any meaningful review of those limits today.” Alston, 141 S. Ct. at 2157; see also id. at 2167 (Kavanaugh, J., concurring) (“The Court makes clear that the decades-old ‘stray comments’ about college sports and amateurism made in [Board of Regents] were dicta and have no bearing on whether the NCAA’s current compensation rules are lawful.”) (citation omitted). The NCAA’s athlete compensation rules, after all, were not even at issue in Board of Regents. That case instead concerned the NCAA’s attempt to exercise monopoly control over television broadcast agreements.

Responding to an argument like the one that the Appellants make here, the Supreme Court in Alston noted that the NCAA had “not adopted any consistent definition” of amateurism and acknowledged that the organization’s “rules and restrictions on compensation have shifted markedly over time,” which further undermined the NCAA’s reliance on the concept. Id. at 2163 (citations omitted). The Court’s disapproval of amateurism as a legal defense was only strengthened by a point made by Justice Kavanaugh in concurrence that we now adopt: the argument “that colleges may decline to pay student athletes because the defining feature of college sports . . . is that the student athletes are not paid,” is circular, unpersuasive, and increasingly untrue. Id. at 2167.

Nevertheless, this is the argument Appellants most heavily rely upon to characterize the economic realities of the college

37 athlete’s alleged employment relationship. They argue that the District Court should have adopted the Court of Appeals for the Seventh Circuit’s reasoning in Berger, 843 F.3d at 291, where the Court declined to apply the Glatt test to determine whether a group of track and field athletes from the University of Pennsylvania were employees under the FLSA. In the eyes of both the Berger Court and the Appellants, no multifactor test is appropriate. Id. Rather, a general economic realities analysis that centers on amateurism and college athletes’ historical lack of bargaining power should be used. Id. In other words, Appellants ask us to elevate amateurism to a quasi-legal status in a way the Supreme Court has already rebuffed.

The Seventh Circuit Court of Appeals did indeed decline to apply a multifactor test because doing so “‘fail[ed] to capture the true nature of the relationship’ between [the] athletes and their schools and [was] not a ‘helpful guide.’” Id. (quoting Vanskike v. Peters, 974 F.2d 806, 807 (7th Cir. 1992)). Instead, it concluded that the “longstanding tradition [of amateurism] defines the economic reality of the relationship between [] athletes and their schools,” and held that existing multifactor tests could not adequately account for this tradition. Id. A “more flexible standard” was needed. Id. Ultimately, the Court held that college athletes were not employees entitled to minimum wage under the FLSA because their “amateur” status made it such that their “‘play’ is not ‘work.’” Id. at 293.

To reach its conclusion, the Berger Court relied on its own precedent in Vanskike, which considered whether incarcerated people had any rights under the FLSA. Id. at 291. In Vanskike, the Court similarly declined to use a multifactor test because any test would fail to account for what it called a “free labor situation” in the prison context. 974 F.2d at 809. This

38 “situation” exists in prisons because the Thirteenth Amendment permits involuntary servitude, meaning that the work incarcerated people perform is not based on voluntary employment relationships. Id. at 809–10. The Vanskike Court also pointed out that some factors typically found in FLSA multifactor tests could not logically be applied in the prison context. 974 F.2d at 809. For example, one common factor among FLSA tests is a consideration of the amount of control the employer has over the worker. Id. Given that prisons have almost complete control over prisoners’ lives, the Vanskike Court reasoned that such control was incidental to the workers’ custodial status. Id.

We disagree with our sister circuit court’s comparison of college athletes to prisoners and refuse to equate a prisoner’s involuntary servitude, as authorized by the Thirteenth Amendment, to “the long-standing tradition” of amateurism in college athletics. Berger, 843 F.3d at 291. Nor are we the only ones. See, e.g., Dawson v. NCAA, 932 F.3d 905, 908 n.2 (9th Cir. 2019) (“We do not adopt Berger’s analytical premises nor its rationales.”). But, in a limited sense, we agree that existing multifactor tests are inadequate when applied to the college athlete. As noted above, we believe that such tests either improperly assume that the alleged employee engages in compensable work or account for factors not relevant to college athletics.

In sum, for the purposes of the FLSA, we will not use a “frayed tradition” of amateurism with such dubious history to define the economic reality of athletes’ relationships to their schools. Berger, 843 F.3d at 294 (Hamilton, J., concurring). Instead, we believe that the amateurism that Judge Hamilton calls into question in his “note of caution” highlights the need

39 for an economic realities framework that distinguishes college athletes who “play” their sports for predominantly recreational or noncommercial reasons from those whose play crosses the legal line into work protected by the FLSA. Id. (“I am less confident, however, that our reasoning should extend to students who receive athletic scholarships to participate in so- called revenue sports like Division I men’s basketball and FBS football.”). Accordingly, we also hold that college athletes cannot be barred as a matter of law from asserting FLSA claims simply by virtue of a “revered tradition of amateurism” in D-I athletics. Board of Regents, 468 U.S. at 120.

III. CONCLUSION

In light of the foregoing, we will vacate the District Court’s order, remand for further proceedings in compliance with this opinion, and direct the District Court to grant leave to amend.62

62 Our colleague’s concurring opinion points out perceived problems with our analysis, but those are inherent to the unique nature of the legal question presented. And while he is correct that factual issues remain unresolved, the District Court must know what law to apply to the facts it ultimately determines upon remand. Accordingly, nothing would be accomplished but delay if we were to await resolution of every underlying factual dispute before deciding what law the trial court must apply.

40 PORTER, Circuit Judge, concurring in the judgment.

I write separately to explain why I concur only in the judgment.

I. We should decline to assert interlocutory jurisdiction.

A. The question presented is too fact- intensive.

Respectfully, we should not have accepted this interlocutory appeal. 28 U.S.C. § 1292(b) allows us to assert appellate jurisdiction over a non-final “order involv[ing] a controlling question of law” if doing so would “materially advance the ultimate termination of the litigation.” Section 1292(b) is thus reserved for dispositive legal questions. It is not suitable for reviewing mixed questions of law and fact. See Resol. Tr. Corp. v. Cityfed Fin. Corp., 57 F.3d 1231, 1236 n.6 (3d Cir. 1995) (declining to decide an issue certified as part of an interlocutory appeal under § 1292(b) because “the fact- intensive nature of the law in this area” required “greater factual development” by the district court), vacated on other grounds by Atherton v. F.D.I.C., 519 U.S. 213 (1997); Link v. Mercedes-Benz of N. Am., Inc., 550 F.2d 860, 863 (3d Cir. 1977) (“[Section] 1292(b) is not designed for review of factual matters but addresses itself to a ‘controlling question of law.’”).1 That is particularly true where, as here, the facts are

1 At least eight of our sister circuits have acknowledged the limitation of 28 U.S.C. § 1292(b) to pure questions of law.

1 See Nice v. L-3 Commc'ns Vertex Aerospace LLC, 885 F.3d 1308, 1313 (11th Cir. 2018) (dismissing appeal and vacating order under § 1292(b) where a question was “far from being one of pure law”); Steering Comm. v. United States, 6 F.3d 572, 575 (9th Cir. 1993) (recognizing a general rule forbidding “mixed questions of law and fact” on § 1292(b) review but finding a narrow exception for liability in multidistrict litigation); Ahrenholz v. Bd. of Tr. of Univ. of Ill., 219 F.3d 674, 676 (7th Cir. 2000) (defining “question of law” to be “the meaning of a statutory or constitutional provision, regulation, or common law doctrine”); Clark-Dietz & Assocs.-Eng’s, Inc. v. Basic Const. Co., 702 F.2d 67, 69 (5th Cir. 1983) (denying § 1292(b) review where “most [of the questions] appear to be merely fact-review questions” and “[e]ven those . . . that are legal may be foreclosed by the fact findings of the district court”); United States ex rel. Michaels v. Agape Senior Cmty., Inc., 848 F.3d 330, 341 (4th Cir. 2017) (rejecting an appeal as not presenting a “pure question of law”); Foster Wheeler Energy Corp. v. Metro. Knox Solid Waste Auth., Inc., 970 F.2d 199, 202 (6th Cir. 1992) (“pure questions of law”); Harriscom Svenska AB v. Harris Corp., 947 F.2d 627, 631 (2d Cir. 1991) (“Section 1292(b) . . . allows certification only of questions of law. Where, as here, the controlling issues are questions of fact[.]” (cleaned up)); Pittway Corp. v. Fyrnetics, Inc., 1993 WL 452621, 9 F.3d 977, at *1 (Fed. Cir. 1993) (Section 1292(b) “contemplates review of pure questions of law” where “‘the order involve[s] a clear-cut question of law against a background of determined and immutable facts.’” (quoting 9 James W. Moore et al., Moore’s Federal Practice ¶ 110.22[2] (2d ed. 1993)) (unpublished).

2 rapidly changing and highly disparate among putative class members. See In re Tutu Wells Contamination Litig., 120 F.3d 368, 407 (3d Cir. 1997) (discouraging § 1292(b) certification where the issues involved “may change as matters proceed before the district court”).

Even if this appeal were limited to just the six named plaintiffs, it would be fact-intensive. Determining whether nearly 200,000 Division I student-athletes playing on nearly 6,700 teams can be employees under the Fair Labor Standards Act (FLSA) requires us to assess countless facts. Maj. Op. at 23–24. On an interlocutory appeal from the denial of a Rule 12(b)(6) motion, that fact-bound exercise exceeds the scope and purpose of § 1292(b).

Indeed, almost no question is as fact-intensive as determining employee status under the FLSA. See Burrell v. Staff, 60 F.4th 25, 43 (3d Cir. 2023) (“FLSA coverage is a highly factual inquiry[.]”). We have said that it is “a mixed question of fact and law.” Verma v. 3001 Castor, Inc., 937 F.3d 221, 229 (3d Cir. 2019). But that is an understatement because, as the Supreme Court has cautioned, “[m]ixed questions are not all alike.” U.S. Bank Nat. Ass’n v. Vill. at Lakeridge, LLC, 138 S. Ct. 960, 967 (2018). Some mixed questions involve the articulation of a “pristine legal standard” or the development of “auxiliary legal principles” from a “broad legal standard.” Id. Others “immerse courts in case-specific factual issues— compelling them to marshal and weigh evidence[.]” Id. When considering mixed questions, we must ask whether “it entails primarily legal or factual work.” Id. If the mixed question involves primarily factual work, appellate courts should defer to the court that is most adept at clarifying and resolving factual disputes. Id. at 968.

3 Here, it is not even close: the mixed question is overwhelmingly factual. The majority’s test posits four factual questions while emphasizing that the “[u]ltimate[] touchstone” is the “economic reality” of “the cumulative circumstances . . . between the athlete and college or NCAA[.]” Maj. Op. at 36. And where, as here, the inquiry is essentially factual or presents conflicting inferences that can be drawn from undisputed facts, the ultimate question must be resolved by the fact-finder. See Razak v. Uber Techs., Inc., 951 F.3d 137, 145 (3d Cir. 2020) (reversing summary judgment because question of employee status under FLSA must go to a fact-finder); Verma, 937 F.3d at 229 (same).

The majority does not return this interlocutory appeal to the District Court with a case-dispositive legal rule. This lack of a clear rule is unsurprising. The question presented2 necessarily invites finding, weighing, and balancing a multitude of as-yet undeveloped facts that will vary widely across many thousands of student-athletes, teams, sports, colleges, and universities. In that sense, it is like the question presented in Link, 550 F.2d at 861 (whether an antitrust action with potentially 300,000 claimants is manageable as a class action), which we declined to answer because it was fact- dependent, id. at 863. Similarly, a yes-or-no answer is impossible under the majority’s test, inevitably leading to over- or under-inclusion. Fashioning a pure legal rule is impossible

2 “Whether NCAA Division I student-athletes can be employees of the colleges and universities they attend for purposes of the Fair Labor Standards Act solely by virtue of their participation in interscholastic athletics.” App. 35.

4 here, and I do not fault the majority for failing to provide one. But we should have refused to certify the order and even now we should decline to answer this intensely fact-bound question.

B. The Plaintiffs’ pleading and the parties’ briefing makes this case even less suitable for interlocutory review.

Plaintiffs’ manner of pleading and the parties’ briefing compounds these problems by failing to discuss Plaintiffs’ own experiences and the heterogeneity that exists between different types of college athletes. The economic relationship between the quarterback of a Power Four conference football team and his university is presumably different than the relationship between, say, a Conference USA school and a member of its bowling team. But the First Amended Complaint (FAC) treats all student-athletes the same.

Collective-action plaintiffs must plead plausible claims at a sufficient level of specificity. For example, in Davis v. Abington Mem’l Hosp., we held that the plaintiffs’ failure to allege a specific week in which they worked more than 40 hours made their claims “insufficient.” 765 F.3d 236, 242–43 (3d Cir. 2014). Merely alleging that they “typically” or “frequently” did so, without more, was not enough. Id. (internal quotation marks omitted). Though exact dates and times for every week aren’t necessary, pleadings lacking any specific instances do not establish a plausible claim under Rule 8. Id. at 243.

And in Karlo v. Pittsburgh Glass Works, LLC, we discussed the “two-step certification process” employed in FLSA collective actions. 849 F.3d 61, 85 (3d Cir. 2017). First,

5 under conditional certification, named plaintiffs must make a “modest factual showing to demonstrate a factual nexus between the manner in which the employer’s alleged policy affected him or her and the manner in which it affected the proposed collective action members.” Id. (internal quotation marks and quoted source omitted) (emphasis added). Second, under final certification, “[t]he named plaintiffs bear the burden of showing that the opt-in plaintiffs are similarly situated to them for FLSA purposes.” Id. (internal quotation marks and quoted source omitted).

Here, the named plaintiffs played football at Villanova University (Johnson), swimming and diving at Fordham University (Kerkeles), baseball at Fordham University (Labella), tennis at Sacred Heart University (Ruiz), soccer at Cornell University (Willebeek-Lemair), and tennis at Lafayette College (Cooke). Rather than describing their own, individual circumstances, Plaintiffs’ FAC describes the alleged experiences of student-athletes generally. The FAC is interspersed with anecdotes about non-party student-athletes and their schools, but it says little about the individual Plaintiffs and their universities. See App. 107–12. It sweepingly refers to “NCAA D1 member schools, like Villanova,” but fails to describe lead-Plaintiff Ralph Johnson’s experiences at Villanova except to allege that his football- related responsibilities could have limited his ability to sign up for certain classes. App. 86 (FAC ¶ 93).

The other named Plaintiffs’ alleged experiences are even less particularized: their names appear almost solely in paragraphs reciting jurisdictional requisites, class definition, and FLSA collective allegations. App. 65–66 (FAC ¶¶ 19–24). Plaintiffs’ generic pleading masks an underlying heterogeneity

6 among them and the thousands of student-athletes playing different sports at different universities across the country. At this interlocutory stage, the prudent course is for us to abstain.

II. The majority’s historical and sociological survey is inappropriate.

This appeal is taken from the District Court’s denial of a motion to dismiss under Fed. R. Civ. P. 12(b)(6). In this posture, we, like the District Court, may consider “only the allegations contained in the complaint, exhibits attached to the complaint, and matters of public record.” Doe v. Univ. of Sciences, 961 F.3d 203, 208 (3d Cir. 2020) (internal quotation marks and quoted source omitted).

The majority opinion, however, opens with a seventeen- page discussion of the history, sociology, and economics of intercollegiate sports. Maj. Op. at 6–23. It includes fifty-five references to news articles, books, journal articles, and online sources—none of which is in the appellate record. The factual assertions in these sources and the inferences drawn therefrom have not been litigated and lack the benefit of party presentation. Although the entire section is dicta, I find it objectionable.

III. Traditional multifactor tests and the amateurism principle are unhelpful guides.

I agree with the majority’s rejection of the tests in Glatt v. Fox Searchlight Pictures, Inc., 811 F.3d 528 (2d Cir. 2016), Donovan v. DialAmerica Mktg., Inc., 757 F.2d 1376 (3d Cir. 1985), and other cases distinguishing employee workers from independent-contractor workers or interns. In those cases, the

7 analysis began with the undisputed premise that individuals performed “work” that was necessary and integral to their employer’s business. See Donovan, 757 F.2d at 1385. Here, the critical antecedent question is whether student-athletes are “workers” providing “services” to an employer. The Glatt and Donovan multi-factor tests do not even attempt to answer that question, so they are inapposite.

I do not question the existence or virtue of amateurism in college athletics.3 But a combination of market forces, decades-spanning behavior of the NCAA and some Division I teams and athletes, and the Supreme Court’s relevant antitrust decisions4 have enervated that concept, at least for some student-athletes. Whatever legal force amateurism once had in the Division I context, it is now insufficient to decide cases like this one. Instead, we must look to the language and rules provided by statute and Supreme Court decisions.

IV. Play is not work.

The FLSA applies only to “employees” who perform “work” for an “employer.” Walling v. Portland Terminal Co., 330 U.S. 148, 150–51 (1947). So an obvious starting point is to ask whether a student-athlete may play her chosen sport because she wants to play, not to work primarily for her university’s benefit. Play is arguably a basic human good that

3 See, e.g., Lincoln Allison, Amateurism in Sport (2001). 4 See NCAA v. Alston, 141 S. Ct. 2141, 2157–58 (2021).

8 many pursue for its own sake.5 It is not work, even though it may involve sustained, regulated, physical, or intellectual exertion and combine with other goals such as competition, teamwork, fitness, or personal glory.6

If a student-athlete participating in an NCAA- sponsored sport—fencing, water polo, rifle, track and field, golf, beach volleyball, or skiing, for example—is engaged in play rather than work, then none of the commonly used tests will be useful because the FLSA simply does not apply.

The FLSA does not define “work.” The Supreme Court interprets it as denoting “physical or mental exertion (whether burdensome or not) controlled or required by the employer and performed necessarily and primarily for the benefit of the employer and his business.” IBP, Inc. v. Alvarez, 546 U.S. 21, 25 (2005) (quoting Tennessee Coal, Iron & R. Co. v. Muscoda Local No. 123, 321 US. 590, 598 (1944)). But the dictionary from which the Supreme Court derived that definition specifically distinguishes work “from something undertaken primarily for pleasure, sport, or immediate gratification, or as merely incidental to other activities[.]” Tennessee Coal, 321 U.S. at 598 n.11 (quoting Websters Int’l Dictionary (2d ed. unabridged)) (emphasis added). So even the Court’s broad definition of work does not encompass play or sport.

5 See, e.g., John Finnis, Natural Law and Natural Rights 87, 140–42 (1980). 6 See, e.g., Anthony J. Celano, Play and the Theory of Basic Human Goods, 28 Am. Phil. Q. 137, 139–40 (1991).

9 Division I student-athletes perform at the top of their highly competitive sports, and some are world-class. They certainly exert themselves physically and mentally. In colloquial terms, they “work out,” just as lesser athletes and fitness buffs do. But not all exertion is “work” for purposes of the FLSA.

In Walling, the Court explained that the FLSA does not cover a person who, “without promise or expectation of compensation, but solely for his personal purpose or pleasure, work[s] in activities carried on by other persons either for their pleasure or profit.” 330 U.S. at 152. The Department of Labor makes the same distinction. See United States Dep’t of Labor Wage and Hour Division, Field Operations Handbook § 10b03(e) (activity of college students participating in interscholastic athletics primarily for their own benefit as part of the educational opportunities provided to the students by the school is not “work”).7

Plaintiffs alleged that their college athletic experiences constitute work. App. 61, 63 (FAC ¶¶ 1, 9). But that allegation has not been proven, and unlike in the independent-contractor and intern cases, it is not a given here. Even at the Rule 12(b)(6) stage, we “are not bound to accept as true a legal conclusion couched as a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544, 555 (2007) (internal quotation marks and quoted source omitted).

7 Available at https://www.dol.gov/sites/dolgov/ files/WHD/legacy/files/FOH_Ch10.pdf (last visited June 18, 2024).

10 The majority opinion helpfully acknowledges this nettlesome issue. Maj. Op. at 30. But in my view, its definitional test8 does not adequately probe the distinction between play and work, nor explain how district courts should do so. In the following sections, I will explain my other scruples about the majority’s proposed test and offer some affirmative thoughts.

V. The economic-reality test continues to apply in FLSA cases.

Congress and the Supreme Court have created a patchwork of tests for determining employee status under federal labor and employment laws. Initially, the Court used an “economic realities” test in cases applying the National Labor Relations Act (NLRA), FLSA, and Social Security Act (SSA). See NLRB v. Hearst Publications, 322 U.S. 111 (1944) (NLRA); Walling, 330 U.S. 148 (1947) (FLSA); United States v. Silk, 331 U.S. 704 (1947) (SSA). Almost immediately, Congress passed a joint resolution (the “Gearhart Resolution”) rejecting the economic-realities test for the NLRA and SSA, and reiterating its intention that employee status under those statutes should be determined by traditional agency law principles. 62 Stat. 438 (1948). But Congress did not similarly amend the FLSA.

Since then, the Supreme Court has applied the common- law definition of “employee” to federal statutes that do not define “employee” or define it circularly. See Cmty. For Create Non-Violence v. Reid, 490 U.S. 730 (1989) (interpreting the

8 See Maj. Op. at 34, 40.

11 Copyright Act of 1976); Nationwide Mut. Ins. Co. v. Darden, 503 U.S. 318 (1992) (interpreting the Employee Retirement Income Security Act of 1974); Clackamas Gastroenterology Assocs., P.C. v. Wells, 538 U.S. 440 (2003) (interpreting the Americans with Disabilities Act of 1990).

But the Court has continued to apply the economic- reality test in FLSA cases. See Walling, 330 U.S. at 150–51 (rejecting “common law employee categories” and considering economic reality of the parties’ relationship); Rutherford Food Corp. v. McComb, 331 U.S. 722, 729 (1947); Goldberg v. Whitaker House Co-op, Inc., 366 U.S. 28, 33 (1961); Tony & Susan Alamo Found. v. Sec’y of Labor, 471 U.S. 290, 301 (1985). So have we. See Burrell, 60 F.4th at 43; Razak, 951 F.3d at 144; Verma, 937 F.3d at 230; Thompson v. Real Estate Mortg. Network, 748 F.3d 142, 148 (3d Cir. 2014); In re Enterprise Rent-A-Car Wage & Hour Emp’t Prac. Litig., 683 F.3d 462, 467–68 (3d Cir. 2012); Haybarger v. Lawrence Cnty. Adult Prob. & Parole, 667 F.3d 408, 417 n.8 (3d Cir. 2012) (“Since [1961], we and other circuits have applied the economic reality test to decide whether entities qualify as employers under the FLSA.”); Reich v. Gateway Press, Inc., 13 F.3d 685, 695 n.12 (3d Cir. 1994) (“This court and others have often applied an ‘economic reality’ test when interpreting the FLSA.”); E.E.O.C. v. Zippo Mfg. Co. 713 F.2d 32, 36 (3d Cir. 1983) (“The ‘economic realities’ standard is generally used in cases involving the Fair Labor Standards Act (‘FLSA’).”); Donovan, 757 F.2d at 1383–84.

So do other courts of appeals. See, e.g., Adams v. Palm Beach Cnty., 94 F.4th 1334, 1337 (11th Cir. 2024) (“The touchstone of the employee inquiry is one of ‘economic reality.’”) (quoted source omitted); Klick v. Cenikor Found.,

12 94 F.4th 362, 369 (5th Cir. 2024) (“The ultimate determination turns on the ‘economic reality’ of the relationship between the parties involved.”); Vallone v. CIS Solutions Group, LLC, 9 F.4th 861, 866 (8th Cir. 2021) (“The test of employment is one of economic reality.”); Mendel v. City of Gibraltar, 727 F.3d 565, 569 (6th Cir. 2013) (“The Supreme Court has adopted an ‘economic reality’ test to determine whether an individual is an employee under the FLSA.”).

Other courts wrestling with the FLSA-employee question in the specific context of student-athletes have also applied an economic-reality test. See Dawson v. Nat’l Collegiate Athletic Ass’n, 932 F.3d 905, 909 (9th Cir. 2019); Berger v. Nat’l Collegiate Athletic Ass’n, 843 F.3d 285, 290 (7th Cir. 2016).

The FLSA shares the Employee Retirement Income Security Act’s (“ERISA’s”) “completely circular” definition of “employee.” In Darden, the Supreme Court held that ERISA’s use of the term “employee” “incorporate[s] traditional agency law criteria.” 503 U.S. at 319. So Darden arguably supports abandoning the economic-reality test in favor of “common-law agency doctrine” in FLSA cases too. Darden, 503 U.S. at 323.

However, doing so would abrogate Walling’s clear rejection of “common law employee categories” and decades of precedent using the economic-reality standard in FLSA cases. Only the Supreme Court can abrogate its precedents. State Oil Co. v. Khan, 522 U.S. 3, 20 (1997) (“[I]t is this Court’s prerogative alone to overrule one of its precedents.”). So while I appreciate the majority’s attempt to fashion a test using common-law agency principles from NLRA cases, Maj.

13 Op. at 31–34, I respectfully decline to join that analysis without clearer direction from the Supreme Court. Instead, I think that the proper test for this case is to determine the economic reality of the parties’ relationships considering the circumstances of the whole activity.9

VI. The majority’s test raises but does not answer some important questions.

In addition to my doubts about relying on common-law agency principles in the FLSA context, I find the majority’s four-part test wanting in some respects.

a

The test begins by asking whether the student-athlete performs “services” for his college or university. The majority does not define “services,” but its test largely tracks the

9 Particularly in the independent-contractor context, courts have sometimes attempted to discern economic reality by applying a multifactor test such as the one we used in Donovan. As noted above and in the majority opinion, such tests are inapposite and unhelpful here. There is no “single ‘economic realities’ test consisting of uniform factors” that apply to every case. Brown v. New York City Dept. of Educ., 755 F.3d 154, 167 (2d Cir. 2014). Because the plaintiffs in this case are “sui generis,” Maj. Op. at 29, the economic-reality question does not turn on “isolated factors” but can only be answered by considering “the circumstances of the whole activity.” Rutherford Food Corp., 331 U.S. at 730.

14 Restatement (Third) of Agency’s definition of servant. That definition has not materially changed since the first Restatement of Agency (1933):

A servant is a person employed to perform services for another in his affairs and who with respect to the physical conduct in the performance of the services is subject to the other’s control or right to control.

Restatement (Third) of Agency § 220 (2006).

Webster’s New International Dictionary 2288 (2d ed. 1950) defines “service” as the “[p]erformance of labor for the benefit of another or at another’s command.” See also Black’s Law Dictionary (11th ed. 2019) (“labor performed in the interest or under the direction of others”); Webster’s New International Dictionary (3d ed. 1993) (“the performance of work commanded or paid for by another”).

So the first part of the majority’s test immediately raises—but does not clarify—the critical distinction between “service,” “labor,” or “work,” as distinguished from play or sport. In a general sense, student-athletes serve the teams for which they play. But that is true of anyone who has ever played on a team: each player contributes her measure of skill and effort—her services, as it were—for the good of the entire team. That’s the whole point of team sports. But one’s contribution in the service of teamwork does not necessarily create an employment relationship.

15 b

The second part of the majority’s test asks whether the student-athlete’s team participation is necessarily and primarily for the university’s benefit. Again, there is a sense in which student-athletes obviously play for the benefit of their university’s team. The NCAA Transfer Portal offers student- athletes the flexibility each year to choose where they wish to play. Once the student-athlete chooses, he enrolls in his chosen college or university and becomes a member of its team. But that has little or nothing to do with employment status; it’s a basic correlate of matriculation and team membership. Naturally, the student’s athletic prowess benefits his chosen team and university because that is how team sports operate. Division I student-athletes play or provide athletic “services” for the benefit of their team just as Division II, Division III, and high school athletes play or provide athletic “services” for the benefit of their respective teams. But something more is required to convert the majority’s university-as-beneficiary factor into a useful indicia of employment.

For example, in Alamo Foundation, the religious foundation doubled as a commercial enterprise through its operation of profit-seeking “businesses serv[ing] the general public in competition with ordinary commercial enterprises.” 471 U.S. at 299. The enterprise included “service stations, retail clothing and grocery outlets, hog farms, roofing and electrical construction companies, a recordkeeping company, a motel, and companies engaged in the production and distribution of candy.” Id. at 292. The putative volunteers’ work for those “ordinary commercial businesses” produced economic benefits for the foundation, so it was appropriate to characterize them as employees. Id. at 298. Here, the factfinder

16 should consider whether a university’s sports team is (a) economically comparable to one of the Alamo Foundation’s profit-seeking businesses, or (b) essentially an extra-curricular activity creating at best indirect and attenuated economic benefit for the university.

c

The third factor of the majority’s test asks whether the student-athlete plays under the university’s control or right of control. This principle of agency law is not particularly helpful in the context of intercollegiate sports. High school students do not set their own rules for recruitment and college students do not set their own rules for eligibility and participation. Such autonomy would invite chaos, undermine teamwork, and destroy competition. Because team sports are collective actions, all teams have coaches and administrators that evaluate players, assemble rosters, allocate playing time, make personnel changes, determine strategy, call plays, set practice and game schedules, arrange transportation, and so forth. The players do not act independently of each other and the coaches because, again, team sports are collective actions requiring significant direction and coordination. The control or right-of- control factor does not go very far to distinguish Division I athletes from Division II athletes, Division III athletes, or other organized team-sport participants.

d

The fourth factor of the majority’s test asks whether the student-athlete provides services “in return for ‘express’ or ‘implied’ compensation of ‘in-kind benefits.’” I agree that this factor is relevant to the work/play and employee/non-employee

17 distinctions. The Supreme Court declared so in Walling, 330 U.S. at 152, and Alamo Found., 471 U.S. at 301–02.

In Alamo Foundation, the Court held that self- proclaimed volunteers who were “entirely dependent upon the Foundation for long periods” were actually employees because they accepted “in-kind benefits . . . in exchange for their services.” Id. at 293, 301 (internal quotation marks and quoted source omitted). The benefits, according to the Court, amounted to “wages in another form.” Id. at 301. So even atypical modes of compensation can create employment relationships under the FLSA. What matters is the existence of some express or implied compensation arrangement and economic dependence. Theoretically, this approach might allow the would-be employer to avoid FLSA coverage simply by refusing to pay would-be employees as a matter of policy. See Alston, 141 S. Ct. at 2167 (Kavanaugh, J., concurring). But if the Supreme Court’s compensation rule is enforced, such avoidance tactics will be futile. If universities offer in-kind benefits—such as, perhaps, scholarships10 that can be

10 The significant grant-in-aid and in-kind benefits that Division I schools have long given through athletic scholarships are excluded from income under IRS rules. Rev. Rul. 77-263, 1977-2 C.B. 47. That raises a legislative question rather than an adjudicative question: what effect federal tax policy has, or should have, on the economic-reality analysis.

18 cancelled mid-year if an athlete quits her team11—they must navigate the rule of Alamo Foundation.12

What if an alleged employment relationship is voluntary and truly implicates no compensation arrangement or wage- like benefits for work in a commercial setting? In that case, the purported employee might be a “person who, without promise or expectation of compensation, but solely for his personal purpose or pleasure, work[s] in activities carried on by other persons either for their pleasure or profit.” Walling, 330 U.S. at 152. The FLSA was “obviously not intended” to classify all such persons as employees, “[o]therwise, all students would be employees of the school or college they attended.” Id.

11 The NCAA Division I bylaws allow member institutions to reduce or cancel athletic grants-in-aid during the period of the award if the student “withdraw[s] from their sport.” Suppl. Br. for Appellants Cornell University, Villanova University, Fordham University, Lafayette College, NCAA, and Sacred Heart University at 3 (citing NCAA Div. I Bylaws § 15.3.4.2(d)). Plaintiffs do not argue that they received aid that amounted to compensation under Alamo Foundation. To do so, they would have to offer pleadings as to their own personal experiences and circumstances, not a sociological survey of the life of a representative student-athlete at a typical Division I university. 12 Ivy League universities do not award athletic scholarships, further highlighting the heterogeneity among Division I conferences, schools, sports, teams, and student- athletes.

19 In Alamo Foundation, the disguised wages were paid for work performed in the foundation’s various commercial businesses. How those businesses compare to any given sports team at any given college is another knotty factual question. And although we may not consider facts that are not alleged in the FAC, the economic reality surrounding the compensation- bargain factor is in flux and will dramatically change even as the ink on this opinion is drying.13

According to the majority, “profit” has always influenced “college athletics.” Maj. Op. at 10. The majority emphasizes the enormous revenue that “college athletes” generate annually. Id. at 11. But revenue is not profit. And the majority’s historical discussion diminishes the role of so-called nonrevenue generating sports at colleges and universities. In this pre-discovery posture, however, my general understanding is that for most student-athletes, the economic reality is that their athletic service, and their team’s existence, is revenue- negative. Football Bowl Subdivision (“FBS”) football and March Madness-level men’s basketball are spectacular exceptions because they attract lucrative television deals.

Compared to FBS schools, the revenue vs. nonrevenue issue is presumably even more pronounced in the smaller Division I Football Championship Subdivision (“FCS”) athletic programs. But the majority offers no guidance about

13 See, e.g., NCAA Agrees to Share Revenue With Athletes in Landmark $2.8 Billion Settlement, The Wall Street Journal (May 23, 2024), available at https://www.wsj.com/sports/basketball/ncaa-revenue-athletes- settlement-0b53306d (last visited June 20, 2024).

20 how courts or factfinders applying an economic-reality test should consider student-athlete participation in nonrevenue sports.14 Are they part of the “business” of a college or university? For that matter, are athletics—though obviously important for various reasons—incidental to the university’s business or essential to it? Does a college benefit from revenue- negative athletic programs? If so, how does that benefit differ from the cash produced by football, men’s basketball, or the profit-seeking businesses in Alamo Foundation? And how is the “economic reality” of a nonrevenue student-athlete’s relationship with his university different from that of a musician whose performing arts scholarship is conditioned on her time-consuming participation in a band or orchestra? Or from a member of the school’s competitive esports team who may also receive a scholarship? We cannot begin to answer

14 Obviously, I do not suggest that only workers in profitable companies can be employees under the FLSA. But in this “sui generis” context, Maj. Op. at 29, the distinction between revenue sports and nonrevenue sports may help to separate—as a matter of economic reality—those student- athletes who “work” for the tangible benefit of their university from those whose “play” confers no comparable benefit.

21 such questions in this interlocutory appeal.15 Nor can they be answered in gross. The answers will likely differ among individuals, teams, sports, and schools.

I tend to agree with Judge Hamilton’s intuition that the economic-reality question probably shakes out differently for FBS football players and March Madness-level men’s basketball players than it does for other student-athletes.16 See

15 We cannot declare as a matter of law which activities are essential or incidental to a particular university’s business, let alone all universities’ businesses. That question is both fact- intensive and value-laden. A few college sports would be considered big business by any standard—so much so that Power Four conferences are now considering private equity investments. On the other hand, as the Supreme Court has said in a different labor-law context, “[t]he ‘business’ of a university is education[.]” N.L.R.B. v. Yeshiva Univ., 444 U.S. 672, 688 (1980). If one’s favorite college team were to evaporate overnight, the institution would presumably continue to teach, research, publish, confer degrees, and perform the work that all universities have carried on for nearly a millennium. Economists, sociologists, accountants, historians, philosophers, professors, students, student-athletes, and sports fans might all offer different perspectives on the essential-or-incidental-question. This and other questions generated by this interlocutory appeal are more legislative than adjudicative. That is another reason for this Court to proceed slowly and cautiously, lest the courts be used to precipitate sweeping and essentially legislative changes. 16 Exceptions surely exist from time to time, such as the 2024 Women’s Final Four.

22 Berger v. NCAA, 843 F.3d 285, 294 (7th Cir. 2016) (Hamilton, J., concurring). See also NCAA v. Alston, 141 S. Ct. 2141, 2166–69 (2021) (Kavanaugh, J., concurring) (focusing on the “enormous sums of money” generated in college athletics and noting distinction between revenue and “nonrevenue-raising sports”); Agnew v. NCAA, 683 F.3d 328, 340–41 (7th Cir. 2012) (positing a relevant labor market, for purposes of the Sherman Act, consisting of “big-time college football programs”). That is a factual matter that the parties can develop in discovery. But any test that purports to gauge “economic reality” must be sensitive to the glaring difference between revenue generating and nonrevenue intercollegiate sports.

VII. The FLSA-employee test should account for longstanding precedent and existing law.

For over 65 years, courts across the country have determined that student-athletes do not qualify as employees of their universities. See Dawson, 932 F.3d at 911; Berger, 843 F.3d at 293; Kavanagh v. Trs. of Bos. Univ., 440 Mass. 195 (Mass. 2003); Korellas v. Ohio State Univ., 121 Ohio Misc.2d 16 (Ohio Ct. Cl. 2002); Shephard v. Loyola Marymount Univ., 102 Cal.App.4th 837 (Cal. Ct. App. 2002); Waldrep v. Texas Emps. Ins. Ass’n, 21 S.W.3d 692 (Tex. Ct. App. 2000); Townsend v. State of Cal., 191 Cal.App.3d 1530 (Cal. Ct. App. 1987); Rensing v. Ind. State Univ. Board of Trs., 444 N.E.2d 1170 (Ind. 1983); Coleman v. Western Mich. Univ., 125 Mich. App. 35 (Mich. Ct. App. 1983); State Comp. Ins. Fund v. Indus. Comm’n, 135 Colo. 570 (Colo. 1957).

As the Supreme Court emphasized in the antitrust context, changing market realities can throw such precedent into doubt. See Alston, 141 S. Ct. at 2158. But in that event, our

23 test for employee status under the FLSA should isolate the changed facts and market realities that distinguish the venerable line of precedent. Again, that exercise may highlight the growth of a unique and robust labor market for FBS football and Division I basketball players. See, e.g., O’Bannon v. NCAA, 802 F.3d 1049, 1056 (9th Cir. 2018) (discussing antitrust market for FBS football and Division I men’s basketball).

This case also presents difficult collateral legal issues that should give us pause. For example, the related-statutes canon requires harmonious interpretation of statutes. See Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 252–55 (2012). Employee- employer relationships are governed by Title VII, among other things, but Title VII’s prohibition against employment discrimination because of sex sits uneasily with Title IX regulations and policy interpretations mandating equal “participation opportunities” (read, “participants”) between the two sexes. See, e.g., Michael E. Rosman, Gender Identity, Sports, and Affirmative Action, 53 St. Mary’s L. J. 1093, 1119– 39 (2022). FLSA employee status for student-athletes would also roil the percolating debate under Title IX over transgender athletes’ participation on opposite-sex teams because Title VII, which would apply to collegiate athletics if student-athletes have employee status under FLSA, prohibits employment discrimination on the basis of gender identity. See Bostock v. Clayton Cnty., 140 S. Ct. 1731 (2020). The notion that sports are integral to a university’s educational purpose, rather than employment programs themselves, is the basis for several tax- advantageous rules benefitting universities and student- athletes, such as unrelated business income tax, and the taxation of athletic scholarships. And our disposition of this

24 interlocutory appeal could impact student-athletes’ eligibility for federal student aid, state worker’s compensation regimes, student-athletes’ immigration status, and the employment status of students participating in other college-supervised extracurricular activities. These potentially disruptive collateral effects implicate many other statutory schemes, revealing the legislative rather than adjudicative nature of plaintiffs’ claims and providing another reason to slow down and proceed warily.

VIII. Conclusion

To the extent that the majority holds simply that it is factually possible for a Division I student-athlete to be an employee under the FLSA, I concur in that judgment.