Abstract
College football and basketball generate huge broadcast and attendance revenues. The NCAA, however, limits the compensation to athletes in these sports to cover, at most, the cost of attending college, which is well below the financial contribution to the schools of the top level college athletes. These compensation restrictions have recently come under attack in the courts in the O’Bannon and Jenkins cases, among others, as being anticompetitive. The NCAA has argued that its restrictions have procompetitive justifications. In this paper we set out first to evaluate the purported justification for the NCAA’s pay restrictions related to amateurism. We argue that the evidence from a number of Olympic and other now professional sports reveals little relationship between amateurism and demand. Rather than amateurism, we propose that the defining and distinguishing characteristic of college sports is the linkage to the universities. That athletes are also students is integral to the demand for college sports rather than whether these students get paid. We next examine the competitive balance justifications for restraints on payments to athletes. We summarize evidence that the current restrictions do not contribute to making college sports more competitively balanced. Nonetheless, we argue against going too far and lifting all restrictions on athlete compensation due to the inefficiency arising from “positional” competition, sometimes known as an “arms race.” We suggest instead that colleges and universities follow the models set up by professional sports for dealing with excessive spending, and that such models can reduce the anticompetitive effects from the NCAA’s salary restrictions. Such compensation schemes can take the form of player unions bargaining for payroll caps, tiered salary caps, and team revenue sharing.
I. Introduction
College football and basketball are big business. These sports generate huge broadcast and attendance revenues. For example, the top-earning public school in the nation, Texas A&M, earned $192.6 million in athletics revenues in 2014–2015. 1 Top-tier college coaches in these sports earn seven-figure compensation packages, and schools spend millions on training facilities, specialized residence halls for athletes, and stadiums. The NCAA, however, regulates the compensation to the primary “inputs” generating the revenues—the college athletes—and limits that compensation to cover at most the cost of attending college, which for the top-level college athletes is well below their financial contribution to the schools. This has generated tremendous controversy and has led to several lawsuits and many opinion pieces. 2
In this paper, we analyze whether there are legitimate efficiency justifications for the NCAA’s restrictions surrounding athlete compensation. In most markets, the courts deem price restrictions among competitors per se illegal. However, the Supreme Court has treated sports differently due to the collaborative nature of creating the product. 3 The Court applies a rule of reason standard that allows for a balancing of the negative effects of restrictions on competition with any efficiency considerations for the restrictions.
With regards to college level sports, in a landmark decision, NCAA v. Board of Regents of Oklahoma University (1984), the Supreme Court granted schools unusual cooperation control over essential prices; namely compensation to athletes. In that decision, the Court stated: “In order to preserve the character and quality of the ‘product,’ athletes must not be paid…” (emphasis added). 4 Despite the special considerations made for college sports, the restrictions on athlete compensation, as in any instance of price fixing, generate anticompetitive effects. For example, in the context of college sports, these restrictions generate a welfare transfer from athletes to colleges and likely a reduced supply of talented athletes. Moreover, they engender a wasteful amount of regulation necessary to ensure that the restrictions on athlete pay are not circumvented.
The NCAA’s limits on compensation of college athletes have been allowed despite these inefficiencies on the notion that it preserves amateurism in college athletics. The Supreme Court has been quite explicit that the amateur nature of college sports is an important feature of the product such that controls designed to foster it are given a wide berth: “It is reasonable to assume that most of the regulatory controls of the NCAA are a justifiable means of fostering competition among amateur athletic teams, and therefore procompetitive because they enhance public interest in intercollegiate athletics” (emphasis added). 5 However, the Court made such a declaration without any type of analysis. Moreover, these compensation restrictions have recently come under attack in the courts in the O’Bannon and Jenkins cases, among others. In this paper, we set out first to evaluate this purported procompetitive justification for the NCAA’s pay restrictions—the amateurism justification.
We conclude that the available evidence shows little relationship between amateurism and the demand for college sports. We base this conclusion on an examination of a number of Olympic and other sports that have transitioned from amateur to professional. We find no negative impact on consumer demand from the relaxation of amateur restrictions. Rather than amateurism, we argue that the distinguishing characteristic of college sports is the linkage to the universities. The association with schools generates a “college spirit” in fans; in effect, a kind of “tribalism” develops that results in strong loyalty to a school and its sports teams. The student nature of college athletes incites the remarkable demand for college sports rather than whether these students get paid. Fans desire to watch athletes with connections to the school they feel loyal to. Thus, we conclude that the essential “procompetitive” restrictions to support top level collegiate athletics concerns requirements that the athletes be students associated with specific colleges, not that they be amateurs. 6
Having reached this conclusion, we next examine alternative economic justifications for restraints on payments to athletes. These include the promotion of competitive balance, which is related to uncertainty in the outcome of games, and the avoidance of an inefficient form of competition for talent known as “positional” competition, or the “arms race” phenomenon. 7
With regards to competitive balance, we show that the current restrictions do not contribute to making college sports more competitive. In fact, college sports currently exhibit significant competitive imbalance despite the payment restrictions imposed by the NCAA. The same small group of teams ends up being successful year after year. Thus, we conclude that easing the pay restrictions would not impact negatively on competitive balance and might even help it.
However, we argue against going too far and lifting all restrictions on athlete compensation. The reason is based on a fundamental feature of sports competition—that the value of the product to a particular team’s fans depends largely on that team doing better than its rival. A team that does not play its best but wins tends to be preferred to one playing very well but losing. Unlike typical goods, the value of the product in sports is in part a function of doing better than the opponent.
Moreover, there is a differential value to finishing first in sports. The widely known and variously attributed saying “only the second-place finisher remembers who finished second” exemplifies this. 8 This differential value from winning gives rise to inefficient positional competition or an arms race. This type of competition is inefficient because of the presence of negative economic externalities. One team’s investment in improving its performance generates a negative effect in its competitors–namely a reduced chance of winning. Because teams do not internalize this negative externality, unbridled competition for college athletes will likely lead to excessive and inefficient compensation.
Thus, one potential procompetitive justification for the current pay restrictions is that they may help prevent inefficient positional competition for athletes. However, the existing restrictions cannot be justified on this basis. First, they do nothing to alleviate the existing arms race in expenditures on facilities and coaches, both of which are widely used to recruit athletes. Colleges and universities compete excessively for athletic talent via these non-wage avenues, despite the pay restrictions. The current compensation restrictions become merely a mechanism to transfer value from players to colleges and coaches. Second, limiting player salaries in the current manner–via a cap essentially the same for all players—precludes variations depending on the relative values of different athletes.
What are needed are more flexible alternatives to the current pay restrictions that alleviate all arms race inefficiencies. We recommend that colleges and universities follow the models for compensating athletes set up by professional sports. The virtue of such models is that they deal with excessive spending while avoiding the significant anticompetitive effects of the NCAA’s current salary restrictions. Such compensation schemes can take the form of player unions bargaining for payroll caps, tiered salary caps, and team revenue sharing.
As a concluding point we note that absent the amateurism justification, the antitrust laws would forbid schools from collectively determining what they deem to be competitive but not excessive compensation. The models from professional sports call for the adoption of more flexible pay restrictions in conjunction with the creation of an athletes’ union to collectively negotiate on behalf of collegiate athletes. Without such a union, the resulting outcomes would likely be inefficient and potentially illegal because the NCAA would be able to exploit its monopsony power over athletes.
II. The NCAA Compensation Restrictions in College Sports
The NCAA has rules in place that limit payments from universities to college athletes. 9 The limits are currently set at cost of attendance (COA) at so-called “Power Conference” schools 10 and Grants-in-Aid (GIA) at other schools. 11 The NCAA also forbids all compensation to players from any other source that bear some relationship to the participation in college sports, for example, video game royalties. 12 Finally, the NCAA imposes other restrictions on payments to players such as forbidding them from signing contracts with professional teams and earning salaries for doing sports in general. 13
However, agreements among competing teams in college sports have not been deemed per se anticompetitive. The most prominent decision highlighting the rule of reason approach is Board of Regents. 14 The NCAA restrictions in this case were (1) limits on total number of games that could be televised, (2) limits on the number of times any one team could appear on TV, and (3) a minimum aggregate payment to be made to the NCAA for the games. 15 These restrictions among competitors clearly and directly limited a form of output, and would typically be per se illegal. However, in the NCAA case, the court applied the “ancillary restraints” doctrine developed in 1979 in Broadcast Music to analyze the NCAA restrictions under the rule of reason. 16
The Court stated that “what is critical is that this case involves an industry in which horizontal restraints on competition are essential if the product is to be available at all.” The Court specifically noted that the NCAA’s restrictions “are designed to preserve amateurism.” 17
At a general level, the courts have treated collaboration among schools in the context of sports under the rule of reason because of the efficiency benefits from collaboration. For example, there are obvious benefits to agreeing about the rules of the athletic events, the timing of matches and playoff rules. Indeed, professional sports have elaborate agreements with regards to league operations. It should come as no surprise that the NCAA has imposed rules and restrictions for member colleges and universities with regards to playing rules, size of athletic teams, timing and length of the season, the size of the field, the type of equipment, and other areas. 18
Given the necessity for significant collaboration in order to produce athletic events and competition, it is not surprising that such collaboration might be extended to anticompetitive activities, most obviously, collusive behavior in the labor market for athletes. The collusive pay restrictions result in anticompetitive effects, including a transfer of wealth from players to universities; a reduced supply of players at the very top level; and a lower quality of competition. However, as mentioned above, procompetitive rationales have been offered for the NCAA pay restrictions. We will address those in a subsequent section.
III. Anticompetitive Effects from Compensation Restrictions
The pay restrictions invariably impose anticompetitive effects in the marketplace for athletes, which we summarize below. In the section that follows, we consider whether those negative effects might be offset by any associated benefits that could only be achieved with those restrictions in place.
First, the pay restrictions result in significant monitoring and enforcement costs. Participating schools employ significant resources to ensure compliance with NCAA regulations. For example, the number of employees involved with compliance at the individual schools can be as high as ten employees per school. 19 As colleges push the edge of what constitutes “pay,” extensive rules and monitoring effort expand. The rules and regulations defining what constitutes pay have resulted in an NCAA manual that is over 800 pages long. 20
Second, the pay limits on athletes shifts the locus of competition among schools for the best athletes to other, less efficient dimensions. These include coaches, stadiums, training facilities, and residence and dining halls. 21 This shift in spending on such complementary factors of production is socially wasteful for at least two reasons. As a recruiting tool, generalized spending on stadiums or training facilities may be less efficient than direct payments targeted at the specific athletes sought after by the school. This is analogous to the difference in advertising efficiency between targeted and untargeted online advertising, where there is a greater bang-for-the-buck for targeted ads. 22 A similar principle applies in recruiting tools used to attract college athletes. Moreover, by eliminating the most direct signal of an athlete’s value to a school, the school’s willingness to pay in the form of direct compensation, there is likely a misallocation of talent across schools.
Third, a significant literature exists demonstrating that many athletes participating in revenue sports are currently paid considerably less than their marginal revenue product (MRP). 23 For example, it has been estimated that the MRP for draft-level college football players was nearly a half a million dollars in 1995 and nearly triple that for men’s college basketball. 24 This is clearly far above the value of the allowed benefits to the players. 25 The widespread incidence of under-the-table payments for many top players corroborates the below-market compensation given to the athletes.
Fourth, because the supply curve for labor in college sports is upward sloping (due to alternatives to turn pro at some point during college career), the current restrictions lead to an inefficiently low number of players at the high end of quality. Many players who would have continued to play in college turn pro before graduation due to the higher compensation in pro sports. This is a substantial inefficiency because it deprives the sport of the substantial number of players that fans most want to watch. This premature exit reduces the absolute quality of college sports competition. Furthermore, the fact that college athletes turn pro before completing their degrees deprives the athletes of the longer-run benefits of education.
Thus, lifting the compensation restrictions may increase the quality of top-level college athletes and the level of competition. With an upward-sloping labor supply curve, the number of athletes that will choose to play college sports at the highest level will increase due to higher wages offered. Rather than turn pro and be middle-of-the-pack players, they can be stars in their college teams at a lower remuneration sacrifice. 26 The effect would be higher average quality teams and increased competitiveness of conferences.
Furthermore, even though there would likely be exit of some schools from top-level competition due to higher costs of paying athletes, these exiting schools would be expected to be those with less committed fan bases and entrenched athletic traditions. The schools that remain in top-level competition would be the ones that value athletics the most, and are likely the most successful. Hence, as the best athletes from schools dropping top-level sports migrate to the remaining schools, the overall competitiveness of top-level conferences would increase. That is, the best athletes will be clustered on fewer teams, increasing the quality of the output. (Those schools that exit top-level nonamateur competition might choose to play in amateur divisions, e.g., Divisions A–AAA.) Whether these effects increase overall demand depends on whether a potentially smaller set of top-level nonamateur teams is offset by the higher value of more competitive conferences.
A related effect is that those schools that continue to play top-level college sports will now play primarily against other teams with long-established sports traditions, with dedicated fan bases, and higher-quality players. This would create more competitive conferences, which in turn may lead to higher demand for college sports. There is also a positive externality here of increasing one team’s quality. Some studies have found that raising the level of the opposition raises a team’s home attendance. 27
In conclusion, the evidence suggests that the anticompetitive effect from the pay restrictions could be significant. In addition to the loss of players to the professional leagues, thereby depriving teams and fans of the highest level of competition in college sports, the pay restrictions impede the most powerful indicator—price—of how school values most highly particular players. 28 This likely results in a misallocation of players among the competing schools. The elimination of price as a tool for attracting talent also means that schools have to rely on less efficient recruiting instruments such as generalized spending on stadiums and facilities. This leads to overall more wasteful spending.
IV. Are the NCAA Compensation Restrictions Procompetitive Because They Maintain Amateurism?
Preliminaries
It is generally accepted that the top college sports conferences generate large profits from college basketball and football. 29 This is demonstrated by the revenues above costs earned by the top college conferences in those sports. For example, the average profit at the SEC was close to $17 million per school in 2014–2015. 30 This observation suggests that on the output side there might be relevant economic product markets for top-level college football and basketball. 31
In addition, the ability of the NCAA to impose a near ban on payments to the athletes participating in college football and basketball implies market power regarding this input. This in turn indicates that there might be relevant economic markets for the “purchase” of college-level football and basketball athletes. 32
The significance of relevant product markets consisting of top-level college football and basketball concerns whether and to what extent the restrictions on payments to the athletes differentiate these sports from professional alternatives. The question we address is whether these relevant markets would continue to exist absent the NCAA payment restrictions. If top-level college football and basketball become less differentiated (horizontally) relative to their pro counterparts because players are paid, then college and pro sports might be perceived instead as being vertically differentiated, with the former being inferior substitutes for the latter. 33 Alternatively, if the payment restrictions do not affect horizontal differentiation (because other aspects of college sports drive demand), then the welfare losses from removing the restrictions would be small or negligible.
Amateurism
In this section, we examine whether pay restrictions for college athletes are essential to differentiate college from professional sports. 34 We find little evidence that demand for top-level college sports would be adversely affected with more flexible compensation to college athletes, even if it means that athletes would no longer be perceived as amateur. We argue that it is not amateurism itself but the connection of the athletes to the institution and the student body that generates the unique demand for college sports.
Of course, college athletes are currently paid via full tuition scholarships and other support, and college football and basketball are highly professionalized in other ways including large TV revenues, high ticket prices and attendance revenue, and unbridled large salaries for coaches. We nonetheless accept that college sports are “amateur” because athletes’ compensation is restricted to cover no more than the cost of attending college.
The importance of amateurism on the demand for college sports has been asserted by a number of authors. For example, Kahn (2007) states that “[t]he potential effects on fan demand of amateurism…distinguish analyses of sports from those of other industries, where presumably consumers are not concerned about the contractual status of the industry’s workers or the relative status of the firms in the industry.” 35 Similarly, McKenzie and Sullivan (1987) argue that part of the appeal of college sports is their amateurism, in contrast to the professionalism of NFL and NBA. 36 Noll (2009) assumes that amateurism “adds to the success of college sports.” 37 Finally, as mentioned above, the Supreme Court in Board of Regents (1984) stated: “It is reasonable to assume that most of the regulatory controls of the NCAA are a justifiable means of fostering competition among amateur athletic teams, and therefore procompetitive because they enhance public interest in intercollegiate athletics” (emphasis added). 38
However, the proposition that amateurism defines and makes college football and basketball successful has not been deeply scrutinized. Noll writes that “NCAA officials believe that college sports is unique because it is an amateur sport played by students and has linkages to the traditions of a college.” However, even if we assume that it is important that college sports are “played by students [with] linkages to the traditions of a college,” the importance of amateurism in making college sports unique is not at all clear. 39
Kahn asserts the near tautology that “[i]t is theoretically possible that demand for amateur college sports is much higher than demand for college sports would be if players were treated as professionals.” He then notes that “no evidence exists on this point.” (emphasis added) 40
Of course, if amateurism itself is what gives rise to the demand for college sports, a rule controlling pay to athletes has, nearly by definition, procompetitive benefits. Thus, a central issue is whether amateurism actually matters. Others have similarly noted that “…understanding fan preferences for amateurism becomes an important consideration in identifying the most appropriate way to address the NCAA’s arguments [with regards to pay restrictions].” 41
Thus, the fundamental question we address is whether amateurism is necessary to sustain the significant demand for top-level college football and basketball. The essence of the NCAA and others’ claims that amateurism is on balance procompetitive is that if it were eliminated there would be significant detrimental effects on the demand for top-level college sports.
We find that neither logic nor evidence supports the necessity of amateurism to maintain fan interest in top-level college sports. Harrison and Harrison note that the NCAA’s argument in effect makes amateurism the sole distinguishing factor of college football and basketball. 42 This cannot be the case. School loyalty, traditional rivalries, the academic tradition are additional sources of the differential demand for college sports. Schools have built-in fan bases, and demand for competitive college sports would likely continue to exist as long as student-athletes who represented schools compete on the fields or courts.
We base this conclusion on a case study of the Olympics and other now-professional sports. We find that the demand for sports appears to be unconnected to receiving remuneration, or maintaining an “amateur” status.
The Olympics and amateurism
The history of the Olympic Games suggests the importance of amateurism is overstated in terms of fan interest. In 1960, Avery Brundage, then president of the International Olympics Committee, asserted that “[i]f we water down the rules [on amateurism] now, the Games will be destroyed within eight years.” 43 He has been clearly proven wrong. Barring a few exceptions, such as wrestling, the Olympic sports are fully professionalized and provide us with many years of experience to assess the effects on consumer demand of allowing athletes to receive payment. We first provide an overview of the shift away from amateurism by the International Olympic Committee (IOC), the principal organizing body of the Games, and then we discuss the transitions of some illustrative individual Olympic sports.
By the late 1960s, the IOC realized that it was not possible to keep some sports from turning professional. Brundage stated that “[w]e are trying to do the impossible. It’s about time to recognize that some sports and events cannot be kept amateur at international level.” 44 This began the departures from the traditional concept of amateurism. In Eastern Europe, for example, the concept of the “state amateur” developed, whereby governments would support world-class athletes with professional training facilities and subsidies. 45 In 1971, after many discussions about the “amateur problem,” the word “amateur” was dropped from Article 26 of the IOC Charter, which defined the eligibility status for Olympic athletes. However, prosecutions of offences continued despite this change, 46 until 1974, when the IOC allowed athletes to be compensated for the time they spent in training camps or traveling to competitions. 47
The next significant change came in 1981, when the IOC delegated control for designating eligibility to the individual sports federations.
48
The athletes played a key role in this change as they were invited to participate in deliberations for the first time. However, the IOC President Juan Samaranch remained clear in his views that “professionals,” which meant athletes receiving compensation from competing or payments from corporations, had no place in the Olympic Games: Open or professional competitions have no place in the Olympic Games.…The Olympic Games will only remain the Olympic Games if all athletes from all over the world participate except for the real professionals.
49
In late 1981, the IOC added tennis and table tennis as Olympic sports for the 1988 Olympics to be held in Seoul, South Korea. These were both highly professionalized sports and, presumably to have meaningful competition, the IOC allowed participation in the Olympics by fully professional athletes for the first time. 51
Even today, the IOC has not embraced professionalism—in that nowhere in its articles does it say that professional athletes are allowed in the Olympic Games. However, by leaving the decision of eligibility to the individual sports federations, the athletes in most Olympic events are de facto professionals. We examine a few of the transitions away from amateurism in individual Olympic sports to understand if that transition adversely affected demand for these sports.
Track and field
Track and field became fully professional in the 1990s with a long transition that began in the 1970s and accelerated in the 1980s. The requirement that track and field athletes be amateurs was strictly enforced by the modern incarnation of the Olympics in its early decades. This was famously exemplified by the Jim Thorpe incident. In 1913, Jim Thorpe was stripped of the gold medals he won in the 1912 Olympics because he had been a professional minor league baseball athlete. 52
In 1982 the International Amateur Athletic Federation (IAAF) amended its amateur rules to allow athletes to accept endorsements, sponsorships, and prize money, as long as they were deposited into a trust fund. 53 In 1993 the IAAF relaxed the rules further and allowed athletes to receive pay directly. 54 In 1997, the IAAF allowed prize money in its top non-Olympic competitions for the first time. 55 In 2001, the IAAF officially changed its name to International Association of Athletics Federations, dropping the amateur reference from its name.
Even though professional Olympic track and field athletes were competing by 1982, consumer demand for the sport was not adversely impacted. 56 For example, an analysis of consumers’ favorite Olympic events undertaken in 2016 shows that track and field holds the top five spots: 100 meter (m), 4x100 m relay, 4x400 m relay, and 100 m/110 m hurdles. 57 For the London Olympics in 2012, the third most streamed event on the Internet was Usain Bolt’s winning the 100 m dash. 58 Thus, there is simply no evidence that the shift to professionalism has had any adverse impact on the demand for Olympic track and field.
Basketball
Basketball has been an Olympic sport since 1936. 59 For the first time in 1989, to take effect for the 1992 Olympics, the International Basketball Federation allowed professional players, mostly NBA players, to play in the Olympics. In 1992, the USA fielded what was called the “Dream Team” consisting entirely of NBA players. This team went on the win the Olympic gold medal by a wide margin. Rather than adversely impacting the demand for Olympic basketball, allowing professionals to compete resulted in record television viewership and record demand for tickets. 60
Soccer
In 1984, the IOC allowed professional soccer players to play in the Olympics. However, because FIFA did not want Olympic soccer to rival the World Cup, a compromise was reached with the IOC (dictated mainly by FIFA) that allowed only African, Asian, North and Central American, and Caribbean teams as well as those from Oceania to play with their best players regardless of professional status. European and South American players were restricted to players who had never played in a World Cup. 61
In 1992, the IOC further changed the rules and allowed only players under twenty-three years old to play on Olympic soccer teams regardless of professional status. 62 And since 1996, a further modification was made allowing three over-twenty-three players to play on each national squad. 63 Despite these restrictions, Olympic soccer increased in popularity after professional players were allowed to play. The average attendance in the 1984 Olympics in Los Angeles—the first year to include professional players—was almost double that of the prior Olympics in Moscow. Average attendance at soccer games in the nine Olympics since 1984 has been 31,965, an 88% increase over the nine prior Olympics. 64 And the London 2012 Olympics had the highest average attendance in Olympic history. 65
Figure skating
The sport evolved from purely amateur, to allowing token payments for performances in exhibitions, to accepting money from endorsements as long as the money was put into trust funds rather than given to the skaters themselves. 66 In 1990, the International Skating Union relaxed its rules to allow skaters to earn money from exhibitions and shows. In 1992, it further amended its rules to allow competition in sanctioned professional competitions. 67 The only restriction currently in place by the International Skating Union is participating in unsanctioned “pro” competitions—a practice that has been challenged by the European Commission for being anticompetitive. 68 Although demand for figure skating has fallen in the twenty-first century, it grew steadily though the 1990s, well after figure skating athletes relinquished their amateur status. 69
Non-Olympic sports and amateurism
Outside of the Olympics, the competition in those sports that evolved from amateur to professional also did so without any detrimental effect on demand. 70 For some sports, like tennis, demand has exploded. Tennis was largely an amateur sport before the start of the “Open Era” in 1968. Before the Open Era, professional players were not allowed to compete in the sports’ top tournaments, including the “Grand Slam” events: Wimbledon, U.S. Open, French Open, and Australian Open. 71 In the Open Era, the Grand Slams have grown tennis into a hugely popular sport that takes in nearly $1 billion in revenues each year. 72
In Rugby, there was a contentious split in the late nineteenth century from the Rugby Football Union, which required strict amateurism. A separate organization was created called RF League that allowed payments to players. This was mostly in the working-class northern England teams. 73 Now, all top-level rugby (with the exception of some national club leagues, like Argentina’s) allow professionals. 74 Rugby is as popular as ever. 75
We have come across only limited evidence about the importance of amateurism in the demand for college sports comes. The NCAA recently offered survey evidence in support of the importance of amateurism. In the O’Bannon trial, J. Michael Dennis, serving as expert for the NCAA, presented survey results showing that (1) 69% of the public and 61% of sports fans oppose paying college athletes; and (b) 38% of respondents are less likely to watch or attend a game if athletes are paid $20,000, 47% if paid $50,000, and 53% if paid $200,000. 76 However this survey provides little evidence that amateurism is important to the success of college. Firstly, the survey merely reflects the romantic notion of the value of amateurism in the abstract, but it does not demonstrate that consumers would actually stop watching or rooting for their favorite college sports teams if athletes were paid. Judge Wilken, the District Court judge in the O’Bannon case, noted in her opinion that surveys of this type “are inevitably a poor tool for accurately predicting consumer behavior.…” She cited “various polls and surveys which documented widespread public opposition to rule changes that ultimately led to increased compensation for professional baseball players and Olympic athletes.” Consistent with our discussion of such changes above, she pointed this out “even as Major League Baseball and the IOC were experiencing periods of massive revenue growth.” 77
As we have discussed, the shift from amateur to professional athletes in Olympic and non-Olympic sports has not affected their popularity. We thus conclude that there is nothing intrinsic to amateurism that elicits fan loyalty, 78 and that there is no evidence that the absence of payment to athletes plays an important role in the demand for viewing and consuming college sports. Yet the popularity of college sports, which undoubtedly is of a lower quality than its professional alternatives, implies that these sports are clearly differentiated products. As hinted at in many discussions of the role of amateurism, that differentiation appears to be the connection of the sports to the colleges, and not to amateurism or player compensation. 79 The participants in college sports are viewed as “student-athletes,” a category distinct from merely athlete. Regardless of whether they are paid, nonprofessionalism is and would be maintained by ensuring that the main job of the athletes was to be students rather than simply paid sports participants. Thus, we believe that the proper distinguishing characteristic between amateurism and professionalism in regards to college sports is not payment for services but whether the athletes are perceived as being distinct from athletes participating in, say, the NFL and the NBA. 80
Indeed, this is the focus of most marketing in college football and basketball, which deals with school loyalty and rivalries. And as Sanderson and Siegfried have pointed out, many of the top-level teams are located in rural areas with few other entertainment options, where school loyalty is part of the culture. 81
By requiring attendance and progress in degree granting college programs, the athletes are properly associated with the student body as a whole. In addition, the long history of football and basketball in American colleges and universities has engendered a dedicated fan base of current and former students, faculty and administrators, as well as residents of the surrounding areas, all of whom associate with the college. This fan base over time has created a vibrant college sports culture in the United States. Even though that culture may have been based in part on the amateur nature of the competition, it has by now attained an identity independent of amateurism. 82 The appeal of college sports centers on the intensity of competition among the different teams with strong histories and place associations vying to be the best in the country. 83
V. The NCAA Compensation Restrictions, Competitive Balance, and Arms Races
Competitive Balance and Payment Restrictions
A lopsided sporting event is entertaining only to the most masochistic fan. 84 The more typical fan prefers “the thrill of victory and the agony of defeat” that arises when there is uncertainty in the outcome. The notion of competitive balance arises from this “uncertainty about the outcomes of professional sporting events.” 85 One common way to measure competitive balance is by the dispersion of winning percentage within sports leagues. 86 Some authors employ specifically the standard deviation of teams’ winning percentages in a period, defined as the ratio of wins to total games played. 87 A seminal paper by Rottenberg theoretically examines the positive impact on fan demand from greater uncertainty of outcomes. 88 From an empirical perspective, there exists evidence that greater competitive balance increases demand for a sport. Others discuss how “a sporting competition is more entertaining and of higher quality when the game’s outcome is more unpredictable.” 89
The NCAA argues that the current pay restrictions are necessary to maintain competitive balance in top-level college sports. In their view, without the restrictions, only a few select teams would be able to afford the top players, and then only those teams would be contenders for the most prestigious football and basketball titles. However, the NCAA’s argument is flawed.
First, the current compensations restrictions cannot be sufficient for maintaining competitive balance because college sports are currently highly imbalanced. There exists evidence of a strong correlation from year to year on a school’s recruiting success. One author finds that “a small number of teams are getting the lion’s share of the best athletes.…” 90 This suggests that there is not a level playing field now, at least not in recruiting. To the extent that there is a correlation between recruiting and success on the field, and the evidence suggests that there is, 91 then competitive imbalance in recruiting translates into competitive imbalance in outcomes. 92
Second, the NCAA conflates two distinct issues. The first issue is whether payment beyond current levels would disrupt competitive balance. We do not believe it would. For example, each athlete could be paid a wage greater than cost of attendance that is the same across all athletes and schools. While some colleges may drop out of top-level sports with such fixed payments, for those remaining there would be no significant impact on competitive balance. 93 Furthermore, loosening pay restrictions could allow weaker teams to become more competitive. The compensation restrictions prevent weaker teams from targeting select star players with increased compensation. 94 Thus, the restrictions may actually help to maintain the status quo. In a more flexible pay system, weaker schools would be able to easily reshuffle money away from coaches and other parts of athletics to attract top-level talent. 95 This could allow weaker schools to attain better outcomes in the field and obtain a larger share of the significant pool of revenues available to top tier college teams, 96 resulting in more competitive leagues. 97
The second issue is whether unfettered competition for athletes via unrestricted monetary payments might adversely impact competitive balance. We do not fully address this issue because unbridled competition for sports talent likely would engender a related but distinct inefficiency in the form of excessive spending on athletes due to the phenomenon of “arms races.” We explore the unique characteristics of sports markets that cause these inefficiencies next.
Inefficiencies from Unbridled Competition for College Athletes
We have argued that paying college athletes an amount beyond cost of attendance is not likely to adversely impact the demand for top-level college football or basketball. The evidence supports the hypothesis that the differentiation between college sports and professional sports is driven not by the payment itself but by a team’s association to a school. That is, participation by student athletes linked to a particular school rather than by athletes whose primary occupation is athletics differentiates college sports from professional sports.
Nonetheless, this does not imply that an unfettered market for athletes who are also students is the optimal replacement to the current NCAA restrictions on player payment. There are important features of athletic competition that suggest that inefficiencies would arise from such unbridled competition for student-athletes.
Sports generally exhibit the phenomenon known as an “arms race.” 98 The analogy is to warfare and/or its threat. Smith (1980) defines an arms race as “the participation of two or more nation-states in apparently competitive or interactive increases in the quantity or quality of war material and/or persons under arms.” 99 The notion arises in competitive situations in which the goal is be more successful than one’s competitors. This is in contrast to the usual case in which the benefit of an action can be considered independent of the success of competitors.
In the sports context, the value of the output is a function of how the athlete or team does compared to the competitors rather than being only a function of the quality of the output. 100 That is, finishing first itself has value independent of quality. 101 In addition, finishing first (rather than second) can have a differentially large value compared to, say, the difference in value between finishing second and third, which may be very small. This is exemplified by the saying, “Nobody remembers who finished second but the guy who finished second.” 102 Because of the value of being better than the competition, an economic incentive exists in sporting events to engage in, for example, inefficiently excessive training on the part of athletes and excessive spending on facilities, coaches, and athletes on the part of teams. 103
Since there are significant benefits to winning (and in the extreme, none for losing), an owner competing in an unfettered labor market may pay excessively for athletes relative to the efficient amount. 104 The reason is that any additional spending on improving the quality of the team imposes a negative externality on its rivals. The rivals are now motivated to pay more for talent so as to remain competitive. This could lead to a bidding war that might result in salaries for athletes that are higher than their social value. The teams could end up utilizing resources that would have been more efficiently spent elsewhere. 105 A prediction from this literature is that there would be inefficient spending that results in schools generally not making substantial profits, and many suffering losses. 106
We conclude therefore that unbridled competition for athletes would result in socially wasteful spending for athletes that would lead many schools to lose money in their athletics programs and inefficiently utilize their scare resources. 107 We discuss next how to improve on the current compensation system without directly embarking in an arms race for talent.
VI. Towards More Efficient and Less Restrictive Compensation Schemes for College Athletes
Up to this point, we have made the case for the following three conclusions: (1) that the NCAA restrictions on payment to college athletes are not justified by amateurism; (2) that maintaining the high level of demand for top-level college sports requires restrictions that ensure that athletes are also students; and (3) that unfettered competition for college athletes would lead to excessive and inefficient pay such that restraints on such competition are appropriate. We conclude this article by proposing ways in which to impose more efficient and less restrictive compensation schemes for athletes.
The current pay restrictions imposed by the NCAA are not efficient because they make no attempt to compensate better players more than weaker players. At the same time, the current pay restrictions do not alleviate college sports from the inefficiencies arising from positional competition because capping player compensation by itself, as the NCAA currently mandates, does nothing to alleviate the potential arms race in expenditures on facilities and coaches. The NCAA places no restrictions on that spending.
There exist less restrictive mechanisms that preserve the value of college sports as a distinct product offering for consumers without the anticompetitive effects of the current system. 108 One criterion for any alternative compensation scheme for athletes must be that the athletes be students linked to particular schools. Any new system should avoid populating college teams with purely professional players who are not students of the school they play for.
The positional competition problem is, of course, not limited to top-level college sports. It is also a problem impacting professional sports. We therefore suggest that professional sports may offer a model for dealing with excessive spending while avoiding the anticompetitive effects from the NCAA’s salary restrictions. Some of these alternatives can take the form of payroll caps, revenue shares, drafts, and player unions.
Possible models taken from pro sports may need to be put in place in conjunction with transfer rules and requirements of maintaining student status for the athletic eligibility. It seems clear that we should not simply remove the current restrictions and switch to an unregulated pay system. This would exacerbate whatever arms race inefficiencies currently exist with schools’ competing in unbridled fashion for coaches and via building expensive facilities.
Teams and a collective union of players in the NFL, NBA, NHL, and MLS negotiate salary caps, which vary in terms of the degree to which teams can surpass the cap. The NBA and MLS have a “soft” cap that allows teams to go beyond the cap for certain “designated” players, while the NFL and NHL have a “hard” cap. 109 With regards to college sports, any adoption of restrictions modeled after the pros must be made in conjunction with the creation of a college athletes’ union to collectively negotiate on behalf of collegiate athletes. Otherwise, the resulting outcomes will likely be inefficient with the NCAA able to exploit its monopsony power over athletes. In addition, if we are correct that amateurism itself is not of significance to the demand for college sports, and if the courts recognize this, the setting of compensation by the NCAA would likely be a per se violation of the Sherman Act.
With those points in mind, we propose the following model to replace the current pay restrictions in college sports: Designate certain conferences to be the “top-tier” conferences where athletes’ compensation can exceed the “cost of attendance.” Other schools would be in “amateur” conferences. Players in all conferences must be full-time college students. The current transfer restrictions should remain in place. The top-tier players should be allowed to form a labor union to collectively negotiate compensation and other rights.
110
Set up a team salary cap negotiated each year for schools in the top-tier conferences. (Currently, the NBA awards approximately half of its revenues to athletes; the NFL allots slightly under 50%.
111
) Teams exceeding the negotiated cap would face stiff penalties. There could be exceptions for a limited number of “designated” players as in MLS.
112
Teams would be free to spend on designated players with only a max amount counting toward the cap.
VII. Conclusion
In 2014, the NFL had revenues of about $12 billion. 113 Under their collective bargaining agreement, the NFL players received wages and benefits of about $5.8 billion, or 48% of the total revenues received. 114 In that same year, the top fifty NCAA football programs earned revenues of over $5.2 billion on that sport, nearly half the revenues earned by the NFL. In contrast to the NFL, however, because of the NCAA compensation restrictions, the college players received about $130 million in benefits, about 2.5% of the corresponding college football revenues. 115 This shows that the NCAA pay restrictions result in significant redistributions from the most important inputs generating the revenue, the players, to the institutions competing in the sports marketplace.
The justification typically offered for this redistribution is the need to protect the demand for the top-level revenue-producing athletic events, college football and basketball, from the scourge of professionalism. It has been near gospel that the demand for college sports is closely tied to the athletes being perceived as amateurs. Yet we have found no evidence in support of this stylized fact, and we suggest that it is false. We argue that it is not the absence of pay that is important to the fans of college sports; rather, it is the notion of the athletes as students affiliated with colleges and universities that distinguishes these sports from their professional counterparts. With that in mind, regulation by the NCAA, or another regulatory board, is necessary for the continued success of these “businesses.” However, such regulation should be about ensuring that college players continue to be students as well as athletes. This would provide a proper procompetitive justification for collective decision making in a similar spirit as rules on class attendance, progress towards a degree, and limits on transfers.
In addition, we argue that unbridled market competition for top-level college athletes would be contrary to the long-run success of college football and basketball. This is not because of the resulting compensation itself, but because of the externalities that exist in sporting events as teams race to the top. These externalities would lead to an arms race or positional competition that would be socially wasteful. However, unlike regulations that pertain to the student aspect of athletics, collective decision making by schools about compensation would likely not be justified under the Sherman Act. Such coordination could only have as a procompetitive justification to achieve greater efficiency in production. This falls short of what has been deemed an acceptable justification by the Supreme Court so far—ensuring the survival of the product itself.
In order to alleviate the inefficiency of positional competition therefore, we propose that college sports adopt compensation schemes akin to those found in professional sports. College athletes would form their own cartel via unionization, and bargain with the NCAA over both the student requirements for playing sports as well as compensation schemes such as team salary caps. This collective bargaining would need to take place at the national level to avoid the inefficiencies of an “arms race” among schools or conferences.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.