Ilya Klimenko

Econometrics Project

May 3, 2013

Superstars in the NBA: $ Cost or Benefit? $

Introduction:

Growing up we all dreamt of a life of fame and fortune. Many felt they had what it takes to compete professionally in a particular sport. Athletes will most definitely tell you they do it for the love of the game. Despite these professed motives expressed by sports figures, as economists we understand and believe otherwise. We are all aware of the inordinately high salaries that athletes are paid and recognize that that is their motive to put themselves through such a strenuous work regimen. For example, in the 2012-2013 NBA season, the LA Lakers’ Kobe Bryant made over $27.85 million! In most cases teams like the LA Lakers can afford to pay players a lot of money based on the amount of revenue they accumulate. In this upcoming analysis on the NBA, we will answer the question: Do gate revenues increase with the arrival of a superstar?

Standard economic reasoning suggests that a player’s salary will be set to (approximately) equal his expected contribution to the team’s revenues over the season – his so-called “marginal revenue product”. From a fan’s perspective, a player’s contributions mostly relate to the team’s win-rate: can this athlete help win the team more games and eventually secure a championship title? However, when owners analyze this problem they actually think in much more economic terms. They believe that the player will improve the team’s performance and in turn will generate higher revenues; which are generated from gate receipts[1], unshared local television contracts, and distributed national television contracts (Li, 2011). All in all, the only reason we continue to analyze an idea such as this is to ultimately determine if this player is worth it to the team, managers, and owners.

Literary Review:

As stated previously, team managers and owners sign new players based on the value they will bring to the team. Using this basis, to define our superstar variable, we will be using the player’s annual salary. Those players with the highest salaries we will conclude are accredited with being the team’s superstars[2]. These statistics we will carry across all 30 teams. Those teams with the luxury to afford 2 or more superstars we will also incorporate in our analysis, due to the fact that the more superstar presence on the court, the more they contribute to gate revenues and winning percentage.

We study the examination of the superstar further with Berri, Schmidt, and Brook (D. J. Berri, 2004) who extended this work and delved deeper into the superstar effect. Not only did they examine certain star players like Michael Jordan, Shaquille O’Neal and Grant Hill, but they also looked at players who did not have as much popular recognition but were still considered All-Star caliber players by the NBA. They found that these kinds of stars did not have significant effects on gate receipts, however they concluded that a star’s effect on revenue was mostly due to their effects on a team’s win-loss record[3] and not from the popularity of that individual player. A preceding study done by Berri (D. J. Berri, 1999) looked into a players performance and how it contributed to wins. Within this study, the results concluded as were expected; the positive contributing performance of a superstar, the more wins the team had. Which he followed up with his 2004 study mentioned earlier.

Conversely, Hausman (Hausman, 1997) proposed that certain players’ stardom was a significant revenue driver in the NBA. He illustrated this point through data analysis, concluding that Michael Jordan was responsible for over $200,000 of the New Jersey Nets’ revenue in a season where the Bulls only played twice at New Jersey. On top of that, Hausman suggested that superstars such as Larry Bird and Michael Jordan in fact had a significant effect on team attendance; which in turn positively affected the team’s revenues. It seemed that a major factor in team revenue was not the number of wins they could accrue over a season but how many NBA stars they could acquire in order to attract more fans.

Empirical Model:

The main objective of this paper will be to analyze the aforementioned relationship and connect it with the relationship of a players marginal revenue product and gate receipts. Taking into account the statistics used by Berri, Schmidt, and Brook, we will be using similar variables such as team wins and gate revenues. Combining them with data gathered from Forbes and ESPN on more recent data for player’s salaries, average ticket price, and gate revenues. Also we will incorporate per capita personal income gathered from the Bureau of Economic Analysis to determine what the likelihood of the population[4] of the city the team resides in will spend on a ticket to a game. All other information considered our model would be as follows:

GATE= β1+ β2 (SALARY) + β3 (WINS) + β4 (AVGTIX) + β5 (PCAPINC) + β6 (POP) + e

Our dependent variable GATE represents the gate receipts and revenues for the current NBA season as well as the year before the player was signed. Our independent variables SALARY, WINS, AVGTIX, PCAPINC, and POP are as follows: SALARY is equal to the amount paid to the superstar(s) during the year of play as well as the salaries of the players before the current superstar. WINS will be the team win percentage for both seasons, AVGTIX is the variable representing the average ticket price in the arena for both years; PCAPINC[5] is equal to the per capita personal income of the metropolitan area the team resides; and POP is the population of the metropolitan area and surrounding relevant areas as reported by the Bureau of Economic Analysis.

Looking at the model presented, we will expect all the independent variables to have a significant enough impact on the dependent variable. The specific independent variables such as Player Salary and Team Win Percentage we expect there to be a positive (+) impact on Gate Revenues whereas Average Ticket Price will be looking to have a more negative (-) impact. The final variables, Per Capita Personal Income and Population, can go either positive or negative (+/-). The reason we look for Player Salary and Team Win Percentage to have a positive impact is simply because an increase in both variables presumably yields a positive outcome on Gate Revenues. Along with that, Average Ticket Price will have a negative affect simply because the more superstars that are on the court, the more expensive it is to see them, resulting in a negative impact on Gate Revenues. Finally, Per Capita Personal Income and Population will have varying affects on Gate Revenues because the entire population will not be basketball fans and even those that are do not necessarily spend their income to attend games.

Data Description:

All the data sets collected will feature all 30 teams for the 2012-2013 season and one year prior to the superstar joining that team. Considering the current NBA rosters and our proposed superstars, the year prior will vary due to different players entering in different seasons. We have accounted for this using the CPI to convert salaries and all other financial data. The Player Salary, Team Win Percentage, Average Ticket Price and Gate Revenue variables in the above model will be determined based off the current NBA season (2012-2013) with all players in their current roles on their current teams. Considering the fact that most teams cannot afford to support more than three superstar salaries, we have accounted for this using the top three paid players on all team rosters; as in the players signed to higher salaries presumably contribute the most to the teams success and increase in revenues. The incorporation of the data set including the gate revenues one year prior to the superstar arriving to the team is important in comparing if the gamble to invest in the player joining the team actually benefited the teams gate revenues.[6]

The Results:

The results of the OLS model 1 are reported on page 10. It is evident in this model that these factors provide a great deal of variation in gate revenues. Heteroskedastic tendencies have been accounted for in all analysis, as you will see on page 10. The results yield the explanatory powers of our model. The initial test results on page 10 show that every variable except SALARY and POP have a significant impact on GATE revenues. In fact, given our t-value statistics, we can be certain at the 99% confidence interval that the variables contribute significant revenues to the teams. For example, our model shows that for every 1 more win a team had, it contributed nearly $1.5 million to gate revenues. Conversely, our model predicts against our main hypothesis. The variable SALARY had a minimal affect on gate revenues. Justifying the study done by Berri. As stated previously and solidified by our model, the team win percentage had a much greater affect on gate revenues than superstar salaries. Our RMSE does provide some support to this test yielding a possible variation of $6 million, which compared to our team values, is large but not overly skewed. Perhaps the most controversial result in this analysis is the results yielded by AVGTIX. The t-value for AVGTIX is much larger than all other variables, which we should assume because the ticket price is what contributes mostly to gate revenues. Taking this into consideration, a test for multicollinearity follows the initial test with variable AVGTIX removed and retested. With the results on page 11, we yield an entirely different result. In these findings we see that the most influential variable is SALARY with a t-value of 4.57. Second most influential variable being WINS. The RMSE in the second model was almost nearly $12 million dollars, which is, quiet a large number making this model skeptical; leading me to believe there exists an omitted variable bias we did not account for.

The Conclusion:

In order to observe the impact of a superstar on the gate revenues of an NBA franchise an OLS regression was estimated with several variables that are predicted to have some influence on gate revenues. The results of the models run lead us to believe that despite the fact that teams pay superstar players outrageous salaries, does not lead to a major change in gate revenues.

For the results of model one we conclude that average ticket price has the largest impact on gate revenues; followed by per capita personal income and team win percentage. Interestingly in this model population had a negative impact on gate revenues and salary was insignificant in explaining the model. This result leads us to believe that the marginal revenue product of a superstar is minimal. In the second model assuming multicollinearity, we omitted the average ticket price to determine if there was a change in the results. As it goes to show most of the variables had a distinctive change in effects on gate revenues. Most significant is the fact that the superstar salary variable took a complete turn and became the most significant variable to gate revenues, with team win percentage a close second.

As with most models, there are certain to be limitations. On teams such as the LA Lakers and Boston Celtics, current superstars such as Paul Pierce, Kobe Bryant and Tim Duncan entered the league over ten years ago and NBA contracts were not as large. Also, superstars like Lebron James, have not been with the same team and when he was the salary difference is much different. Many years have passed for other teams since a superstar played with them and those who are paid the most are not necessarily superstars. Lagged attendance to the events must also be considered as not everyone will follow a team the initial year of the superstar’s arrival. Finally, the population figures do not attest to the fact that those in the metropolitan area are fans of the sport or the team.

Given time and full statistical support, further research using several more variables will most definitely yield other results.

Variable / Definition / Source
GATE / Gate Revenues($millions) / FORBES. http://www.forbes.com/nba-valuations/
SALARY / Players Salary($millions) / ESPN. http://espn.go.com/nba/salaries
SALARY variable determines superstar status.
WINS / Team win Percentage / ESPN. http://espn.go.com/nba/standings/_/type/expanded
WINS variable determines superstars MRP to team success
CAPINC / Per Capita Income / U.S. Department of Commerce: Bureau of Economic Analysis.
http://bea.gov/iTable/iTable.cfm?reqid=70&step=1&isuri=1&acrdn=5#reqid=70&step=26&isuri=1&7023=7&7024=Non-Industry&7001=720&7090=70&7029=20&7031=5&7022=20
POP / Population / U.S. Department of Commerce: Bureau of Economic Analysis
http://bea.gov/iTable/iTable.cfm?reqid=70&step=1&isuri=1&acrdn=5#reqid=70&step=26&isuri=1&7023=7&7024=Non-Industry&7001=720&7090=70&7029=20&7031=5&7022=20
AVGTIX / Average Ticket Price / FORBES. http://www.forbes.com/nba-valuations/

Works Cited

1. Berri, David J. "Who is Most Valuable? Measuring the Players Production of Wins in the National Basketball Association." 1999. Online Wiley Library. <http://onlinelibrary.wiley.com/doi/10/1002/1099-1468(199912)20:8%3C411::AID-MDE957%3E3.0.CO;2-G/pdf>.

2. David J. Berri, Martin B. Schmidt and Stacey L. Brook. "Stars at the Gate: The Impact of Star Power on NBA Gate Revenues." Journal of Sports Economics (2004): 33-50.

3. ESPN-NBA. NBA Salaries- 2012-2013. 2013. <http://espn.go.com/nba/salaries>.

4. ESPN-NBA(2). NBA Expanded Standings- 2012-2013. 2013. <http://espn.go.com/nba/standings/_/type/expanded>.

5. Forbes. NBA Team Values: The Business of Basketball. 2013. <http://www.forbes.com/nba-valuations/>.