Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
For the love of the game? ownership and control in the NBA
(USC Thesis Other)
For the love of the game? ownership and control in the NBA
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
1
For the Love of the Game? Ownership and Control in the NBA
Kari Joseph Olsen
University of Southern California
May 15, 2015
The degree being conferred is: PhD in Business Administration—Accounting
I gratefully acknowledge the support on this project from my dissertation chair S. Mark Young,
along with dissertation committee members Eric Allen and Peer Fiss. I also recognize support
received from the University of Southern California’s Marshall School of Business along with
the Deloitte Foundation’s Doctoral Fellowship. I appreciate comments and suggestions on this
research project from H. Scott Asay, Kurt Badenhausen, James N. Cannon, Dane Christensen,
Jeremy Douthit, Kelsey K. Dworkis, Derek Harmon, D. Kip Holderness, Christo Karuna, Ed
Lamb, Kenneth A. Merchant, Troy Pollard, Bryce Schonberger, Steven D. Smith, James M.
Stekelberg, Monte R. Swain, Todd A. Thornock, Marshall Vance, workshop participants at the
University of Southern California, and participants at the BYU Accounting Research
Symposium.
University of Southern California, Marshall School of Business, Los Angeles, CA 90089.
Contact: kari.olsen.2015@marshall.usc.edu
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
2
TABLE OF CONTENTS
List of Tables……………………………………………………………..…………………..pg. 3
List of Figures………………………………………………………………………….……..pg. 4
Abstract…………………………..………………………………………..……………….….pg.5
I. Introduction………………………………………………………………………..…….pg.6-12
II. Background and Related Literature…………………………………………………...pg.13-28
Managerial Myopia
Brief History of the NBA
NBA Owners and Managerial Objectives
Player Compensation Decisions
On-the-court Outcomes (Winning)
Off-the-court Outcomes (Financial Performance)
III. Research Design……………………………………………………………………..pg.28-46
Sample and Variables
Descriptive Statistics and Correlations
Hypotheses Tests
IV. Results…………………………………………………………………………..…...pg.46-83
Primary Test Results.
Results of Hypotheses Tests
Timing of Owner Objectives
Additional Analysis
Determinants of Team-Level Player Compensation
Robustness Tests
Alternative Measures of Team Performance
Alternative Measure of Player Expenses
V. Conclusion……………………………………………...……………………….…….pg.84-86
VI. Future Directions………………………………………………………………….....pg. 86-90
Accounting Future Directions
Sports Entertainment Future Directions
References……….……………………………………………………………...…….…..pg.91-99
Appendix……………………………………………………………………….….......pg.100-101
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
3
LIST OF TABLES
Table 1 Road Attendance Values by Team…………………………………………………...pg.30
Table 2 Descriptive Statistics…………………………………………………………………pg.32
Table 3 Correlations…………………………………………………………………………..pg.33
Table 4 Player Expense………………………………………………………………….pg.34 - 35
Table 5 Profit Measures…………………………………………………………………pg. 39 - 42
Table 6 Regression Estimating Team Performance…………………………………………pg. 47
Table 7 Regressions Estimating Revenue……………………………………………….pg.49 - 51
Table 8 Regression Estimating Road Attendance……………………………………………pg.52
Table 9 Regression Estimating Team Value…………………………………………………pg.55
Table 10 Timing of Owner Objectives……………………………………………………pg.56-57
Table 11 Regression Estimating Team Level Player Compensation…...………………pg. 61 - 62
Table 12 Robustness Test of Regression Estimating Team Performance………………pg. 65 - 66
Table 13 Robustness Test of Regression Estimating Revenue…………………………pg. 68 - 73
Table 14 Robustness Test of Regression Estimating Revenue…………………………pg. 74 - 75
Table 15 Robustness Test of Regression Estimating Home Game Attendance……….pg. 77 - 78
Table 16 Robustness Test of Regression Estimating Road Game Attendance…..……..pg. 79 - 80
Table 17 Robustness Test of Regression Estimating Team Performance……………...........pg.82
Table 18 Robustness Test of Regression Estimating Road Game Attendance……………...pg.83
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
4
LIST OF FIGURES
Figure 1 Geography of NBA Franchises as of 2014…………………………………………pg. 16
Figure 2 Hypothesized On-the Court and Off-the-Court Outcomes…………………………pg.24
Figure 3 League-wide Averages of Select Variables…………………………………..pg. 36 -38
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
5
For the Love of the Game? Ownership and Control in the NBA
Abstract
The National Basketball Association (NBA) presents a setting wherein agency frictions and other
causes of managerial myopia, which are common in many organizational settings, are limited for
NBA team owners. Much like managers of any business, NBA team owners make decisions
about how to pursue their objectives, which in this setting include clearly defined strategic
objectives of winning games and maximizing profits. I examine NBA team owners’ resource-
allocation decisions related to these objectives by analyzing the effect of team-level player
compensation on on-the-court (winning) and off-the-court (financial performance) outcomes.
Results show that spending more on team-level player compensation has a positive effect on
team’s winning, yet is also associated with lower current operational profits. However, results
also show that spending more on team-level player compensation leads to higher team values due
to an increase in revenue generation ability and brand value from winning more games. Rather
than winning and profits being alternative objectives to satisfy, the realization of the one
objective (long-term profits) is contingent upon investments leading to the satisfaction of the
other objective (winning). Hence, results suggest that NBA team owners who pursue short-term
operational profits by spending less on team-level player compensation likely do so at the
expense of real-economic future value. That is, despite not facing typical causes of managerial
myopia, team owners may pursue myopic behavior.
Keywords: Ownership Objectives, Performance Measurement, Myopia, Brand Value, Firm
Value
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
6
For the Love of the Game? Ownership and Control in the NBA
I. INTRODUCTION
Business managers are often faced with multiple objectives and must make choices about
which objective they will take actions to achieve (Jensen 2001; Kaplan & Norton 1996; Simon
1964; Cyert & March 1963). Managers must make tradeoffs between satisfying objectives
because actions taken to achieve one objective may hinder the achievement of another objective,
such as when those objectives have varying time horizons (Laverty 1996). For example, prior
literature has found that managers frequently take actions to achieve a short-term objective even
at the expense of real-economic value and the realization of a longer term objective (Graham et
al. 2005).
1
This managerial myopia, or the tendency to focus on short-term rather than longer-term
outcomes, typically arises from agency problems associated with information asymmetry and
goal conflict between the principal and agent (Eisenhardt 1989; Jensen & Meckling 1976).
Managerial myopia has been shown to be affected by contract design, performance measure
characteristics, performance standards or budgets, capital market pressures, ownership structure,
and public accounting reporting requirements (Abernethy et al. 2013; Mizik 2010; Seybert 2010;
Bhojraj & Libby 2005; Graham et al. 2005; Lundstrum 2002; Bushee 1998; Cooper & Selto
1991; Dechow & Sloan 1991; Merchant 1990; Stein 1989; Healy 1985). Privately held firms
have been shown to have less myopic behavior because they are subject to fewer short-term
pressures than publicly held firms, although agency problems can still arise depending on
ownership concentration and owner involvement in management (Asker et al. 2014). A setting
wherein agency frictions and other common causes of managerial myopia are limited could be
1
Short-term objectives may include meeting analysts’ earnings forecasts, bonus thresholds, budget targets, debt
covenants, or cash requirements. Longer-term objectives may include building skills and competencies, creating
innovative ideas, improving operational efficiencies, and increasing owner or shareholder value.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
7
informative to observe how managers make tradeoffs between multiple objectives and whether
managerial myopia is still present.
The National Basketball Association (NBA) in the sports entertainment industry presents
such as a setting wherein owners
2
of large privately-held firms face multiple objectives in a
competitive environment (Wolfe et al. 2005). Furthermore, as very closely-held private firms
whose management and control of decisions are restricted to the primary decision agent and
residual claimant (i.e., the team owners), typical agency problems are limited (Fama & Jensen
1985, 1983). Much like managers of any business, NBA team owners make decisions about how
to pursue their objectives, which in this setting include clearly defined strategic objectives of
winning games and maximizing profits (Wolfe et al. 2005; Rosner & Shropshire 2004; Fort &
Quirk 1995; Quirk & Fort 1992).
Prior research in the sports management literature has generally viewed team owners’
objectives of winning games and maximizing profits to be competing as opposed to
interdependent (Leeds and von Allmen 2013; Dietl et al. 2011; Fort & Quirk 2004, 1995;
Alexander 2001; Vrooman 1995; Quirk & Fort 1992). On the one hand, team owners face an
objective of winning games which places demand on expending current-period financial
resources to maximize games won. On the other hand, team owners also have the objective of
maximizing profits (Fort & Quirk 2004; Keidel 1984). Thus, from this perspective, team owners
must make an economic trade-off between current-period profits and winning games.
However, the manner in which objectives are defined and measured is critical to
understanding manager behavior and what objectives will be pursued (Abernethy et al. 2013;
Artz et al. 2012). The sports management literature has defined the profit objective to be the
2
NBA teams take an organizational form of a proprietorship or partnership, although there is almost always a
majority owner (Fama & Jensen 1985).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
8
teams’ current year profits (revenue minus expenses) (Késenne 2014; Leeds and von Allmen
2013). This definition is convenient in that it fits within economic models of profit maximizing
behavior and organizations often use current year profits as a performance measure for goal
setting, evaluation, and compensation purposes (Késenne 2014; Merchant & Van der Stede 2012;
Fort and Quirk 2004, 1995). However, measuring the profit objective using current-period profits
may lead to myopic decisions, or interpreting owner behavior to be myopic, inasmuch as
winning games in the current period could lead to greater profits in the future.
3
Thus, how
objectives are measured is informative to understanding the interchange between the objectives,
manager behavior, and which objectives will be pursued.
More specifically, an alternative conceptualization of the profit objective is to measure
profits as the long-term financial return that comes from operational profits and team value. This
way of measuring profits recognizes the value derived from capital appreciation, takes a longer
performance measurement horizon, and provides a more holistic view of factors that could affect
managers’ decision-making and behavior (Feltham & Xie 1994). By measuring the profit
objective this way, it is possible that the relationship between team owners’ objectives of
winning games and maximizing profits can be viewed as interdependent rather than competing
(Silver 2014; Zimbalist 2003; Gladden et al. 2001; Jensen 2001). Hence, team owners must trade
off the future economic benefits from winning games against the present negative profit-impact
of expending resources to win games. In this sense, team owners’ objectives can be strategically
linked, recognizing a fundamental connectedness between the two objectives. That is, rather
3
For example, from a competing objectives perspective with profits defined as current year profits, team owners
who spend more money on team-level player compensation could be viewed as prioritizing winning instead of
profits. However, by altering the measurement of the profit objective to be the long-term financial return that comes
from operational profits and team value, the objectives could be viewed as interdependent. In which case, team
owners who spend more on team-level player compensation could be viewed as prioritizing profits by prioritizing
winning. Hence, rather than viewing team owners as choosing between their love of the game and love of money,
team owners who prioritize winning games, doing so for the love of the game, could also realize greater financial
gain, for the love of money.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
9
than winning and profits being alternative objectives to satisfy, the realization of the one
objective (long-term profits) can be contingent upon investments leading to the satisfaction of
the other objective (winning).
4
Important to this setting, both the winning and profit objectives are arguably linked to
team owners’ decisions regarding the most substantial expense teams incur: team-level player
compensation. All teams operate under the same contracting conditions; however, there is
significant variation across teams in team-level player compensation. Team-level player
compensation is a particularly useful measure to examine owner behavior because team-level
player compensation is a large, separately measurable and focal expense, and team owners have
clear involvement in the player compensation levels for their teams. Further, by observing team-
level player compensation, inferences can be made about team owners’ emphasis on their
strategic objectives.
In this paper, I empirically examine NBA team owners’ resource-allocation decisions
directed towards objectives of winning games and maximizing profits by analyzing the effect of
team-level player compensation on both on-the-court (winning) and off-the-court (financial
performance) outcomes. I first examine whether greater expenditures on team-level player
compensation leads to greater on-the-court performance. This link is important because it
establishes that decisions about how much owners spend on team-level player compensation
does in fact have an effect on a team’s winning. If so, then team owners can make strategic
compensation choices in pursuit of their winning and profit objectives.
4
This resonates with a substantial empirical literature which shows that certain objectives (e.g., objectives related to
product quality, service quality, or customer satisfaction) can be leading indicators of the realization of other
objectives (e.g., objectives related to profits or returns) (Abernethy et al. 2013; Bryant et al. 2004; Sedatole 2003;
Zeithaml 2000; Ittner & Larcker 1998).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
10
I next examine the effect that team-level player compensation has on off-the-court
outcomes (financial performance). I argue that spending more money on team-level player
compensation leads to an improved on-the-court product (winning), which in turn, leads to
greater financial returns from appreciated team values and operating profits (financial
performance) through two mechanisms: (1) increasing revenue generation and (2) increasing
brand values. First, the increased ability of a firm to generate revenue can increase firm value
(Chandra & Ro 2008; Lee 1999). Therefore, team values are likely to increase as additional
revenues are gained from increased fan interest in and attendance at games of a winning team.
Second, team values are likely to increase due to an increase in brand value, incremental to
accounting information (Barth et al. 1998). Brand value is arguably especially important in the
professional sports setting due to the visibility of sports teams through media coverage and the
enthusiasm, loyalty, and commitment of sports fans to their teams (Ertug & Castellucci 2013;
Gladden et al. 2001). If fielding a winning team and a more popular team through greater
expenditures on team-level player compensation leads to higher brand value, then team values
could increase (Madden et al. 2006; Keller 1997; Aaker & Jacobson 1994).
I employ a sample of 386 firm-years covering 13 seasons of the NBA from the 2001
season to the 2013 season. Empirical results show that owners who spend more on team-level
player compensation realize higher winning records. I also find evidence that team-level player
compensation is negatively correlated with current operating income. This empirical result is
consistent with a competing view of team owners’ objectives wherein investing more to field a
team can result in winning more games at the expense of current year profits.
However, empirical results also suggest that team owners are able to improve the overall
financial value of their business by spending more on team-level player compensation.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
11
Specifically, I find that team values are significantly affected by teams’ revenue generation and
brand values, both of which are significantly affected by a team’s winning performance (which
in turn depends on the amount spent on team-level player compensation). Thus, spending more
on team-level player compensation can lead to lower current year operating profits, but can also
lead to higher team values.
Additionally, I find support for timing differences in achieving objectives. As mentioned,
I find a significant positive contemporaneous relationship of team-level player compensation and
how a team performs during a season. I also find a significant positive relationship between
lagged amounts spent on team-level player compensation (beginning three years prior and
continuing up to 10 years prior) and measures of overall financial return. These results suggest
that building a team’s revenue-generating ability and brand value takes time to develop and grow
and is dependent upon the greater investments in team-level player compensation. Overall,
results suggest that NBA team owners who pursue short-term operational profits by spending
less on team-level player compensation likely do so at the expense of real-economic future value.
That is, despite being privately-held firms and not facing agency problems associated with
diffused ownership and lack of owner involvement in management or facing other typical causes
of managerial myopia, managers may pursue myopic behavior.
I explore potential influencing factors on decisions about team-level player
compensation, and find that team owners’ personal net worth has a significant positive effect on
team-level player compensation, even after controlling for the team’s composition and financial
resources available to the team. These results suggest that owners’ personal wealth is positively
associated with long-term thinking or willingness to spend money, which based on prior results,
ultimately leads to greater overall financial returns.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
12
This study contributes to the accounting literature by demonstrating that even within a
private firm setting wherein many of the typical causes of managerial myopia are either absent or
limited, managers can and do take actions that have short-term benefits but which harm
achievement of long-term objectives. Personal net worth is shown to be a significant factor
affecting managers’ investment behavior, even amongst extremely wealthy individuals. This
study also contributes to the accounting literature by highlighting how the measurement of
objectives faced by business managers can affect how the relationship between the objectives
and how manager behavior are understood. Finally, this paper helps resolve tension in the
competing objectives view of sports franchise owner behavior presented in prior sports
management literature by recognizing an alternative measure of the profit objective that leads to
an interdependent conceptualization of objectives. Understanding that objectives can be viewed
as either interdependent or competing depending on the performance measured used, and that
multiple objectives can be achieved but may not be realized contemporaneously, can be
beneficial broadly for business owners and managers involved with strategic decision making,
incentive compensation design, and performance evaluation.
The remainder of this study is organized as follows. Section II provides background
information, related literature, and hypothesis development. Section III provides the research
design, including description of the sample, variables used, and empirical models. Section IV
presents results , Section V concludes, and Section VI discusses future directions and
implications.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
13
II. BACKGROUND AND RELATED LITERATURE
Managerial Myopia
Physiologically, myopia refers to the inability to see things clearly when they are far
away. This concept has been applied to managerial settings such that managerial myopia refers
to a common tendency of managers to focus on short-term rather than longer-term outcomes.
Such myopic behavior typically arises in a business setting from agency problems associated
with information asymmetry and goal conflict between the principal and agent (Eisenhardt 1989;
Jensen & Meckling 1976). For example, prior literature has found that managers frequently take
actions to achieve a short-term objective even at the expense of real-economic value and the
realization of a longer term objective (Graham et al. 2005). Short-term objectives considered in
prior literature include meeting analysts’ earnings forecasts, bonus thresholds, budget targets,
debt covenants, or cash requirements whereas longer-term objectives include building skills and
competencies, creating innovative ideas, improving operational efficiencies, and increasing
owner or shareholder value.
Managerial myopia is considered a first-order effect in that it is assumed to occur
frequently and to have a significant effect on actions taken by managers (Stein 1989).
Managerial myopia has been shown to be affected by contract design, performance measure
characteristics, performance standards or budgets, capital market pressures, ownership structure,
and public accounting reporting requirements (Abernethy et al. 2013; Mizik 2010; Seybert 2010;
Bhojraj & Libby 2005; Graham et al. 2005; Lundstrum 2002; Bushee 1998; Cooper & Selto
1991; Dechow & Sloan 1991; Merchant 1990; Stein 1989; Healy 1985). Despite the many
studies that examine managerial myopia, myopia is difficult to test empirically for at least two
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
14
reasons: 1) difficulty operationalizing behavioral intentions and 2) ex-post assessments of
optimality or inefficiency of managerial actions.
In the first place, it is problematic for researchers to ascertain managers’ intentions and
whether their actions can be called myopic ex-ante. Accessing the intentions of an individual are
notoriously difficult without full disclosure and cooperation from the individual. Furthermore,
without knowing the end outcome of an action, it is hard to label the action as being myopic.
Granted, past experiences can be used as reference point to evaluate an action. For example, it
seems reasonable based on past observations of managerial actions that cutting expenses related
to research and development (R&D), marketing, and/or corporate investments in order to boost
short-term earnings could be labeled myopic behavior. Nevertheless, for any specific instance,
there may be other countervailing reasons for the behavior rendering the decision rational rather
than myopic.
Researchers also face a difficult task of ex-post labeling actions as myopic as there needs
to be a clear economic tradeoff made between short-term and long-term outcomes. This can
require a projection or model of what the optimal outcome would have been or could have been
had actions been taken otherwise. Furthermore, proving that indeed a sacrifice of economic value
has been made to achieve a short-term objective can be difficult. In particular, the passage of
time is required in order to measure long-term objectives and then look back on previous actions
to label them myopic due to the loss of economic value. The proper period of time over which to
measure a long-term outcome can be unclear.
This study presents a setting wherein agency frictions and other common causes of
managerial myopia are limited. Privately held firms have been shown to have less myopic
behavior because they are subject to fewer short-term pressures than publicly held firms,
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
15
although agency problems can still arise depending on ownership concentration and owner
involvement in management (Asker et al. 2014). Thus, the study of privately held firms can be
informative to observe how managers make tradeoffs between multiple objectives and whether
managerial myopia is still present. This setting is also one wherein myopia is likely to be present
as the firm value for a sports franchise and ultimate realization of the economic profits when sold
can be uncertain and elusive. Thus, with the future economic benefit seemingly far away, it can
be more difficult for managers to see it clearly, leading them to focus more on the present, short-
term outcomes. The dataset analyzed herein covers a 13 year time period over which the tradeoff
between short-term and long-term outcomes can be reasonably assessed and inferences drawn
about the sacrifice of long-term economic value for short-term financial gain. I next provide
details about the setting and present hypotheses.
Brief History of NBA
In June of 1946, a group of businessmen established 11 basketball team franchises that
would compete in two divisions, East and West. These owners were mostly from the Arena
Association of America, controlling the arenas in major U.S. cities. Each team paid a $10,000
franchise fee, and their basketball league was known as the Basketball Association of America
(BAA) (Goldaper 1996). Only three years later in 1949, the Basketball Association of America
(BAA) merged with the National Basketball League (NBL, founded in 1937) to form the
National Basketball Association (NBA). The NBA initially had 17 teams from small towns and
large cities, yet due to dwindling fan support the NBA had only 8 teams in 1955 (History.com).
5
In 1955, the NBA introduced the 24-second shot clock which revolutionized the pace and
5
All of these eight franchises are still in the league today (Atlanta Hawks, Boston Celtics, Detroit Pistons, Golden
State Warriors, Los Angeles Lakers, New York Knicks, Sacramento Kings, and Philadelphia 76ers), although some
have moved cities.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
16
strategy of the game. Popularity of the sport increased and the league began to expand again. In
the late 1960’s and early 1970’s, the NBA added nine expansions teams. In 1976, the NBA
added four more teams from the disbanding American Basketball Association (ABA). The NBA
has since grown to include 30 teams as of 2014. Figure 1 provides a map showing the locations
of these 30 teams.
FIGURE 1
Geography of NBA Franchises as of 2014
Source: Wikipedia.com, “National Basketball Association”, accessed at
http://en.wikipedia.org/wiki/National_Basketball_Association
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
17
NBA Owners and Managerial Objectives
The National Basketball Association (NBA) in the sports entertainment industry presents
a setting wherein owners of large privately-held firms face multiple objectives in a competitive
environment (Wolfe et al. 2005). Furthermore, as very closely-held private firms whose
management and control of decisions are restricted to the primary decision agent and residual
claimant (i.e., the team owners), typical agency problems are limited (Fama & Jensen 1985,
1983). Managerial myopia, or the tendency of managers to focus on short-term rather than
longer-term outcomes, typically arises from agency problems associated with information
asymmetry and goal conflict between the principal and agent (Eisenhardt 1989; Jensen &
Meckling 1976). Privately held firms are subject to fewer short-term pressures known to be
associated with myopic behavior than publicly held firms, although agency problems can still
arise depending on ownership concentration and owner involvement in management (Asker et al.
2014). The NBA is unique in that it presents a setting wherein agency frictions and other causes
of managerial myopia are limited and can therefore be useful to observe how owners make
tradeoffs between multiple objectives and whether managerial myopia is still present.
Owners of the NBA franchises are extremely wealthy individuals from diverse
backgrounds, such as lawyers, venture capitalists, entrepreneurs, bankers, vehicle dealers,
professional athletes, real estate developers, and business owners. As of 2013, NBA owners had
an average net worth around $2.7 billion.
6
NBA team owners own a unique asset to which they
have a long-term commitment and from which they derive many benefits such as earning money
and gaining satisfaction from winning games as well as the power, prestige, and position of
being a sports franchise owner (Simmons 2014a; Rosner & Shropshire 2004; Zimbalist 2003).
6
Owners’ net worth as of 2013 is obtained from Forbes.com (url:
http://www.forbes.com/sites/tomvanriper/2013/01/23/the-nbas-billionaire-owners/).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
18
As of 2013, NBA owners had an average tenure of 13 years with a standard deviation just over
10 years indicating a wide range of ownership tenure, and an average age of 61 years old with
youngest being age 35 and the oldest being age 87 (HoopsHype 2014; Van Riper 2013).
Much like managers of any business, NBA team owners make decisions about how to
pursue their objectives, which in this setting include clearly defined strategic objectives of
winning games and maximizing profits (Wolfe et al. 2005; Rosner & Shropshire 2004; Fort &
Quirk 1995; Quirk & Fort 1992). Prior research in the sports management literature has generally
viewed sports team owners’ objectives of winning games and earning profits to be competing as
opposed to interdependent (Leeds and von Allmen 2013; Dietl et al. 2011; Fort & Quirk 2004,
1995; Alexander 2001; Vrooman 1995; Quirk & Fort 1992). On the one hand, team owners face
an objective of winning games which places demand on expending current-period financial
resources to maximize games won. On the other hand, team owners also have the objective of
maximizing profits (Fort & Quirk 2004; Keidel 1984). Thus, from this perspective, team owners
must make an economic trade-off between current-period profits and winning games.
Player Compensation Decisions
Both the winning and profit objectives are arguably linked to team owners’ decisions
regarding the most substantial expense teams incur: team-level player compensation. It is clear
from recent labor disagreements that player compensation is a weighty matter for NBA teams,
and that it significantly influences teams’ financial performance.
7
Team-level player
7
For example, the 2011 NBA lockout was largely centered on owners trying to reduce the amount of money spent
on players’ compensation so that their teams would be more profitable as many teams were argued by the NBA to
be losing money. Owners were successful in reducing the amount of basketball-related income that would be
allotted for player compensation from 57% down to 50%. For the 2010-2011 season, this difference represented an
estimated basketball-related income transfer from players to owners of $270 million (Rishe 2011). The 1998 NBA
lockout, where many NBA teams also claimed to be losing money, was also focused in part on limiting player
salaries (Wise 1998).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
19
compensation is a particularly useful measure to examine owner behavior because team-level
player compensation is a large, separately measurable and focal expense, and team owners have
clear involvement in the player compensation levels for their teams. Further, by observing team-
level player compensation, inferences can be made about team owners’ emphasis on their
strategic objectives.
In this paper, I empirically examine NBA teams’ resource-allocation decisions directed
towards objectives of winning games and maximizing profits by analyzing the effect of team-
level player compensation decisions on both on-the-court (winning) and off-the-court (financial
performance) outcomes. NBA teams each operate under the same Collective Bargaining
Agreement (CBA) with the National Basketball Players Association (NBPA). Under the CBA,
there is a range of total player compensation that teams can spend to field their teams, meaning
there is a lower and upper bound to the amount of money that can be spent on player
compensation. There is significant variation across teams in the amount spent on team-level
player compensation.
The upper bound on team-level player compensation, commonly referred to as the “salary
cap”, is actually a soft cap that allows teams to exceed the salary cap under certain
circumstances. For example, the soft cap allows teams to retain their own players or sign their
rookie draft picks when doing so would otherwise cause them to exceed a hard salary cap (Coon
2012). Despite the variety of exceptions for which a team can exceed the salary cap, teams can
face a luxury tax penalty for high spending amounts on player compensation.
8,9
This paper does
8
The tax threshold level is higher than the salary cap level. For example, the salary cap for the 2011-2012 season
was set at $58.044 million while the tax level was set at $70.307 million (Coon 2012).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
20
not attempt to detail all of the rules and complexities of player compensation.
10
Rather, the paper
focuses on the team owners’ discretion and willingness to pay a certain level of player
compensation to field their teams.
Team owners typically hire general managers to help make basketball personnel
decisions. Even though owners may not specifically be making decisions about which players to
sign, trade for, or draft, they do set the tone for which objectives the organization will pursue.
And ultimately, owners must give their consent on the amount of compensation offered to
players.
11
Thus, team owners make decisions about how much to spend on team-level player
compensation while other involved decision makers are likely to assist in making decisions about
who the specific players are.
On-the-court Outcomes (Winning)
I first examine whether greater expenditures on team-level player compensation leads to
greater on-the-court performance. This link is important because it establishes that decisions
about how much teams spend on team-level player compensation do in fact have an effect on a
team’s winning. There are several reasons why spending more on team-level player
compensation might not necessarily lead to winning more games (Berri et al. 2006). Individual
players often play on contracts that could be considered under- or over-paying them for their
9
The luxury tax started in the 2002 – 2003 season. The luxury tax rates are the same over the sample period
examined in this paper. During the sample period, the tax provision was a dollar-for-dollar system such that teams
paid one dollar for each dollar of player compensation exceeding the tax threshold. Terms of the luxury tax penalty
were significantly modified as part of the 2011 collective bargaining agreement with the major changes being
implemented at the beginning of the 2013 – 2014 season, which is not included in the sample. The new tax rates are
an incremental system and include a repeat offender tax, all of which make it more expensive for teams to exceed
tax thresholds.
10
The interested reader is referred to Larry Coon’s website that provides exhaustive details on the CBA (Coon 2012,
url: http://www.cbafaq.com/salarycap.htm).
11
Team owners could be viewed as taking on a role of manager and owner in their franchise because they are the
primary decision agent and the residual claimant (Fama & Jensen 1985).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
21
abilities. For example, when players enter the league, they play on rookie contracts that have
relatively low specified salaries based on their draft position (Coon 2012). If a younger player
performs at a high level (e.g., like making the all-star team), then the team could win more
games despite having a lower-salaried player that is likely being underpaid for his performance.
Players can also sign very lucrative contracts, yet fail to live up to the expected level of
performance due to a variety of factors such as deterioration of skills from aging, injuries, or lack
of motivation. Teams can also make poor judgments about a player’s ability or potential and
thereby overpay the player (Quirk and Fort 1992). In the NBA, player contracts are typically
fully guaranteed which means teams can be stuck with paying a player whose contribution to the
team winning is not commensurate with a high paying contract (Simmons 2014b).
12, 13
This
could lead to teams having high payrolls and lower winning percentages.
On the other hand, player salary can be a considered a reasonable, although somewhat
noisy signal of ability. Especially given the free agent market in the NBA, high performing
players are able to command and negotiate for the higher paying contracts. Teams also have a
unique right (known as the “Bird Exception”) to sign their own players to higher contract
amounts than the player could be given by a new team (Coon 2012). Teams are likely to try and
retain those players who are contributing to winning games and therefore would have higher
player compensation. Hence, teams with higher team-level player compensation are likely to
12
In contrast, the National Football League (NFL) does not have guaranteed contracts, so it is less clear how much
money owners will have to actually commit to a player when a contract is signed (Bryant 2014). Additionally,
compared to the Major League Baseball (MLB) which has a fair amount of player mobility due to minor league
arrangements, the NBA has more identifiable rosters of regular players which makes expected commitments to
team-level player compensation more apparent. Furthermore, although MLB does not have a salary cap which might
permit salaries to better reflect a player’s ability, MLB players do not become free agents until after six years of
service (compared to 2-3 years in the NBA), effectively delaying the usefulness of salary as a signal of ability. MLB
does have an arbitration process which allows some players to circumvent the six year requirement. The interested
reader is directed to url: http://mlb.mlb.com/pa/info/faq.jsp.
13
NBA contracts also make much less extensive use of performance incentives compared to the NFL and MLB
player contracts, which makes the expected level of money committed to a player more clear in the NBA (Stiroh
2007). This also limits revere causality of team-level player compensation being dependent upon how a team
performs.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
22
have, on average, better players, which likely leads to winning more games. That being said, this
study does not seek to examine or explain decisions about individual player contracts, but rather
focuses on the team-level player compensation. Because player salary is generally a reasonable
reflection of players’ ability levels, I hypothesize that teams with higher levels of team-level
player compensation will have players of higher ability leading to better team performance.
H
1
: Team-level player compensation has a positive effect on team performance.
Off-the-court Outcomes (Financial Performance)
Inasmuch as decisions about team-level player compensation can have a significant effect
on a team’s basketball performance, then team owners can make strategic compensation choices
in pursuit of their winning and profit objectives. However, the manner in which objectives are
defined and measured is critical to understanding manager behavior and what objectives will be
pursued (Abernethy et al. 2013; Artz et al. 2012). The sports management literature has defined
the profit objective to be the teams’ current year profits (revenue minus expenses) (Késenne
2014; Leeds and von Allmen 2013). This definition is convenient in that it fits within economic
models of profit maximizing behavior and organizations often use current year profits as a
performance measure for goal setting, evaluation, and compensation purposes (Késenne 2014;
Merchant & Van der Stede 2012; Fort and Quirk 2004, 1995). However, measuring the profit
objective using current-period profit may lead to myopic decisions, or interpreting owner
behavior to be myopic, inasmuch as winning games in the current period could lead to greater
profits in the future.
Thus, how objectives are measured is informative to understanding the
interchange between the objectives, manager behavior, and which objectives will be pursued.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
23
More specifically, an alternative conceptualization of the profit objective is to define and
measure profits as the long-term financial return that comes from operational profits and changes
in team value over the life of the franchise investment. This way of defining profits recognizes
the value derived from capital appreciation, takes a longer performance measurement horizon,
and provides a more holistic view of factors that could affect managers’ decision-making and
behavior (Feltham & Xie 1994). By measuring the profit objective in this manner, it is possible
that the relationship between team owners’ objectives of winning games and maximizing profits
could be viewed as interdependent rather than competing (Silver 2014; Zimbalist 2003; Gladden
et al. 2001; Jensen 2001). Rather than winning and profits being alternative objectives to satisfy,
the realization of the one objective (long-term profits) could be contingent upon investments
leading to the satisfaction of the other objective (winning).
I next discuss possible mechanisms by which winning more games can increase long-
term profits. I argue that improved on-the-court product (winning) leads to greater financial
returns from appreciated team values and operating profits (financial performance) through two
mechanisms: (1) increasing revenue generation and (2) increasing brand values (Chandra & Ro
2008; Gladden et al. 2001; Barth et al. 1998). Figure 2 presents an overall model of the proposed
hypotheses related to these mechanisms.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
24
FIGURE 2
Hypothesized On-the Court (Winning) and Off-the-Court (Financial) Outcomes
Note: This figure lists only the main variables of interest. Numerous control variables are included and are listed in the tabulated regressions. Dotted line
indicates a mediated path. Solid lines indicate a main effect.
H4
H5
H
7
Player
Expenses
Home
Attendance
Team
Performance
Revenue
Team Value
Brand Value
(Road
Attendance)
H
1 H
2
H
6
H3
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
25
Teams who win more often are likely to draw greater fan interest. When fans are
interested in and excited about their team, then attendance at home games and all the
accompanying revenues (gate receipts, concessions, parking, memorabilia sales, sponsorships,
advertisements, etc.) are likely to increase. Home game revenues make up a substantial portion
of a team’s total revenues. Therefore, I hypothesize that team performance has a positive effect
on revenue that is mediated by home attendance.
H
2
: Team performance has a positive effect on revenue as mediated by home attendance.
Greater team performance (i.e., winning more games) can also increase the quality of a
team’s brand. Higher brand quality means that consumers have a high level of awareness and a
willingness to either pay higher product prices or make purchases more frequently (Barth et al.
1998; Simon & Sullivan 1993). When a team wins more often, consumer awareness of the team
increases because the team is featured more prominently in media outlets such as newspapers,
web pages, newscasts, and radio shows. Host teams will be able to charge higher prices for
tickets and sell more tickets to games featuring teams with higher brand quality (Berri et al.
2006; Berri et al. 2004; Benjamin & Podolny 1999). I therefore hypothesize that team
performance has a positive effect on brand value.
14
H
3
: Team performance has a positive effect on brand value.
14
The proxy measure for brand value used in this study is a team’s average road game attendance during the regular
season. This is discussed in further detail in Section III.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
26
I also note that a team’s brand value is likely directly affected by greater expenditures on
team-level player compensation. Players who have higher public profiles, like all-star players,
generally are able to command higher salaries. Therefore, teams who spend more on team-level
player are likely to have a higher level of awareness of their team due to the popularity of their
players. Thus, while greater team-level player compensation can affect a team’s winning
percentage, it is also likely to increase the brand value of team. As such, I hypothesize that team-
level player compensation has a positive effect on brand value.
H
4
: Team-level player compensation has a positive effect on brand value.
From both winning more games and having a higher profile team, a team’s greater brand
value is likely to increase revenues. This can happen through several revenue streams, such as
increases in memorabilia sales and sponsorships, as fans and business will want to be associated
with the team. Also, teams with higher brand values are likely able to negotiate larger local-
television broadcast deals due to the consumer demand to watch their games (Settimi 2014;
Keller 1997). While revenues from nationally broadcast deals with the NBA (such as those with
ESPN, ABC, and TNT) are equally shared among the teams, revenues from local television
broadcast deals are kept by the team. These local TV deals are increasingly a major source of
revenue for teams, and teams with higher brand values are likely able to be in stronger
negotiation positions (Keller 1997).
15
As such, I hypothesize that brand value has a positive
effect on revenue.
15
The Los Angeles Lakers, who have arguably the strongest brand value in the NBA (Giratikanon et al. 2014),
signed a 20-year local television broadcast rights contract with Time Warner Cable in 2011. This contract averages
over $200 million a year and included the launch of two new regional sports networks, one in English and one in
Spanish (Settimi 2014).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
27
H
5
: Brand value has a positive effect on revenue.
The increased ability of a team to generate revenue can lead to higher team values (Lee
1999). While sport franchise values often depend less on current profitability than other factors
such as metropolitan area and stadium arrangements, the ability to generate greater revenues is a
significant value factor (Badenhausen 2014; Silver 2014; Alexander & Kern 2004). This is likely
because revenue for sports franchises can contain information about future earnings and cash
flows that is lost when revenue and expenses are aggregated as earnings (Chandra & Ro 2008).
In this sense, NBA teams are similar to technology firms where revenue is an informative
valuation factor because of volatile earnings from mismatching of current expenditures and
future revenue benefits (Amir & Lev 1996). I hypothesize that revenue is more informative to
team values than earnings.
H
6
: Revenue is more informative to team values than earnings.
Finally, brand values can have an incremental effect on firm values over accounting
information and have been shown to be associated with positive returns (Barth et al. 1998; Aaker
& Jacobson 1994). In addition to improving revenues through improved sales volume and
product price, strong brand values can bring benefits such as greater loyalty from customers,
stronger negotiation positions, less vulnerability to competitive marketing actions, greater trade
cooperation and support, possible licensing opportunities, and additional brand extension
opportunities (Keller 1997). Brand value is arguably especially important in the professional
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
28
sports setting due to the visibility of sports teams through media coverage and the enthusiasm,
loyalty, and commitment of sports fans to their teams (Ertug & Castellucci 2013). Therefore, I
hypothesize that brand value has a positive effect on team value.
H
7
: Brand value has a positive effect on team value.
III. RESARCH DESIGN
Sample and Variables
Data related to the NBA has been used in studies across disciplines, including
management, organizational behavior, economics, and social psychology (Ertug & Castellucci
2013; Simmons & Berri 2011; Wang 2009; Wolfe et al. 2005; Berman et al. 2002; Taylor &
Trogdon 2002; Hausman & Leonard 1997; Staw & Hoang 1995; Harder 1992; Pfeffer & Davis-
Blake 1986). I employ a sample of 386 firm-years covering 13 seasons of the NBA, from the
2001 season to the 2013 season. There are 4 seasons with 29 teams (seasons ending 2001 – 2004)
and 9 seasons with 30 teams (seasons ending 2005 – 2013) after the addition of the Charlotte
Bobcats franchise. This represents this entire population of NBA teams.
The NBA does not release its financial information publicly. The individual teams are
closely held, private businesses with no legal requirement to release their financial information
publicly. However, Forbes provides yearly estimates of NBA teams’ finances. This data source
has been used in various academic studies (Ertug & Castellucci 2013; Alexander & Kern 2004;
Quirk and Fort 1992). From Forbes, I gather financial variables including team value, revenue,
operating income, and player expenses. I gather additional team and arena information from a
variety of sources including nba.com, espn.com, team websites, and basketball-reference.com.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
29
Luxury tax information is based on calculations from Mark Deeks at www.shamsports.com. The
Fan Cost Index is produced by www.teammarketing.com and obtained from Rodney Fort’s
Sports Business Data website. Variable definitions are included in the Appendix.
Most variables in this study are direct measures of the construct of interest. However, to
capture brand value, I use a team’s average road game attendance as a proxy measure. By
scheduling convention, teams typically play two road games each season against teams within
their conference (East or West) and one road game against teams in the other conference (East or
West). Each team plays a total of 41 road games in a regular season. Thus, each team within a
conference will play the same teams on the road the same number of times. I argue that average
regular season road game attendance is a reasonably proxy for a team’s brand value as higher
average road game attendance compared to other teams with similar schedules indicates a greater
level of awareness and higher demand to watch a given team play.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
30
TABLE 1
Road Attendance Values by Team
Team
Sample Period
Average Road
Attendance
Std. Dev.
Sample Period
Average Road
Attendance Rank
Std. Dev.
LA Lakers 19,150 390 1.69 0.63
Miami 18,044 1,210 7.38 8.94
Boston 17,856 813 8.08 5.79
Philadelphia 17,732 680 10.23 6.39
New York 17,673 411 8.08 3.25
Cleveland 17,579 1,313 13.62 12.28
Houston 17,482 584 10.92 6.10
Sacramento 17,460 434 12.69 7.88
Phoenix 17,421 652 10.46 6.20
San Antonio 17,389 485 11.31 4.17
Chicago 17,324 804 13.92 7.91
Orlando 17,308 364 12.38 6.49
Dallas 17,268 564 13.85 7.29
Denver 17,253 593 12.62 5.75
Detroit 17,177 803 15.85 8.96
Washington 17,132 1,212 20.31 9.55
Oklahoma City 17,086 680 17.31 8.10
Brooklyn 17,082 558 15.85 6.99
Minnesota 17,034 576 17.23 9.12
Indiana 17,029 436 17.31 6.49
Utah 16,971 284 17.77 5.18
LA Clippers 16,910 729 20.00 8.85
New Orleans 16,875 369 19.15 6.54
Charlotte 16,853 448 22.44 5.32
Atlanta 16,804 369 21.08 3.97
Milwaukee 16,791 312 21.62 5.04
Portland 16,789 389 20.77 6.42
Toronto 16,786 380 21.92 7.97
Memphis 16,714 504 22.69 4.94
Golden State 16,630 452 23.92 4.68
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
31
To test the validity of this measure, I take the average road game attendance for each
team over the sample period and correlate it with the average values of a team’s Facebook Likes
and Twitter Followers as of May 2014. I find that teams’ average road game attendance over the
sample period is significantly and positively correlated with Facebook Likes (r = 0.394) and
Twitter Followers (r = 0.488).
16
Furthermore, in a multiple regression analysis presented later in
the analysis section (see Table 8), I find that factors that are likely determinants of a team’s
brand value are indeed determinants of road attendance. These include variables such as the
number of All-Star players on a team, current year winning percentage, amount spent on player
compensation, number of NBA championships won by a franchise, and size of metropolitan area
(Ertug & Castelluci 2013; Berri & Schmidt 2006). The overall sample average for road game
attendance is 17,257 with a standard deviation of 800, minimum value of 15,761, and maximum
value of 20,102. Table 1 provides the sample average road attendance for each team along with
a sample average of the team’s yearly road attendance ranking.
Descriptive Statistics and Correlations
Tables 2 and 3 provide descriptive statistics and correlations for the variables used in the
study. Panel A of Table 4 provides details about league-wide team-level player compensation for
each year within the sample period while Panel B of Table 4 provides details about player
compensation for each team over the sample period.
16
For a geographic display of NBA teams’ fans based on Facebook Likes, see the interactive NBA Fan Map from
the Giratikanon et al. (2014).
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
32
TABLE 2
Descriptive Statistics
Sample size 386 team years
4 seasons with 29 teams (seasons ending 2001 – 2004)
9 seasons with 30 teams (seasons ending 2005 – 2013)
Variable Mean Std. Dev. Min Max
Home Attendance
17,257 2,237 11,286 22,272
Road Attendance
17,257 800 15,761 20,102
All-Stars
0.87 0.85 0 4
Playoffs
t-1
0.54 0.50 0 1
Winning Percentage
0.50 0.15 0.11 0.82
Average Player Age
26.86 1.71 22.7 32
Population
5,365,747 4,751,636 1,130,300 19,069,800
Conference
0.50 0.50 0 1
NBA Championships
2.22 4.11 0 17
Fan Cost Index
280.84 67.82 173.72 643.78
Arena Capacity
19,477 1,840 16,285 35,000
Arena Age with Renovation
9.90 5.86 1 29
Player Expenses
65,300,000 13,800,000 28,000,000 121,000,000
Revenue
115,000,000 34,800,000 53,000,000 295,000,000
Operating Income
9,264,508 20,072,210 -85,000,000 96,000,000
Team Value
366,000,000
159,000,000
135,000,000
1,400,000,000
Value Change 33,025,974 54,586,842 -121,000,000 350,000,000
Value Change Plus
Operating Income
42,288,571 64,540,144 -118,000,000 416,000,000
Value Return 9.50 % 12.50 % -25.42 % 75.00%
Overall Return 11.69 % 14.35 % -30.74 % 75.87 %
Note: This table presents descriptive statistics for the variables employed in the primary tests. All variables are as
defined in the Appendix.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
33
TABLE 3
Correlations
Variable
Home
Attendance
Road
Attendance
All-
Stars
Playoffs t-1
Winning
%
Player
Age
Population Conf.
NBA
Champion
-ships
Fan
Cost
Index
Arena
Capacity
Arena
Age
with
Reno.
Player
Expenses
Rev.
Operating
Income
Road
Attendance
0.45
**
All-Stars 0.32
**
0.55
**
Playoffs t-1 0.41
**
0.35
**
0.43
**
Winning % 0.47
**
0.50
**
0.66
**
0.53
**
Player Age 0.33
**
0.33
**
0.45
**
0.55
**
0.52
**
Population 0.16
**
0.21
**
0.06 -0.05 0.05 0.08
Conference -0.02 0.02 -0.00 -0.02 -0.17
**
-0.03 0.20
**
NBA
Championships
0.25
**
0.44
**
0.28
**
0.17
**
0.20
**
0.16
**
0.24
**
0.06
Fan Cost Index 0.40
**
0.53
**
0.31
**
0.23
**
0.25
**
0.32
**
0.60
**
0.06 0.51
**
Arena Capacity 0.47
**
-0.02 -0.05 0.07 0.03 0.06 0.15
**
0.25
**
0.07 0.05
Arena Age with
Renovation
-0.10 -0.11
**
-0.19
**
-0.06 -0.11 -0.16
**
0.16
**
0.10
*
-0.15
**
0.03 -0.02
Player
Expenses
0.31
**
0.37
**
0.14
**
0.24
**
0.24
**
0.25
**
0.24
**
-0.01 0.16
**
0.45
**
-0.02 0.14
**
Revenue 0.55
**
0.56
**
0.27
**
0.24
**
0.31
**
0.22
**
0.43
**
0.05 0.40
**
0.77
**
0.12
*
-0.02 0.63
**
Operating
Income
0.39
**
0.33
**
0.20
**
0.10
*
0.14
**
0.02 0.25
**
0.06 0.38
**
0.48
**
0.14
**
-0.13
*
-0.16
**
0.58
**
Team Value 0.43
**
0.50
**
0.24
**
0.19
**
0.24
**
0.18
**
0.37
**
0.02 0.37
**
0.71
**
0.08 0.02 0.52
**
0.89
**
0.55
**
Note: this table presents Pearson correlations. All variables are as defined in the Appendix. The symbols * and ** denote statistical significance (two-tailed) at
the 5%, and 1% levels, respectively.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
34
TABLE 4
Player Expenses
Panel A: League Wide Average Team Level Player Compensation by Year
League Wide
Team Player
Expenses
Season Ending
2001 2002 2003 2004 2005 2006 2007
Average $ 53,482,759 $ 52,551,724 $ 58,448,276 $ 59,551,724 $ 61,300,000 $ 66,166,667 $ 68,700,000
Std. Dev. $ 12,656,508 $ 10,432,046 $ 11,431,742 $ 12,376,782 $ 14,825,770 $ 15,186,390 $ 13,022,924
Min $ 33,000,000 $ 34,000,000 $ 46,000,000 $ 38,000,000 $ 28,000,000 $ 37,000,000 $ 46,000,000
Max $ 90,000,000 $ 86,000,000 $ 100,000,000 $ 96,000,000 $ 100,000,000 $ 118,000,000 $ 121,000,000
Season Ending
2008 2009 2010 2011 2012 2013 2001 to 2013
Average $ 72,100,000 $ 75,400,000 $ 73,433,333 $ 71,266,667 $ 59,466,667 $ 76,200,000 $ 65,331,606
Std. Dev. $ 10,684,665 $ 9,137,267 $ 8,826,619 $ 10,523,974 $ 8,033,264 $ 10,283,566 $ 13,828,218
Min $ 58,000,000 $ 59,000,000 $ 59,000,000 $ 50,000,000 $ 45,000,000 $ 61,000,000 $ 28,000,000
Max $ 103,000,000 $ 99,000,000 $ 91,000,000 $ 95,000,000 $ 76,000,000 $ 110,000,000 $ 121,000,000
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
35
Panel B: Team Level Player Compensation by Team
Team Player Expenses 2001 to 2013 Luxury Tax
Team
Average
Std. Dev.
Min
Max
Seasons
paying tax
in sample
period
Total tax paid
in sample
period
New York $ 92,076,923 $ 15,574,881 $ 68,000,000 $ 121,000,000 8 $ 205,250,551
Dallas $ 82,461,538 $ 14,785,214 $ 58,000,000 $ 103,000,000 9 $ 150,530,433
Portland $ 78,153,846 $ 11,238,670 $ 63,000,000 $ 100,000,000 4 $ 89,052,474
LA Lakers $ 77,846,154 $ 14,559,339 $ 56,000,000 $ 110,000,000 8 $ 113,676,992
Orlando $ 68,769,231 $ 16,995,852 $ 40,000,000 $ 95,000,000 3 $ 38,951,508
Boston $ 68,615,385 $ 13,853,630 $ 48,000,000 $ 88,000,000 7 $ 47,275,853
Philadelphia $ 68,230,769 $ 9,426,504 $ 53,000,000 $ 83,000,000 2 $ 17,857,635
Miami $ 67,076,923 $ 11,842,449 $ 49,000,000 $ 90,000,000 5 $ 36,014,230
Brooklyn $ 66,000,000 $ 9,574,271 $ 54,000,000 $ 94,000,000 3 $ 28,004,030
San Antonio $ 65,000,000 $ 11,633,286 $ 46,000,000 $ 84,000,000 5 $ 12,597,554
Indiana $ 64,615,385 $ 9,224,410 $ 48,000,000 $ 79,000,000 3 $ 8,889,087
Minnesota $ 64,538,462 $ 9,106,915 $ 50,000,000 $ 75,000,000 3 $ 24,657,542
Cleveland $ 63,769,231 $ 16,614,019 $ 46,000,000 $ 94,000,000 3 $ 43,126,121
Milwaukee $ 63,769,231 $ 6,029,840 $ 54,000,000 $ 75,000,000 1 $ 4,734,000
Houston $ 63,615,385 $ 9,717,154 $ 49,000,000 $ 78,000,000 1 $ 757,145
Toronto $ 62,923,077 $ 11,124,010 $ 41,000,000 $ 78,000,000 2 $ 6,771,836
Denver $ 62,769,231 $ 13,417,363 $ 43,000,000 $ 87,000,000 3 $ 21,157,439
Sacramento $ 62,692,308 $ 9,818,220 $ 45,000,000 $ 73,000,000 2 $ 30,518,745
Phoenix $ 62,538,462 $ 9,820,178 $ 45,000,000 $ 80,000,000 4 $ 15,632,239
Memphis $ 61,769,231 $ 8,757,414 $ 45,000,000 $ 75,000,000 2 $ 11,297,452
Utah $ 61,461,538 $ 12,332,987 $ 38,000,000 $ 80,000,000 2 $ 8,103,619
Golden State $ 61,384,615 $ 10,145,101 $ 45,000,000 $ 77,000,000 0 $ -
Detroit $ 60,615,385 $ 11,288,183 $ 40,000,000 $ 76,000,000 1 $ 756,627
Washington $ 60,461,538 $ 8,646,920 $ 49,000,000 $ 73,000,000 0 $ -
Oklahoma City $ 59,230,769 $ 7,928,365 $ 46,000,000 $ 74,000,000 0 $ -
Chicago $ 59,076,923 $ 12,277,768 $ 33,000,000 $ 75,000,000 1 $ 3,932,336
New Orleans $ 59,076,923 $ 10,119,795 $ 45,000,000 $ 74,000,000 0 $ -
Atlanta $ 58,538,462 $ 11,398,943 $ 42,000,000 $ 75,000,000 2 $ 4,382,199
Charlotte $ 55,333,333 $ 15,748,016 $ 28,000,000 $ 73,000,000 0 $ -
LA Clippers $ 54,461,538 $ 13,787,769 $ 33,000,000 $ 80,000,000 0 $ -
Note: Luxury tax information is based on calculations from Mark Deeks at www.shamsports.com.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
36
Over the sample period, the average amount of player compensation is $65.3 million per
team with a standard deviation of $13.8 million. The average operating income is $9.3 million
with a standard deviation of $20.0 million. The average revenue for a team is $115 million with a
standard deviation of $34.8 million. Team values have an average of $366 million with a
standard deviation of $159 million. Figure 3 provides times series graphs of league-wide
averages for teams’ player expenses, operating income, revenue, and team values. Panels A, B,
and C of Table 5 provide details about league-wide profit measures per team for each year within
the sample period, including Operating Income, Value Return which is the change in team value
divided by the prior year’s team value, and Overall Return which is the change in team value
plus operating income divided by the prior year’s team value. Panel D of Table 5 provides details
about these profit measures for each team over the sample period.
FIGURE 3
League-wide Averages of Select Variables
Panel A: Average Team Player Expenses
Note: Total team level amount spent on players’ compensation gathered from Forbes. Dotted lines represent 95%
confidence intervals.
$-
$20,000,000
$40,000,000
$60,000,000
$80,000,000
$100,000,000
$120,000,000
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Average Team Player Expenses
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
37
Panel B: Average Operating Income
Note: Team level operating income is estimated by Forbes. Dotted lines represent 95% confidence intervals.
Panel C: Average Revenue
Note: Total revenue, net of revenue sharing and stadium debt service, is estimated by Forbes. Dotted lines represent
95% confidence intervals.
$(60,000,000)
$(40,000,000)
$(20,000,000)
$-
$20,000,000
$40,000,000
$60,000,000
$80,000,000
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Average Operating Income
$-
$50,000,000
$100,000,000
$150,000,000
$200,000,000
$250,000,000
$300,000,000
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Average Revenue
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
38
Panel D: Average Team Value
Note: Team values are the estimated values of a team by Forbes. Dotted lines represent 95% confidence intervals.
$-
$200,000,000
$400,000,000
$600,000,000
$800,000,000
$1,000,000,000
$1,200,000,000
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
Average Team Value
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
39
TABLE 5
Profit Measures
Panel A: League Wide Average Team Operating Income by Year
League Wide Season Ending
Operating Income 2001 2002 2003 2004 2005 2006 2007
Average $ 7,906,897 $ 14,013,793 $ 1,434,483 $ 3,289,655 $ 16,966,667 $ (746,667) $ 1,263,333
Std. Dev. $ 9,914,633 $ 11,938,166 $ 27,895,728 $ 11,882,230 $ 31,373,702 $ 8,566,645 $ 13,683,781
Min $ (16,000,000) $ (17,000,000) $ (85,000,000) $ (23,000,000) $ (42,000,000) $ (20,000,000) $ (25,000,000)
Max $ 29,000,000 $ 39,000,000 $ 40,000,000 $ 28,000,000 $ 96,000,000 $ 17,000,000 $ 29,000,000
Season Ending
2008 2009 2010 2011 2012 2013 2001 to 2013
Average $ 22,750,000 $ 11,633,333 $ 14,946,667 $ 5,263,333 $ 16,043,333 $ 5,326,667 $ 9,264,508
Std. Dev. $ 17,336,721 $ 17,920,290 $ 16,811,628 $ 19,823,853 $ 27,875,359 $ 15,784,124 $ 20,072,210
Min $ (12,000,000) $ (17,000,000) $ (26,000,000) $ (34,000,000) $ (25,000,000) $ (24,000,000) $ (85,000,000)
Max $ 66,000,000 $ 64,000,000 $ 47,000,000 $ 52,000,000 $ 59,000,000 $ 47,000,000 $ 96,000,000
Note: Operating Income is the team’s total operating profits, excluding interest, taxes, depreciation, and amortization.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
40
Panel B: League Wide Average Team Value Return by Year
League Wide Season Ending
Value Return 2001 2002 2003 2004 2005 2006 2007
Average 8% 10% 7% 8% 11% 8% 9%
Std. Dev. 11% 10% 17% 10% 12% 10% 14%
Min -11% -6% -13% -5% -6% -11% -13%
Max 30% 28% 75% 33% 41% 34% 37%
Season Ending
2008 2009 2010 2011 2012 2013 2001 to 2013
Average 12% 9% 9% 12% 7% 13% 9%
Std. Dev. 12% 12% 12% 14% 13% 15% 12%
Min -1% -9% -25% -25% -13% -8% -25%
Max 40% 36% 35% 44% 33% 51% 75%
Note: Value Return is the percent return in the team’s value from the prior year.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
41
Panel C: League Wide Average Team Overall Return by Year
League Wide Season Ending
Overall Return 2001 2002 2003 2004 2005 2006 2007
Average 11% 14% 6% 9% 14% 8% 9%
Std. Dev. 12% 11% 21% 11% 15% 11% 15%
Min -10% -7% -31% -8% -6% -13% -15%
Max 36% 36% 76% 33% 52% 32% 40%
Season Ending
2008 2009 2010 2011 2012 2013 2001 to 2013
Average 17% 12% 13% 13% 10% 14% 12%
Std. Dev. 12% 15% 14% 16% 16% 15% 14%
Min 0% -13% -18% -25% -14% -14% -31%
Max 44% 48% 43% 45% 39% 55% 76%
Note: Overall Return is the percent return based on the change in team value and operating income.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
42
Panel D: Average Operating Income, Value Return, and Overall Return by Team
2001 to 2013 Operating Income Value Return Overall Return
Team Average Min Max Average Min Max Average Min Max
Chicago $ 47,869,231 $ 34,000,000 $ 59,000,000 10% -2% 33% 21% 10% 39%
LA Lakers $ 39,000,000 $ 23,000,000 $ 66,000,000 11% -1% 40% 18% 4% 44%
New York $ 28,230,769 $ (42,000,000) $ 96,000,000 11% -4% 41% 15% -4% 52%
Houston $ 26,538,462 $ 4,000,000 $ 64,000,000 11% -6% 36% 18% 2% 48%
Phoenix $ 24,769,231 $ 13,000,000 $ 40,000,000 7% -5% 26% 14% -1% 33%
Detroit $ 23,876,923 $ 7,700,000 $ 47,000,000 6% -25% 28% 12% -18% 36%
Boston $ 17,761,538 $ 4,200,000 $ 47,000,000 12% -3% 51% 17% 0% 55%
San Antonio $ 15,769,231 $ (5,000,000) $ 39,000,000 10% -4% 26% 15% 0% 33%
Cleveland $ 15,000,000 $ 3,000,000 $ 33,000,000 9% -25% 32% 14% -25% 38%
Toronto $ 14,492,308 $ (1,000,000) $ 29,000,000 11% -6% 28% 15% -7% 36%
LA Clippers $ 13,269,231 $ 9,100,000 $ 22,000,000 11% -1% 34% 17% 3% 37%
Miami $ 12,153,846 $ (6,000,000) $ 29,000,000 11% -7% 37% 14% -6% 40%
Golden State $ 12,000,000 $ (3,000,000) $ 43,000,000 13% -6% 35% 16% -2% 43%
Washington $ 9,176,923 $ (6,000,000) $ 29,000,000 7% -11% 30% 10% -10% 36%
Oklahoma City $ 8,923,077 $ (9,000,000) $ 33,000,000 10% -5% 36% 12% -4% 45%
Utah $ 6,846,154 $ (16,000,000) $ 28,000,000 7% -4% 29% 10% -7% 33%
Sacramento $ 5,384,615 $ (17,000,000) $ 21,000,000 11% -13% 75% 12% -14% 76%
New Orleans $ 4,507,692 $ (8,000,000) $ 22,000,000 9% -6% 27% 11% -6% 38%
Philadelphia $ 1,723,077 $ (10,000,000) $ 18,000,000 6% -5% 33% 7% -8% 33%
Denver $ 1,676,923 $ (26,000,000) $ 25,000,000 9% -2% 35% 9% -6% 39%
Orlando $ 846,154 $ (23,000,000) $ 13,000,000 10% 0% 22% 11% -4% 25%
Atlanta $ (69,231) $ (19,000,000) $ 13,000,000 7% -8% 34% 7% -14% 33%
Indiana $ (1,000,000) $ (17,000,000) $ 12,000,000 6% -9% 35% 6% -13% 39%
Milwaukee $ (2,469,231) $ (16,000,000) $ 11,000,000 9% -9% 30% 8% -11% 33%
Minnesota $ (2,769,231) $ (20,000,000) $ 9,000,000 8% -11% 34% 7% -13% 32%
Charlotte $ (4,888,889) $ (25,000,000) $ 12,000,000 5% -8% 30% 2% -10% 32%
Dallas $ (8,300,000) $ (34,000,000) $ 38,000,000 13% -4% 44% 10% -8% 45%
Brooklyn $ (8,861,538) $ (24,000,000) $ 5,000,000 14% -13% 48% 11% -14% 44%
Memphis $ (9,230,769) $ (25,000,000) $ 11,000,000 12% -13% 40% 8% -15% 36%
Portland $ (18,646,154) $ (85,000,000) $ 30,000,000 7% -9% 28% -1% -31% 35%
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
43
Among the univariate correlations of note for this study, I find that team-level player
compensation is positively correlated with winning (r = 0.24, p < 0.01) and significantly
negatively correlated with current operating income (r = – 0.16, p < 0.01). That is, spending
more on team-level player compensation is associated with greater on-the-court performance, yet
at the sacrifice of current profits. This empirical result is consistent with a competing view of
team owners’ objectives wherein investing more to field a team can result in winning more
games at the expense of lower current year profits. Thus, when viewed from a one-season time
horizon with profits defined as operating profits, there is a contemporaneous tradeoff of
objectives. However, empirical results also suggest that winning can improve the overall
financial value of team. Specifically, I find that winning percentage is significantly positively
correlated with home attendance, revenue, and brand value, and that revenue and brand value are
significantly positively related to team value.
Hypotheses Tests
To test my first hypothesis that team-level player compensation has a positive effect on
team performance, I estimate the following regression (model 1):
(1) Team Performance = β
0
+ β
1
Player Expenses + Controls + Year Fixed Effects + ε.
Where team performance is measured as the team’s regular season Winning Percentage
(wins/games played) (Bloom 1999), and Player Expenses is the total team level amount spent on
players’ compensation. As control variables, I include the number of All-Stars on a team,
whether the team made the playoffs in the previous year, the metropolitan population of the
team’s city, the conference the team plays in, and year fixed effects. All variables are defined in
the Appendix.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
44
My second hypothesis is that team performance has a positive effect on revenue as
mediated by home attendance. To test this mediational hypothesis, I use a series of regressions
(Kenny et al. 1998; Baron & Kenny 1986). First, I examine the effect of team performance on
revenue without including teams’ average home attendance. As control variables, I include the
number of All-Stars on a team, whether the team made the playoffs in the previous year, the
average cost for a fan to attend a team’s game, the age of the team’s arena, the arena’s seating
capacity, the metropolitan population of the team’s city, brand value, and year fixed effects
(Ertug & Castellucci 2013; Berri et al. 2004; Hausman & Leonard 1997). The model is estimated
as follows:
(2) Revenue = β
0
+ β
1
Team Performance + Controls + Year Fixed Effects + ε.
Where Revenue is a team’s revenue for a season as estimated by Forbes, and all other variables
are as defined earlier and in the Appendix. I then examine if team performance has a positive
effect on a team’s average home attendance while controlling for other factors that could affect a
team’s average home attendance including the number of All-Stars on a team, whether the team
made the playoffs in the previous year, the average cost for a fan to attend a team’s game, the
age of the team’s arena, the arena’s seating capacity, the metropolitan population of the team’s
city, brand value, and year fixed effects.
(3) Home Attendance = β
0
+ β
1
Team Performance + Controls + Year Fixed Effects + ε.
Where Home Attendance is measured as the average home game attendance for team during the
regular season, and all other variables are defined as earlier and in the appendix. Finally, I
examine the effect of home attendance and team performance on revenue when both variables
are included in the regression.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
45
(4) Revenue = β
0
+ β
1
Team Performance + β
2
Home Attendance + Controls + Year Fixed
Effects + ε.
All variables are as defined earlier and in the Appendix.
My third and fourth hypotheses are that team performance and team-level player
compensation have positive effects on brand value. To test these hypotheses, I estimate the
following regression:
(5) Brand Value = β
0
+ β
1
Team Performance + β
2
Player Expenses + Controls + Year Fixed
Effects + ε.
Where a team’s average road game attendance, Road Attendance, is a proxy for brand value, and
all variables are defined as earlier and in the appendix. As control variables, I include the
number of All-Stars on a team, whether the team made the playoffs in the previous year, the
conference the team plays in, the number of NBA championships the team has won, the
metropolitan population of the team’s city, and year fixed effects (Berri & Schmidt 2006).
My fifth hypothesis is that a team’s brand value has a positive effect on revenue. To test
this hypothesis, I use the same revenue model as estimated in model 4 while focusing on the
significance of the proxy measure for brand value, Road Attendance.
Finally, my sixth hypothesis is that revenue is more informative to team values than
earnings while my seventh hypothesis is that brand value has a positive effect on team value.
These hypotheses are tested using variations of the following regression:
(6) Team Value = β
0
+ β
1
(Revenue or Operating Income) + β
2
Brand Value + Controls + Year
Fixed Effects + ε.
Where Team Value is the estimated market value of a team by Forbes, Operating Income is a
team’s operating income for a season as estimated by Forbes, and all variables are as defined
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
46
earlier or in the appendix. As control variables, I include the team’s operating expenses as
estimated by Forbes, number of All-Stars on a team, whether the team made the playoffs in the
previous year, the team’s regular season winning percentage, the age of the team’s arena, the
arena’s capacity, the metropolitan population of the team’s city, the team’s average home
attendance, and year fixed effects (Alexander & Kern 2004; Rosner & Shropshire 2004).
IV. RESULTS
Primary Test Results
Results of Hypotheses Tests
1
Results for Hypothesis 1 are presented on Table 6. Consistent with Hypothesis 1, I find
that Player Expenses has a statistically significant positive effect on team performance (p = 0.01,
one-tailed). This supports that teams that spend more on team-level player compensation win
more games. I note the significance of the control variables which indicate that teams with more
All-Stars, who made the playoffs in the prior year, who have older players, who are from smaller
metropolitan areas, and who play in the Western Conference have higher team performance
within the sample period.
Results for Hypothesis 2 are presented on Table 7, Panels A-C. In Panel A of Table 7, I
find that Winning Percentage has a statistically significant positive effect on Revenue (p < 0.01,
one-tailed). This means that teams who win more games during the regular season take in higher
amounts of revenue. To test the mediating effect of home attendance on this relationship, I first
find in Panel B of Table 7 that Winning Percentage is positively related to Home Attendance at a
statistically significant level (p < 0.01, one-tailed).
1
All of the models are estimated in double-logged form to allow for potential non-linear relationships while linear
form model estimates are also provided (Berri et al. 2004).
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
47
TABLE 6
Regression Estimating Team Performance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Winning Percentage Dependent Variable: Winning Percentage
Variable Coefficient p-value Coefficient p-value
Ln Player Expenses 0.197
**
0.01 Player Expenses 0.001
**
0.02
All-Stars 0.179
***
0.00 All-Stars 0.087
***
0.00
Playoffs
t-1
0.125
***
0.01 Playoffs
t-1
0.060
***
0.00
Average Player Age 0.035
***
0.00 Average Player Age 0.014
***
0.01
Ln Population -0.038
**
0.02 Population -0.003
***
0.01
Conference -0.084
***
0.00 Conference -0.043
***
0.00
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.505 Adjusted R
2
= 0.570
Note: Winning Percentage is the team’s regular season winning percentage at time t. Player Expenses (in millions) is the amount spent on players’ compensation
in a season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season
and 0 otherwise. Average Player Age is the average age of the players on a team. Population (in millions) is the team’s 2011 metropolitan area population.
Conference is an indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team plays in the Western Conference.
Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-
tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
48
Next, I include Home Attendance in the revenue regression as shown in Model 4 with results
presented in Panel C of Table 7. When Home Attendance is included in the revenue model, I find
that Home Attendance has a statistically significant positive effect on Revenue (p < 0.01, one-
tailed) while also no longer finding that the previous statistically significant effect of Winning
Percentage (p = 0.19, one-tailed), thus supporting the mediation effect in Hypothesis 2. I also
note from Panel C of Table 7 the statistically significant effect of control variables such that
teams earn higher revenue when it costs fans more to attend games, when they play in newer
arenas, and the teams’ have higher brand values. This last result relates to Hypothesis 5 and will
be discussed more below.
Results for Hypotheses 3 and 4 are presented on Table 8. In support of both Hypotheses 3
and 4, I find that Winning Percentage and Player Expenses have statistically significant positive
effects on the proxy for brand value, Road Attendance (p < 0.01 and p = 0.03, respectively, one-
tailed). This indicates that teams that win more games and have greater team-level player
compensation have higher average attendance for their road games, signaling a greater level of
brand value of the team. I note the significant positive effect on Road Attendance of having more
All-Stars, of teams being from bigger cities, and of teams who have won more championships.
The significant effects of the hypothesized and control variables on Road Attendance arguably
support its use as a proxy for brand value.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
49
TABLE 7
Panel A: Regression Estimating Revenue
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.089
***
0.00 Winning Percentage 26.337
***
0.00
All-Stars -0.148 0.30 All-Stars -2.120 0.28
Playoffs
t-1
0.019 0.41 Playoffs
t-1
0.528 0.85
Ln Fan Cost Index 0.550
***
0.00 Fan Cost Index 0.279
***
0.00
Ln Player Expenses 0.192 0.11 Player Expenses 0.375
**
0.08
Ln Arena Age with Renovation -0.045
**
0.02 Arena Age with Renovation -0.799
**
0.01
Ln Arena Capacity 0.459
**
0.034 Arena Capacity 0.002
*
0.06
Ln Population 0.025 0.55 Population 0.477 0.628
Ln Road Attendance 0.835
***
0.00 Road Attendance 0.004
**
0.01
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.802 Adjusted R
2
= 0.811
Note: Revenue (in millions) is the team’s total revenue for the season. Winning Percentage is the team’s regular season winning percentage at time t. All-Stars is
the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan
Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for
one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’
compensation in a season. Arena Age with Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena Capacity is
the number of attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean
attendance for the team’s road games. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%,
5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
50
Panel B: Regression Estimating Home Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Home Attendance Dependent Variable: Home Attendance
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.104
***
0.00 Winning Percentage 4238.549
***
0.00
All-Stars -0.005 0.59 All-Stars -173.384 0.26
Playoffs
t-1
0.037
***
0.01 Playoffs
t-1
604.435
***
0.01
Ln Fan Cost Index 0.142
**
0.02 Fan Cost Index 7.705
**
0.04
Ln Player Expenses 0.028 0.59 Player Expenses 11.599 0.40
Ln Arena Age with Renovation -0.007 0.56 Arena Age with Renovation -22.343 0.46
Ln Arena Capacity 0.714
***
0.00 Arena Capacity 0.555
***
0.00
Ln Population -0.012 0.60 Population -26.453 0.69
Ln Road Attendance 0.392
*
0.08 Road Attendance 0.371
*
0.10
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.555 Adjusted R
2
= 0.562
Note: Home Attendance is the mean attendance for the team’s home games. Winning Percentage is the team’s regular season winning percentage at time t. All-
Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise.
The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs,
parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on
players’ compensation in a season. Arena Age with Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena
Capacity is the number of attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is
the mean attendance for the team’s road games. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at
the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
51
Panel C: Mediation Test
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.027 0.19 Winning Percentage 10.372 0.11
Ln Home Attendance 0.591
***
0.00 Home Attendance 0.004
***
0.00
All-Stars -0.012 0.33 All-Stars -1.541 0.35
Playoffs
t-1
-0.003 0.88 Playoffs
t-1
-1.75 0.54
Ln Fan Cost Index 0.466
***
0.00 Fan Cost Index 0.250
***
0.00
Ln Player Expenses 0.175 0.12 Player Expenses 0.331
*
0.10
Ln Arena Age with Renovation -0.041
**
0.01 Arena Age with Renovation -0.714
**
0.01
Ln Arena Capacity 0.037 0.86 Arena Capacity 0.000 0.83
Ln Population 0.032 0.39 Population 0.576 0.51
Ln Road Attendance 0.603
**
0.03 Road Attendance 0.002 0.12
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.838 Adjusted R
2
= 0.837
Note: Revenue (in millions) is the team’s total revenue for the season. Winning Percentage is the team’s regular season winning percentage at time t. All-Stars is
the number of All-Stars on the team in a season. Home Attendance is the mean attendance for the team’s home games. Playoffs
t-1
is an indicator variable of 1 if
the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft
beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable
caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years since the arena
was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions) is the team’s
2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols *, **, and ***
denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
52
TABLE 8
Regression Estimating Road Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Road Attendance Dependent Variable: Road Attendance
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.023
***
0.00 Winning Percentage 1103.988
***
0.00
Ln Player Expenses 0.023
**
0.03 Player Expenses 5.201
**
0.04
All-Stars 0.017
***
0.00 All-Stars 282.804
***
0.00
Playoffs
t-1
0.006 0.12 Playoffs
t-1
90.909 0.23
Ln Population 0.007
*
0.08 Population 21.093 0.11
Conference 0.001 0.88 Conference 30.273 0.80
NBA Championships 0.003
*
0.06 NBA Championships 50.735
**
0.05
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.566 Adjusted R
2
= 0.568
Note: Road Attendance is the mean attendance for the team’s road games. Winning Percentage is the team’s regular season winning percentage at time t. Player
Expenses (in millions) is the amount spent on players’ compensation in a season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an
indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. Population (in millions) is the team’s 2011 metropolitan area population.
Conference is an indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team plays in the Western Conference.
NBA Championships is the number of NBA Championships won by a franchise. Year indicators are included. The symbols *, **, and *** denote coefficients
statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
53
Results for Hypothesis 5 are presented on Table 7, Panel C. As mentioned earlier, I find
that the proxy for brand value, Road Attendance, has a statistically significant positive effect on
Revenue (p = 0.03, one-tailed). This provides support for Hypothesis 5 that teams with higher
brand values are able to realize greater revenues. This effect is incremental to the statistically
significant effect of Home Attendance on Revenue, which likely speaks to the variety of revenue
streams that comprise the Revenue variable. I am unable to disaggregate revenue into its different
revenue streams as Forbes only provides estimates of teams’ total revenues.
1
Nevertheless,
results do suggest that having higher brand values have an incremental effect on revenue above
and beyond the major revenue source of home games. For example, higher brand values are
likely to improve a team’s negotiating position for local television broadcast rights and command
higher prices for advertising deals for sponsorships and promotions.
Results for Hypotheses 6 and 7 are presented on Table 9. I find that Revenue has a
statistically significant positive effect on Team Value (p < 0.01, one-tailed) and that, consistent
with Hypothesis 7, brand value has statistically significant positive effect on Team Value (p <
0.01, one-tailed). These results support that teams with higher revenue have higher team values
along with an incremental positive effect of brand value on team value. Related to Hypothesis 6,
I note that team value model presented on Table 9 includes Revenue and Operating Expenses as
separate variables rather than an aggregated number in Operating Income. Theoretical support
for this approach comes from the idea that NBA teams are valued similarly to technology firms
where revenue can be a more informative valuation factor compared to aggregated earnings
(Chandra & Ro 2008; Amir & Lev 1996). In untabulated results, I find empirical support for this
1
From the select few audited financial statements of NBA teams that have been leaked into the public domain, I
observe that revenue can be broken down into different categories such as home game ticket revenue, broadcast and
cable rights revenue, arena related revenues (concessions and parking), advertising revenue, and NBA-related
investments revenue. Anecdotally, home game ticket revenue and broadcast and rights revenue are proportionately
by far the two biggest revenue sources of revenue followed by arena related revenue and advertising revenue.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
54
idea as the adjusted R
2
when including the aggregated variable Operating Income is 0.795
compared to an adjusted R
2
of 0.916
1
when Revenue and Operating Expenses are included as
separate factors. Consistent with Hypothesis 6, the greater adjusted R
2
indicates that the
disaggregated model with Revenue and Operating Expenses has greater explanatory power of
team values than the model with Operating Income (Vuong Z-Statistic = 7.406, p < 0.01)
(Dechow 1994; Vuong 1989). All other results and inferences remain unchanged whether
aggregated earnings or separate factors are used in the team value model.
Timing of Owner Objectives
From an interdependent view of owner objectives, the objectives of maximizing wins and
maximizing profits might not be realized contemporaneously (Keating 2014; Silver 2014;
Zimbalist 2003; Cyert & March 1963). The idea behind these timing differences is that spending
more on player compensation will have a short-term, contemporaneous effect that leads to higher
winning percentage and lower operating income while also having a long-term, value building
effect on overall financial returns. To further support the idea of timing differences in achieving
owner objectives, Table 10 provides correlations of measures of owner objectives of maximizing
wins and maximizing profits with contemporaneous and lagged player expenses.
1
This adjusted R
2
is from the linear specification of the team value model. This is given as a comparison rather than
the double-logged specification because operating income has some negative values which require adjustments to
create positive numbers in order to use the log specification. As such, I provide the comparison of the Adjusted R
2
values based on the linear specification.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
55
TABLE 9
Regression Estimating Team Value
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Team Value Dependent Variable: Team Value
Variable Coefficient p-value Coefficient p-value
Ln Revenue 0.926
***
0.00 Revenue 3.625
***
0.00
Ln Road Attendance 0.570
***
0.00 Road Attendance 0.007
**
0.03
Ln Operating Expenses 0.002 0.97 Operating Expenses -0.217 0.41
All-Stars 0.007 0.38 All-Stars 7.861
*
0.10
Playoffs
t-1
-0.007 0.49 Playoffs
t-1
0.984 0.88
Ln Winning Percentage -0.013 0.53 Winning Percentage -18.039 0.58
Ln Arena Age with Renovation -0.003 0.74 Arena Age with Renovation -0.288 0.77
Ln Arena Capacity 0.013 0.91 Arena Capacity 0.001 0.69
Ln Population 0.030
**
0.01 Population 1.634
*
0.09
Ln Home Attendance 0.019 0.84 Home Attendance -0.003 0.36
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.949 Adjusted R
2
= 0.916
Note: Team Value (in millions) is the team’s estimated team value for the year. Revenue (in millions) is the team’s total revenue for the season. Road Attendance
is the mean attendance for the team’s road games. Operating Expenses (in millions) is the team’s total operating expenses which excludes interest, taxes,
depreciation, and amortization. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs
in the prior season and 0 otherwise. Winning Percentage is the team’s regular season winning percentage at time t. Arena Age with Renovation is the numbers of
years since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in
millions) is the team’s 2011 metropolitan area population. Home Attendance is the mean attendance for the team’s home games. Year indicators are included. The
symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed
otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
56
TABLE 10
Timing of Owner Objectives
Panel A: Correlations of Measures of Owner Objectives with Current and Lagged Player Expenses
Variable
Player
Expenses
PE,
Lag 1
PE,
Lag 2
PE,
Lag 3
PE,
Lag 4
PE,
Lag 5
PE,
Lag 6
PE,
Lag 7
PE,
Lag 8
PE,
Lag 9
PE,
Lag 10
PE,
Lag 11
Winning Percentage
0.24
**
0.13
*
0.03 -0.02 -0.06 -0.01 0.05 0.06 0.03 -0.03 -0.02 -0.03
Operating Income
-0.16
**
-0.12
*
-0.03 0.08 0.16
**
0.16
*
0.19
**
0.23
**
0.20
*
0.22
*
0.23
**
0.24
Value Return
-0.12
*
-0.17
**
0.00 0.12
*
0.26
**
0.31
**
0.33
**
0.31
**
0.25
**
0.29
**
0.10 0.07
Overall Return
-0.24
**
-0.25
**
-0.07 0.08 0.24
**
0.27
**
0.30
**
0.30
**
0.23
**
0.28
**
0.11 0.09
N
386
356
326
296
266
236
206
176
146
116
87
58
Note: Note: this table presents Pearson correlations. The symbols * and ** denote statistical significance (two-tailed) at the 5%, and 1% levels, respectively.
Value Return is the percent return in the team’s value from the prior year. Overall Return is the percent return based on the change in team value and operating
income. Operating Income is the team’s total operating profits, excluding interest, taxes, depreciation, and amortization. Winning Percentage is the team’s
regular season winning percentage at time t. Player Expenses is the amount spent on players’ compensation in a season.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
57
Panel B: Correlations of Measures of Owner Objectives with Player Expenses on a 3 year rolling window
Variable
Player
Expenses,
0-2
PE,
Lag 1-3
PE,
Lag 2-4
PE,
Lag 3-5
PE,
Lag 4-6
PE,
Lag 5-7
PE,
Lag 6-8
PE,
Lag 7-9
PE,
Lag 8-10
PE,
Lag 9-11
Winning Percentage
0.13
**
0.03 -0.05 -0.04 0.01 0.06 0.06 -0.01 -0.05 -0.11
Operating Income
-0.13
**
-0.01 0.12
*
0.17
*
0.22
**
0.29
**
0.30
**
0.32
**
0.28
**
0.28
*
Value Return
-0.11
*
-0.03 0.17
**
0.29
**
0.35
**
0.36
**
0.32
**
0.28
**
0.17 0.03
Overall Return
-0.20
**
-0.08 0.15
*
0.26
**
0.33
**
0.36
**
0.32
**
0.29
**
0.18 0.05
N
326
296
266
236
206
176
146
116
87
58
Note: Note: this table presents Pearson correlations. The symbols * and ** denote statistical significance (two-tailed) at the 5%, and 1% levels, respectively.
Value Return is the percent return in the team’s value from the prior year. Overall Return is the percent return based on the change in team value and operating
income. Operating Income is the team’s total operating profits, excluding interest, taxes, depreciation, and amortization. Winning Percentage is the team’s
regular season winning percentage at time t. Player Expenses are the amount spent on players’ compensation on 3 year rolling windows.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
58
On the first row of Panel A of Table 10, Winning Percentage is seen to have a significant
positive relationship with Player Expenses during the contemporaneous year (r = 0.24) and a
weaker positive relationship with the prior year player expenses (r = 0.12). Thereafter, there is
not a statistically significant or clear relationship between Winning Percentage and Player
Expenses from two or more years prior. This seems to support a strong connection between the
contemporaneous effect of the amount spent on player compensation and how a team performs
during a season.
Rows 2 – 4 of Panel A of Table 10 present various measures of the profit objective
including Operating Income, Value Return, and Overall Return. Each of these measures is shown
to have a significant negative relationship with Player Expenses during the contemporaneous
year and prior year. Hence, spending more on player compensation leads to higher winning
percentage, yet lower profits and returns. This is consistent with the view of owner objectives
being conflicting objectives. However, beginning with a 3-year lag of Player Expenses and
continuing on to a 10-year lag of Player Expenses, the various measures of the profit objective
demonstrate a statistically significant positive relationship with these lagged measures of Player
Expenses. These relationships in concert with the short-term positive relationship of Winning
Percentage and Player Expenses seem to support the interdependent view of owner objectives
such that the effects of spending more on player expenses leads first to greater winning and then
to greater profits. Panel B of Table 10 presents similar results using a 3-year rolling window of
Player Expenses.
These results show that team owners must trade off the future economic benefits from
winning games against the present negative profit-impact of expending resources to win games.
Results also suggest that building a team’s revenue generating base and brand value takes time to
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
59
develop and grow and is dependent upon the greater investments in team-level player
compensation. Thus, both objectives of maximizing wins and maximizing profits can be
achieved, yet may not be realized contemporaneously.
Additional Analysis
Determinants of Team-Level Player Compensation
The above results suggest that team owners who pursue operational profits by spending
less on team-level player compensation are likely to do so at the expense of long-term gains.
1
Absent many of the typical causes of myopia, I explore potential influencing factors on team
owners’ investment behavior by modeling determinants of team-level player compensation. To
do so, I use team-level player compensation as a dependent variable. As independent variables
that could affect the amount spent on team-level player compensation, I include measures of the
resources available to team owners, (Revenue
i,t-1
), the owners’ personal net worth (Owner Net
Worth
i
), the team composition including the average age of players on the team (Average Player
Age
i,t
) and the number of high-profile, top performing players (All-Stars
i,t
), the owner’s age, the
metropolitan population of the city where the team is from, and year fixed effects. By
convention, older players generally command higher salaries due to the tenure-related escalation
of allowed salaries under the NBA’s Collective Bargaining Agreement (Coon 2012). One-year
lagged revenue is included because teams that collect more revenue are likely to have greater
1
While the above results show the on-average effect of owners’ decisions about team-level player compensation, an
anecdote from the Miami Heat’s decision to amnesty Mike Miller after the 2012-13 season seems fitting. After
winning the NBA championship in 2013, wherein Miller played a significant role, the Miami Heat decided to
exercise an amnesty clause that cut Miller from the team and removed his committed salary from the team’s salary
cap number. The Heat still were obligated to pay Miller’s contract in full, an amount equal to $12.8 million, but that
number would not count against their salary cap. This move lowered the Heat’s luxury tax bill for that season by
nearly $17 million. It seems clear from the sporting news that the team’s General Manager, Pat Riley, and Miller’s
teammates wanted Miller to remain on the team to help defend their title (Windhorst 2013). The decision though
was ultimately the team owner’s decision to make and the owner chose to not pay. In the following season, the
Miami Heat made it to the Finals again, but lost to the San Antonio Spurs leaving many to comment on the “what
ifs” for Miami and keeping Miller. Arguably, the value of the Miami Heat would have increased from winning
another championship.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
60
resources to pay player compensation in future seasons. Owner Net Worth is obtained from
usatoday.com
1
and is measured as of 2005.
2
Owner Net Worth is included as a static variable
across sample years, making it a measure of the wealth of an owner relative to other owners.
3
In
the main tests above, team-level player compensation was measured as the total compensation
for player contracts (Player Expenses).
4
This measure does not take into account the amount of
luxury tax that owners might have incurred. As an alternative measure of team-level player
compensation, I define Total Cost to Owner as the sum of the total compensation paid to players
(Players Expenses) and the amount of luxury tax paid.
In results reported in Panels A and B of Table 11, I find that the resources available to
team owners, Revenue
t-1
(p = 0.01; p < 0.01), owner’s personal wealth, Owner Net Worth
i
(p =
0.01; p = 0.01), and Average Player Age (p < 0.01; p < 0.01) have statistically significant
positive effect on with team-level player compensation measured either as Player Expenses or
Total Cost to Owner. These results provide evidence of owners’ personal influence on the
amount of money spent on team-level player compensation. Specifically, owners’ personal
wealth has a significant positive effect on how much an owner spends on team-level player
compensation, even after controlling for the team’s composition and financial resources available
to the team. These results suggest that owners’ personal wealth is positively associated with
long-term thinking or willingness to spend money, which based on prior results, ultimately leads
to greater overall financial returns.
1
url: http://usatoday30.usatoday.com/sports/basketball/nba/2005-salary-owners.htm
2
This data requirement causes a drop in sample size from 386 firm-years to 311 firms-years.
3
I also obtain owners’ net worth as of 2013 from Forbes.com (url:
http://www.forbes.com/sites/tomvanriper/2013/01/23/the-nbas-billionaire-owners/). For those owners whose net
worth are available as of 2005 and 2013, the correlation between the two values is 0.94, suggesting that using 2005
net worth as a static, relative wealth measure is reasonable.
4
Prior results are robust to using Total Cost to Owner rather than Player Expenses.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
61
TABLE 11
Panel A: Regression Estimating Team Level Player Compensation
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Player Expenses Dependent Variable: Player Expenses
Variable Coefficient p-value Coefficient p-value
Ln Owner Net Worth 0.034
**
0.01 Owner Net Worth 0.001
***
0.00
Ln Revenue
t-1
0.276
***
0.01 Revenue
t-1
0.150
***
0.00
All-Stars -0.014 0.46 All-Stars -0.868 0.46
Average Player Age 0.034
***
0.00 Average Player Age 2.114
***
0.00
Owner Birth Year 0.003 0.14 Owner Birth Year 0.115 0.35
Ln Population -0.003 0.89 Population 0.244 0.43
Year Fixed Effects Included Year Fixed Effects Included
n = 311 team-years n = 311 team-years
Adjusted R
2
= 0.596 Adjusted R
2
= 0.583
Note: Player Expenses (in millions) is the amount spent on players’ compensation in a season. Owner Net Worth (in millions) is the owner’s net worth as of 2005.
Revenue
t-1
(in millions) is the one-year lag of the team’s total revenue for the season. All-Stars is the number of All-Stars on the team in a season. Average Player
Age is the average age of the players on a team. Owner Birth Year is the year the owner was born. Population (in millions) is the team’s 2011 metropolitan area
population. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels,
respectively.
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
62
Panel B: Regression Estimating Team Level Player Compensation
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Total Cost to Owner Dependent Variable: Total Cost to Owner
Variable Coefficient p-value Coefficient p-value
Ln Owner Net Worth 0.044
***
0.01 Owner Net Worth 0.001
***
0.00
Ln Revenue
t-1
0.356
***
0.00 Revenue
t-1
0.221
***
0.00
All-Stars -0.018 0.47 All-Stars -1.521 0.41
Average Player Age 0.042
***
0.00 Average Player Age 2.962
***
0.00
Owner Birth Year 0.004
*
0.09 Owner Birth Year 0.195 0.27
Ln Population 0.002 0.91 Population 0.422 0.33
Year Fixed Effects Included Year Fixed Effects Included
n = 311 team-years n = 311 team-years
Adjusted R
2
= 0.566 Adjusted R
2
= 0.513
Note: Total Cost to Owner (in millions) is the sum of the amount spent on players’ compensation and luxury tax payments for the season. Owner Net Worth (in
millions) is the owner’s net worth as of 2005. Revenue
t-1
(in millions) is the one-year lag of the team’s total revenue for the season. All-Stars is the number of All-
Stars on the team in a season. Average Player Age is the average age of the players on a team. Owner Birth Year is the year the owner was born. Population (in
millions) is the team’s 2011 metropolitan area population. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from
zero at the 10%, 5%, and 1% levels, respectively.
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
63
Robustness Tests
Alternative Measures of Team Performance
In the main test of Hypothesis 1 that examined the effect of team-level player
compensation on team performance, team performance was measured as the team’s regular
season Winning Percentage (wins/games played) (Bloom 1999). As a robustness test, I repeat the
analysis of Model 1 with alternative measures of team performance. In the first alternative
specification, I define Playoffs to be a dummy variable coded for 1 if the team made the playoffs
in a given season and 0 otherwise. I run a logistic regression similar to Model 1, but with
Playoffs as the dependent variable. Consistent with the primary test results, I find as reported on
Panel A of Table 12 a statistically significant positive effect of Player Expenses on Playoffs (p =
0.06, one-tailed), although results are slightly weaker compared to using Winning Percentage.
This makes sense because although Playoffs is highly positively correlated with Winning
Percentage (r = 0.81), teams can have a relatively high winning percentage and still not make the
playoffs. This is especially true over the sample period as evidenced by the significant effect of
Conference on Winning Percentage as seen in Table 6. Over the sample period, the Western
Conference teams had higher winning percentages than teams in the Eastern Conference, which
means that teams in the Western Conference missed the playoffs whereas they would have been
in the playoffs based on their winning percentage had they been in the Eastern Conference.
In the second alternative specification, I define Season Performance to be an ordered
variable which takes a value of 1 if the team did not qualify for the Playoffs, 2 if the team
qualified for the Playoffs, 3 if the team advanced to the Conference Semifinals, 4 if the team
advanced to the Conference Finals, 5 if the team advanced to the NBA Finals, and 6 if the team
won the NBA Championship (Ertug & Castellucci 2013). Using an ordered logit regression with
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
64
Season Performance as the dependent variable, I find as reported on Panel B of Table 12 that the
coefficient on Player Expenses is positive but not statistically significant (p = 0.25, one-tailed).
These results in concert with the other specifications of team performance seem to suggest that
paying more for players’ compensation improves a team’s ability to win more regular season
games and make the playoffs, but does not necessarily improve a team’s chances of advancing
through the playoffs. This is likely due to the important complementarities in basketball, like
teamwork or coaching prowess, that are not considered by player compensation measures. These
factors are especially likely to gain importance in high stakes competition of a playoff
atmosphere. Thus paying more for player compensation can help facilitate a team making the
playoffs but basketball-related factors may carry more weight therein.
In the main tests of Hypothesis 2 and Hypothesis 3, team performance was also measured
using Winning Percentage. I repeat the analysis related to these hypotheses using the alternative
measures of team performance, Playoffs and Season Performance. Consistent with the primary
test results for Model 2 presented in Panel A of Table 7, I find in separate regressions reported in
Panels A and B of Table 13 that Playoffs and Season Performance have statistically significant
positive effects on Revenue (p < 0.01, one-tailed for both variables). I also find in separate
regressions of Model 3 presented in Panel B of Table 7 that results are consistent with a
significant positive effect of Playoffs (p < 0.01, one-tailed) and Season Performance (p = 0.04,
one-tailed) on Home Attendance, as reported in Panels C and D of Table 13.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
65
TABLE 12
Panel A: Robustness Test of Regression Estimating Team Performance
1
Double-Logged Specification, Logit Model Linear Specification, Logit Model
Dependent Variable: Playoffs Dependent Variable: Playoffs
Variable Coefficient p-value Coefficient p-value
Ln Player Expenses 1.576
*
0.06 Player Expenses 0.022
*
0.07
All-Stars 1.618
***
0.00 All-Stars 1.586
***
0.00
Playoffs
t-1
1.049
***
0.01 Playoffs
t-1
1.086
***
0.01
Average Player Age 0.215
**
0.05 Average Player Age 0.217
**
0.05
Ln Population -0.353
**
0.04 Population -0.042
*
0.06
Conference 0.200 0.48 Conference 0.155
***
0.59
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Pseudo R
2
= 0.356 Pseudo R
2
= 0.353
Note: Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at time t. Player Expenses (in millions) is the amount spent on players’
compensation in a season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the
prior season and 0 otherwise. Average Player Age is the average age of the players on a team. Population (in millions) is the team’s 2011 metropolitan area
population. Conference is an indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team plays in the Western
Conference. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels,
respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
66
Panel B: Robustness Test of Regression Estimating Team Performance
1
Double-Logged Specification, Ordered Logit Model Linear Specification, Ordered Logit Model
Dependent Variable: Season Performance Dependent Variable: Season Performance
Variable Coefficient p-value Coefficient p-value
Ln Player Expenses 0.590 0.25 Player Expenses 0.006 0.31
All-Stars 1.404
***
0.00 All-Stars 1.381
***
0.00
Playoffs
t-1
1.090
***
0.00 Playoffs
t-1
1.129
***
0.00
Average Player Age 0.106 0.20 Average Player Age 0.107 0.18
Ln Population -0.209 0.29 Population -0.020 0.52
Conference 0.042 0.88 Conference 0.006 0.98
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Pseudo R
2
= 0.200 Pseudo R
2
= 0.199
Note: Season Performance is an ordered variable which takes a value of 1 if the team did not qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if
the team advanced to the Conference Semifinals, 4 if the team advanced to the Conference Finals, 5 if the team advanced to the NBA Finals, and 6 if the team
won the NBA Championship. Player Expenses (in millions) is the amount spent on players’ compensation in a season. All-Stars is the number of All-Stars on the
team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. Average Player Age is the average age
of the players on a team. Population (in millions) is the team’s 2011 metropolitan area population. Conference is an indicator variable set equal to one if the team
plays in the Eastern Conference, and equal to zero if the team plays in the Western Conference. Year indicators are included. The symbols *, **, and *** denote
coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
67
Finally, I find in alternative specifications of Model 4 presented in Panel C of Table 7,
that Playoffs (p = 0.03, one-tailed) and Season Performance (p < 0.01, one-tailed) have
significant direct effects on revenue alongside the significant positive effect of Home Attendance
on revenue (p < 0.01 in both cases), as seen in Panels E and F of Table 13. The significant direct
effects rather than a fully mediated effect when using these two alternative specifications of team
performance likely results from the alternative measures of team performance picking up the
effects of playing games after the regular season, which increases revenue. Overall, results
related to Hypothesis 2 and the various measures of team performance suggest that a team’s
current season winning percentage increases revenue as mediated through home attendance
while playing more games in the playoffs has an incremental effect on revenue.
The two alternative measures of team performance could also be included as
complementary variables to capture the effect of playing playoff games on revenue rather than
being used as alternative measures of team performance. When either of these measures is
included in Model 2, then Winning Percentage could be interpreted as being the effect on
revenue of the team’s performance in regular season games while Playoffs or Season
Performance could be interpreted as the incremental effect on revenue of the team’s post-season
performance. In those specifications, I find in results reported on Panel A of Table 14 that
Winning Percentage continues to have a significant positive effect on revenue (p = 0.04, one-
tailed) when Playoffs (p = 0.06, one-tailed) is included, and as reported on Panel B of Table 14
that Winning Percentage has a significant positive effect on revenue (p = 0.07, one-tailed) when
Season Performance (p = 0.01, one-tailed) is included. These results highlight that while the
three measures of team performance are significantly related, they can be informative in
explaining revenue as separate variables.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
68
TABLE 13
Panel A: Robustness Test of Regression Estimating Revenue
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Playoffs 0.500
***
0.00 Playoffs 5.703
***
0.00
All-Stars -0.012 0.41 All-Stars -1.446 0.46
Playoffs
t-1
0.020 0.36 Playoffs
t-1
0.878 0.75
Ln Fan Cost Index 0.542
***
0.00 Fan Cost Index 0.276
***
0.00
Ln Player Expenses 0.204
*
0.08 Player Expenses 0.400
**
0.06
Ln Arena Age with Renovation -0.046
**
0.02 Arena Age with Renovation -0.807
**
0.01
Ln Arena Capacity 0.449
**
0.04 Arena Capacity 0.002
*
0.06
Ln Population 0.023 0.57 Population 0.421 0.67
Ln Road Attendance 0.927
***
0.00 Road Attendance 0.004
***
0.00
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.801 Adjusted R
2
= 0.809
Note: Revenue (in millions) is the team’s total revenue for the season. Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at time
t. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0
otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size
hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount
spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years since the arena was either built or underwent major renovation.
Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road
Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different
from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
69
Panel B: Robustness Test of Regression Estimating Revenue
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Season Performance 0.024
***
0.00 Season Performance 2.836
***
0.00
All-Stars -0.016 0.26 All-Stars -1.939 0.32
Playoffs
t-1
0.022 0.33 Playoffs
t-1
1.068 0.70
Ln Fan Cost Index 0.551
***
0.00 Fan Cost Index 0.279
***
0.00
Ln Player Expenses 0.214
*
0.06 Player Expenses 0.419
**
0.04
Ln Arena Age with Renovation -0.046
**
0.02 Arena Age with Renovation -0.810
**
0.01
Ln Arena Capacity 0.433
**
0.04 Arena Capacity 0.002
*
0.06
Ln Population 0.022 0.60 Population 0.391 0.69
Ln Road Attendance 0.804
***
0.00 Road Attendance 0.003
***
0.00
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.803 Adjusted R
2
= 0.812
Note: Revenue (in millions) is the team’s total revenue for the season. Season Performance is an ordered variable which takes a value of 1 if the team did not
qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the team advanced to the Conference Semifinals, 4 if the team advanced to the Conference
Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won the NBA Championship. Player Expenses (in millions) is the amount spent on players’
compensation in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™
comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car,
two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a
season. Arena Age with Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena Capacity is the number of
attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the
team’s road games. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels,
respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
70
Panel C: Robustness Test of Regression Estimating Home Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Home Attendance Dependent Variable: Home Attendance
Variable Coefficient p-value Coefficient p-value
Playoffs 0.037
***
0.00 Playoffs 641.478
***
0.00
All-Stars 0.004 0.70 All-Stars 12.862 0.94
Playoffs
t-1
0.044
***
0.00 Playoffs
t-1
729.554
***
0.00
Ln Fan Cost Index 0.135
**
0.04 Fan Cost Index 7.287
*
0.05
Ln Player Expenses 0.047 0.39 Player Expenses 16.527 0.25
Ln Arena Age with Renovation -0.007 0.59 Arena Age with Renovation -21.919 0.49
Ln Arena Capacity 0.713
***
0.00 Arena Capacity 0.560
***
0.00
Ln Population -0.015 0.52 Population -38.557 0.57
Ln Road Attendance 0.517
**
0.04 Road Attendance 0.505
*
0.03
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.530 Adjusted R
2
= 0.538
Note: Home Attendance is the mean attendance for the team’s home games. Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at
time t. All-Stars is the number of All-Stars on the team in a season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator
variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two
(2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-
size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years
since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions)
is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols
*, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
71
Panel D: Robustness Test of Regression Estimating Home Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Home Attendance Dependent Variable: Home Attendance
Variable Coefficient p-value Coefficient p-value
Season Performance 0.010
**
0.04 Season Performance 189.980
**
0.03
All-Stars 0.006 0.53 All-Stars 41.853 0.80
Playoffs
t-1
0.048
***
0.00 Playoffs
t-1
806.790
***
0.00
Ln Fan Cost Index 0.141
**
0.03 Fan Cost Index 7.595
**
0.04
Ln Player Expenses 0.056 0.30 Player Expenses 18.718 0.18
Ln Arena Age with Renovation -0.007 0.62 Arena Age with Renovation -20.444 0.52
Ln Arena Capacity 0.714
***
0.00 Arena Capacity 0.559
***
0.00
Ln Population -0.017 0.47 Population -43.620 0.52
Ln Road Attendance 0.479
*
0.06 Road Attendance 0.457
*
0.05
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.523 Adjusted R
2
= 0.533
Note: Home Attendance is the mean attendance for the team’s home games. Season Performance is an ordered variable which takes a value of 1 if the team did
not qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the team advanced to the Conference Semifinals, 4 if the team advanced to the
Conference Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won the NBA Championship. All-Stars is the number of All-Stars on the team in
a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of
four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs
and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with
Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a
team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year
indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if
predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
72
Panel E: Robustness Test of Mediation Test
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Playoffs 0.028
**
0.03 Playoffs 3.252
**
0.02
Ln Home Attendance 0.593
***
0.00 Home Attendance 0.004
***
0.00
All-Stars -0.014 0.25 All-Stars -1.495 0.34
Playoffs
t-1
-0.006 0.78 Playoffs
t-1
-1.910 0.50
Ln Fan Cost Index 0.462
***
0.00 Fan Cost Index 0.248
***
0.00
Ln Player Expenses 0.176 0.11 Player Expenses 0.336
*
0.09
Ln Arena Age with Renovation -0.042
**
0.01 Arena Age with Renovation -0.723
**
0.01
Ln Arena Capacity 0.026 0.90 Arena Capacity 0.000 0.86
Ln Population 0.032 0.38 Population 0.569 0.51
Ln Road Attendance 0.621
**
0.04 Road Attendance 0.002 0.19
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.838 Adjusted R
2
= 0.837
Note: Revenue (in millions) is the team’s total revenue for the season. Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at time
t. All-Stars is the number of All-Stars on the team in a season. Home Attendance is the mean attendance for the team’s home games. Playoffs
t-1
is an indicator
variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two
(2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-
size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years
since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions)
is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols
*, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
73
Panel F: Robustness Test of Mediation Test
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Season Performance 0.018
***
0.00 Season Performance 2.112
***
0.00
Ln Home Attendance 0.590
***
0.00 Home Attendance 0.004
***
0.00
All-Stars -0.020 0.11 All-Stars -2.097 0.19
Playoffs
t-1
-0.007 0.75 Playoffs
t-1
-1.990 0.48
Ln Fan Cost Index 0.468
***
0.00 Fan Cost Index 0.250
***
0.00
Ln Player Expenses 0.182 0.09 Player Expenses 0.348
*
0.08
Ln Arena Age with Renovation -0.042
***
0.01 Arena Age with Renovation -0.732
***
0.01
Ln Arena Capacity 0.012 0.95 Arena Capacity 0.000 0.91
Ln Population 0.032 0.38 Population 0.5556 0.53
Ln Road Attendance 0.522
*
0.06 Road Attendance 0.002 0.31
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.841 Adjusted R
2
= 0.839
Note: Revenue (in millions) is the team’s total revenue for the season. Season Performance is an ordered variable which takes a value of 1 if the team did not
qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the team advanced to the Conference Semifinals, 4 if the team advanced to the Conference
Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won the NBA Championship. Home Attendance is the mean attendance for the team’s home
games. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of
four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs
and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with
Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a
team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year
indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if
predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
74
TABLE 14
Panel A: Robustness Test of Regression Estimating Revenue
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.065
**
0.04 Winning Percentage 20.956
**
0.03
Playoffs 0.025
*
0.07 Playoffs 2.120 0.12
All-Stars -0.017 0.24 All-Stars -2.271 0.26
Playoffs
t-1
0.017 0.46 Playoffs
t-1
0.362 0.90
Ln Fan Cost Index 0.547
***
0.00 Fan Cost Index 0.278
***
0.00
Ln Player Expenses 0.192 0.11 Player Expenses 0.377
**
0.08
Ln Arena Age with Renovation -0.046
**
0.02 Arena Age with Renovation -0.806
**
0.01
Ln Arena Capacity 0.450
**
0.04 Arena Capacity 0.002
*
0.06
Ln Population 0.025 0.54 Population 0.476 0.63
Ln Road Attendance 0.850
***
0.00 Road Attendance 0.004
**
0.01
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.803 Adjusted R
2
= 0.811
Note: Revenue (in millions) is the team’s total revenue for the season. Winning Percentage is the team’s regular season winning percentage at time t. Playoffs is
an indicator variable of 1 if the team made the playoffs and 0 otherwise at time t. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an
indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price
tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-
expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the
numbers of years since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena.
Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are
included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted,
two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
75
Panel B: Robustness Test of Regression Estimating Revenue
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Revenue Dependent Variable: Revenue
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.057
*
0.07 Winning Percentage 15.764 0.11
Season Performance 0.018
**
0.01 Season Performance 1.939
*
0.05
All-Stars -0.021 0.13 All-Stars -2.617 0.20
Playoffs
t-1
0.016 0.46 Playoffs
t-1
0.398 0.89
Ln Fan Cost Index 0.551
***
0.00 Fan Cost Index 0.279
***
0.00
Ln Player Expenses 0.200 0.09 Player Expenses 0.392
*
0.07
Ln Arena Age with Renovation -0.046
**
0.02 Arena Age with Renovation -0.814
**
0.01
Ln Arena Capacity 0.434
**
0.04 Arena Capacity 0.002
*
0.07
Ln Population 0.025 0.55 Population 0.452 0.65
Ln Road Attendance 0.762
***
0.00 Road Attendance 0.003
***
0.01
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.805 Adjusted R
2
= 0.811
Note: Revenue (in millions) is the team’s total revenue for the season. Winning Percentage is the team’s regular season winning percentage at time t. Season
Performance is an ordered variable which takes a value of 1 if the team did not qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the team
advanced to the Conference Semifinals, 4 if the team advanced to the Conference Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won the
NBA Championship. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the
prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks,
four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in
millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years since the arena was either built or
underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan
area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols *, **, and *** denote coefficients
statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
76
The alternative measures of team performance could also be included alongside Winning
Percentage in Model 3 with Home Attendance as the dependent variable. If so, then Winning
Percentage could be interpreted as being the effect of the team’s performance in regular season
games on home attendance while Playoffs or Season Performance could be the incremental
effect of the team’s post-season participation or performance on home attendance. In separate
specifications of Model 3 as reported on Panels A and B of Table 15, Winning Percentage has a
significant positive effect (p < 0.01, one-tailed) when Playoffs (p = 0.41, one-tailed) is included
and a significant positive effect (p < 0.01, one-tailed) when Season Performance (p = 0.40, one-
tailed) is included. These results suggest that current regular season winning percentage is the
most informative team performance variable in explaining a team’s average home attendance for
regular season games. I also note that in all specifications with Home Attendance as the
dependent variable, whether a team made the playoffs in the prior season is a statistically
significant control variable. I interpret these results to signal the different types of ticket buyers,
where making the playoffs likely drives up the next season’s season-ticket holders whereas
current season winning percentage likely drives single game ticket sales. Overall, results from
using various alternative measures of a team’s performance in analyses related to Hypothesis 2
suggest that current season winning percentage is the key team performance variable that drives
home attendance which subsequently leads to more revenue. Additionally, team performance
variables that capture some element of having more games from playoff participation have an
incremental direct effect on revenue.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
77
TABLE 15
Panel A: Robustness Test of Regression Estimating Home Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Home Attendance Dependent Variable: Home Attendance
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.108
***
0.00 Winning Percentage 4612.087
***
0.00
Playoffs -0.004 0.40 Playoffs -147.130 0.30
All-Stars -0.004 0.60 All-Stars -168.681 0.26
Playoffs
t-1
0.038
***
0.00 Playoffs
t-1
615.938
***
0.00
Ln Fan Cost Index 0.143
**
0.02 Fan Cost Index 7.768
**
0.04
Ln Player Expenses 0.028 0.59 Player Expenses 11.481 0.40
Ln Arena Age with Renovation -0.007 0.56 Arena Age with Renovation -21.801 0.46
Ln Arena Capacity 0.715
***
0.00 Arena Capacity 0.556
***
0.00
Ln Population -0.012 0.59 Population -26.440 0.68
Ln Road Attendance 0.389
*
0.08 Road Attendance 0.363
*
0.10
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.555 Adjusted R
2
= 0.562
Note: Home Attendance is the mean attendance for the team’s home games. Winning Percentage is the team’s regular season winning percentage at time t.
Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at time t. All-Stars is the number of All-Stars on the team in a season.
Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4)
average-price tickets, two (2) small draft beers, four (4) small soft drinks, four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two
(2) least-expensive, adult-size adjustable caps. Player Expenses (in millions) is the amount spent on players’ compensation in a season. Arena Age with
Renovation is the numbers of years since the arena was either built or underwent major renovation. Arena Capacity is the number of attendance capacity of a
team’s arena. Population (in millions) is the team’s 2011 metropolitan area population. Road Attendance is the mean attendance for the team’s road games. Year
indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if
predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
78
Panel B: Robustness Test of Regression Estimating Home Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Home Attendance Dependent Variable: Home Attendance
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.107
***
0.00 Winning Percentage 4642.512
***
0.00
Season Performance -0.002 0.40 Season Performance -74.093 0.28
All-Stars -0.004 0.62 All-Stars -157.795 0.29
Playoffs
t-1
0.038
***
0.00 Playoffs
t-1
609.396
***
0.00
Ln Fan Cost Index 0.142
**
0.02 Fan Cost Index 7.768
**
0.03
Ln Player Expenses 0.028 0.60 Player Expenses 10.936 0.43
Ln Arena Age with Renovation -0.007 0.56 Arena Age with Renovation -21.750 0.47
Ln Arena Capacity 0.716
***
0.00 Arena Capacity 0.558
***
0.00
Ln Population -0.012 0.60 Population -25.522 0.69
Ln Road Attendance 0.398
*
0.08 Road Attendance 0.385
*
0.08
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.555 Adjusted R
2
= 0.563
Note: Home Attendance is the mean attendance for the team’s home games. Winning Percentage is the team’s regular season winning percentage at time t.
Season Performance is an ordered variable which takes a value of 1 if the team did not qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the
team advanced to the Conference Semifinals, 4 if the team advanced to the Conference Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won
the NBA Championship. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the
prior season and 0 otherwise. The Fan Cost Index™ comprises the prices of four (4) average-price tickets, two (2) small draft beers, four (4) small soft drinks,
four (4) regular-size hot dogs, parking for one (1) car, two (2) game programs and two (2) least-expensive, adult-size adjustable caps. Player Expenses (in
millions) is the amount spent on players’ compensation in a season. Arena Age with Renovation is the numbers of years since the arena was either built or
underwent major renovation. Arena Capacity is the number of attendance capacity of a team’s arena. Population (in millions) is the team’s 2011 metropolitan
area population. Road Attendance is the mean attendance for the team’s road games. Year indicators are included. The symbols *, **, and *** denote
coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
79
TABLE 16
Panel A: Robustness Test of Regression Estimating Road Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Road Attendance Dependent Variable: Road Attendance
Variable Coefficient p-value Coefficient p-value
Playoffs 0.005 0.14 Playoffs 78.847 0.17
Ln Player Expenses 0.028
**
0.01 Player Expenses 6.839
**
0.02
All-Stars 0.020
***
0.00 All-Stars 366.200
***
0.00
Playoffs
t-1
0.009
**
0.01 Playoffs
t-1
154.238
**
0.01
Ln Population 0.006
*
0.12 Population 18.785 0.16
Conference 0.001 0.87 Conference -19.393 0.87
NBA Championships 0.003
*
0.06 NBA Championships 50.867
**
0.05
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.552 Adjusted R
2
= 0.551
Note: Road Attendance is the mean attendance for the team’s road games. Playoffs is an indicator variable of 1 if the team made the playoffs and 0 otherwise at
time t. Player Expenses (in millions) is the amount spent on players’ compensation in a season. All-Stars is the number of All-Stars on the team in a season.
Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the prior season and 0 otherwise. Population (in millions) is the team’s 2011 metropolitan
area population. Conference is an indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team plays in the
Western Conference. NBA Championships is the number of NBA Championships won by a franchise. Year indicators are included. The symbols *, **, and ***
denote coefficients statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
80
Panel B: Robustness Test of Regression Estimating Road Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Road Attendance Dependent Variable: Road Attendance
Variable Coefficient p-value Coefficient p-value
Season Performance 0.005
*
0.05 Season Performance 82.689
*
0.05
Ln Player Expenses 0.029
**
0.01 Player Expenses 7.073
**
0.02
All-Stars 0.018
***
0.00 All-Stars 325.174
***
0.00
Playoffs
t-1
0.008
**
0.03 Playoffs
t-1
132.513
**
0.05
Ln Population 0.006
*
0.11 Population 18.838 0.17
Conference -0.001 0.88 Conference -15.660 0.89
NBA Championships 0.003
*
0.07 NBA Championships 47.276
*
0.05
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.561 Adjusted R
2
= 0.551
Note: Road Attendance is the mean attendance for the team’s road games. Season Performance is an ordered variable which takes a value of 1 if the team did not
qualify for the Playoffs, 2 if the team qualified for the Playoffs, 3 if the team advanced to the Conference Semifinals, 4 if the team advanced to the Conference
Finals, 5 if the team advanced to the NBA Finals, and 6 if the team won the NBA Championship. Player Expenses (in millions) is the amount spent on players’
compensation in a season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the team made the playoffs in the
prior season and 0 otherwise. Population (in millions) is the team’s 2011 metropolitan area population. Conference is an indicator variable set equal to one if the
team plays in the Eastern Conference, and equal to zero if the team plays in the Western Conference. NBA Championships is the number of NBA Championships
won by a franchise. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%, and 1% levels,
respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
81
With regards to Hypothesis 3 about team performance having a positive effect on a team’s brand
value, I find in alternative specification of Model 5 reported in Panels A and B of Table 15 that
Season Performance has a statistically significant positive effect on Road Attendance (p = 0.05,
one-tailed) while Playoffs has a positive coefficient, but does not quite reach statistical
significance (p = 0.14, one-tailed). The results are consistent with my main finding that relies on
a team’s regular season winning percentage. Hence, teams that have greater team performance as
captured by winning more regular season games and advancing deeper into the playoffs have
higher brand values.
Alternative Measure of Player Expenses
In the main tests of Hypothesis 1 and Hypothesis 4, team-level player compensation was
measured as the total compensation for player contracts. As noted above, an alternative measure
of team-level player compensation is Total Cost to Owner, which is defined as the sum of the
total compensation paid to players (Players Expenses) and the amount of luxury tax paid. As
reported on Table 17, I find consistent results with the main tests such that Total Cost to Owner
has a significant positive effect on team performance in Model 1 (p = 0.01, one-tailed), and as
reported on Table 18 that Total Cost to Owner has a significant positive effect on brand value in
Model 5 ((p = 0.02, one-tailed). These results make sense given the dependent relationship and
high positive correlation between player expenses and luxury tax expense (r = 0.59). Owners that
are willing to spend more on player compensation measured either by total amount given to
players or by actual total cost to owner, realize greater levels of team performance and higher
brand values.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
82
TABLE 17
Robustness Test of Regression Estimating Team Performance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Winning Percentage Dependent Variable: Winning Percentage
Variable Coefficient p-value Coefficient p-value
Ln Total Cost to Owner 0.158
***
0.00 Total Cost to Owner 0.001
**
0.01
All-Stars 0.179
***
0.00 All-Stars 0.087
***
0.00
Playoffs
t-1
0.127
***
0.00 Playoffs
t-1
0.061
***
0.00
Average Player Age 0.035
***
0.00 Average Player Age 0.013
***
0.01
Ln Population -0.039
**
0.02 Population -0.003
***
0.01
Conference -0.083
***
0.00 Conference -0.043
***
0.00
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.505 Adjusted R
2
= 0.570
Note: Winning Percentage is the team’s regular season winning percentage at time t. Total Cost to Owner (in millions) is the sum of the amount spent on players’
compensation and luxury tax payments for the season. All-Stars is the number of All-Stars on the team in a season. Playoffs
t-1
is an indicator variable of 1 if the
team made the playoffs in the prior season and 0 otherwise. Average Player Age is the average age of the players on a team. Population (in millions) is the team’s
2011 metropolitan area population. Conference is an indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team
plays in the Western Conference. Year indicators are included. The symbols *, **, and *** denote coefficients statistically different from zero at the 10%, 5%,
and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
83
TABLE 18
Robustness Test of Regression Estimating Road Game Attendance
1
Double-Logged Specification Linear Specification
Dependent Variable: Ln Road Attendance Dependent Variable: Road Attendance
Variable Coefficient p-value Coefficient p-value
Ln Winning Percentage 0.023
***
0.00 Winning Percentage 1086.153
***
0.00
Ln Total Cost to Owner 0.022
**
0.03 Total Cost to Owner 4.087
**
0.03
All-Stars 0.017
***
0.00 All-Stars 284.027
***
0.00
Playoffs
t-1
0.006 0.12 Playoffs
t-1
93.278 0.21
Ln Population 0.006
*
0.08 Population 20.305 0.12
Conference 0.001 0.87 Conference 31.803 0.79
NBA Championships 0.003
*
0.06 NBA Championships 50.615
**
0.05
Year Fixed Effects Included Year Fixed Effects Included
n = 386 team-years n = 386 team-years
Adjusted R
2
= 0.566 Adjusted R
2
= 0.568
Note: Road Attendance is the mean attendance for the team’s road games. Winning Percentage is the team’s regular season winning percentage at time t. Total
Cost to Owner (in millions) is the sum of the amount spent on players’ compensation and luxury tax payments for the season. Playoffs
t-1
is an indicator variable
of 1 if the team made the playoffs in the prior season and 0 otherwise. Population (in millions) is the team’s 2011 metropolitan area population. Conference is an
indicator variable set equal to one if the team plays in the Eastern Conference, and equal to zero if the team plays in the Western Conference. NBA
Championships is the number of NBA Championships won by a franchise. Year indicators are included. The symbols *, **, and *** denote coefficients
statistically different from zero at the 10%, 5%, and 1% levels, respectively (one-tailed if predicted, two-tailed otherwise).
1
Standard errors are robust standard errors clustered by team.
Running Head: FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
84
V. CONCLUSION
In this study, I examine a setting where managers have multiple objectives in a
competitive environment. In particular, I have examined the outcomes of a key expenditure
decision of NBA team owners related to how much money they spend on team-level player
compensation. These owners have complicated objectives of trying to field a winning team while
also running a for-profit business in which they have invested substantial financial resources
(Zimbalist 2003). Paying more to field a high-quality team can come at the sacrifice of current
profits, and indeed, I find data consistent with this competing view of objectives. Thus, team
owners must make an economic trade-off between current-period profits and winning games.
However, I also find support for an interdependent conceptualization of team owners’
objectives in that spending more on team-level player compensation can improve long-term
financial returns. I document two mechanisms by which owners can build the overall financial
value of their team. Specifically, I find that team values are improved by increasing revenue
generation and brand values, both of which depend significantly on a team’s performance which
depends in turn on investments in player compensation. Furthermore, I find support for timing
differences in realizing team owners’ objectives such that the effects of spending more on team-
level player compensation leads first to greater winning and then to greater overall financial
returns. Significantly, the effect on winning is mostly contemporaneous while the effect on profit
measures is a 3 to 10 year lag. Hence, NBA team owners who pursue short-term operational
profits by spending less on team-level player compensation likely do so at the expense of real-
economic future value. Such myopia is present even in a setting where agency frictions and other
causes of myopia are limited for NBA team owners
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
85
This study contributes to the accounting literature by demonstrating that absent many of
the typical causes of managerial myopia, business owners and managers can and do take actions
that have short-term benefits but which harm achievement of long-term objectives. This study
also contributes to the accounting literature by highlighting how the measurement of objectives
faced by business managers can affect how the relationship between the objectives and how
manager behavior are understood. That is, multiple objectives can be viewed as competing or
interdependent based on how the objectives are measured. Finally, this paper sheds light on
tension in prior sports management that has viewed objectives of sports franchise owners to be
competing by recognizing an alternative measure of the profit objective that leads to an
interdependent conceptualization of objectives.
Limitations
Data from professional sports teams is unique in that the data is complete, reasonably
accurate, and generally publicly available along with being in a setting that has fairly clear rules,
processes, and outcomes (Wolfe et al. 2005). However, a limitation of this study is the
generalizability of results to other organizations and settings. While the professional sports
setting shares common features with other industry settings, unique features of the setting do
limit the inferences that can be made. Financial data for the NBA used in this study comes from
estimates provided by Forbes which introduces noise in the accuracy of financial numbers. While
estimates can be noisy, the relative estimations of a team compared to other teams is likely more
reasonable.
Results in this study are average effects and rely more on the relative differences in
team’s financial and performance data than on absolute accuracy. As such, the overall tenor of
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
86
the results would likely remain consistent if actual team financial data were used. Finally, this
study is limited in as much as brand value is not directly measured but rather is captured by a
proxy, road game attendance. While arguments are provided to support its reasonableness as a
proxy measure, it is nevertheless an imperfect measure of brand value.
VI. FUTURE DIRECTIONS
Accounting Directions
This study examined a setting wherein many of the typical causes of managerial myopia
are either absent or limited due to the sample firms being closely-held private firms, normally
having one individual as the majority owner who is the primary decision agent and the residual
claimant (Fama & Jensen 1985). The majority of the accounting literature related to managerial
myopia has examined public firm settings and has found that managerial behavior can be
affected by such factors as capital market pressures and public accounting reporting requirements
(Mizik 2010; Bhojraj & Libby 2005; Lundstrum 2002; Stein 1989). This study provides further
evidence of managers taking actions to achieve a short-term objective even at the expense of
real-economic value and the realization of a longer term objective (Graham et al. 2005).
Typically, the accounting literature examining managerial myopia studies how manager
behavior is affected by how accounting is done, such as external or internal reporting
requirements creating incentives for individuals to act a certain way. In contrast, this study uses
accounting to better understand behavior of business managers by altering how fundamental
relationships of business objectives are portrayed. More specifically, this study uses accounting
knowledge of different performance measures for business objectives to provide insight into
behavior of sports franchise owners. In so doing, this study helps resolve tension in prior sports
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
87
management literature about sports franchise owner behavior related to their business objectives
of winning games and making profits. Thus, the accounting is informative to understanding the
setting and interpreting observed phenomenon rather than accounting rules being presented as
the explanatory cause, such as a reporting requirement would be. This approach of using
accounting rationale to explain observed phenomenon could be applied to other settings wherein
business managers face multiple objectives that could be measured in various ways and along
varying time horizons (Jensen 2001; Kaplan & Norton 1996; Simon 1964; Cyert & March 1963).
An advantage of the private firm setting, such as the one used in this study, is a
simplification of the setting as public reporting requirements are removed and firms are usually
smaller and organizationally less complex. However, the difficulty with private firms is that
availability of data is restricted. Hence, this study has heretofore relied on what publicly
available there is about the private firms in the sample. Going forward, I am working to cultivate
contacts within the NBA to gain access to additional firm specific data. At this point, I have had
meaningful communications with a Big 4 accounting partner who has done valuations on NBA
teams, the CFO of one NBA team, and the senior editor of Forbes magazine who oversees their
coverage of NBA teams’ finances and valuations. I plan to continue developing these contacts to
learn more about accounting and control issues faced by NBA teams and to obtain internal data
that can facilitate future research. For example, NBA teams may include performance measures
in their employees’ contracts (e.g., General Manager, CFO, and other team executives) that could
have internal reporting requirements. This could be a useful setting to examine how tradeoffs are
made between financial measures affected by reporting requirements and non-financial measures
related to team operations and on-the-court performance. Given the high value placed on
nonfinancial objectives (e.g., the presentation and performance of what firms produce (i.e.,
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
88
winning games)), the sports entertainment industry could be an interesting setting for balanced
scorecard related research, building upon this study’s examination of tradeoffs between making
profits and winning games (Kaplan & Norton 1996).
Access to internal data would also allow examination of multiple levels of decision
making and behavior within the sports franchise. For example, employment contracts for general
managers could reveal significant variation in delegation of decision rights from owners and
variation in performance evaluations. Thus, while the present study focuses on team owners,
making agency problems limited to a large extent, agency conflicts are likely to exist lower in
the organization with general managers, coaches, and players as information asymmetry and goal
conflict likely increase (Eisenhardt 1989; Jensen & Meckling 1976). For example, if individual
players are assumed to be self-interested in line with economic theory, which seems to be a
reasonable assumption, then goal conflict may increase as players’ interests may not be aligned
with those of team owners. On the other hand, information asymmetry is likely to be low because
an unique feature of the sports entertainment setting is that the performance of individual
employees (e.g., the players) is widely observable with generally complete and accurate
performance measures (Wolfe et al. 2005). This feature could allow for clean, powerful tests of
the efficiency and effectiveness of incentive contracts. One potential research setting could be
comparing the performance of players with expiring contracts against prior and future
performance with the intent of examining if there is a performance spike in contract years (Stiroh
2007). If performance does spike in years when contracts are expiring, then future research could
explore what other incentives could be included in contracts or what control system mechanisms
could be put in place to achieve the higher level of performance in years when contracts are not
expiring.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
89
Another natural extension of the present study is to delve deeper into the characteristics
of team owners that are likely to have an effect on their decision making and behavior. This
study finds that the net worth of individual owners has a significant effect on how much financial
resources they commit to team-level player compensation. The accounting, finance, and
management literatures have recently had a growing interest in the effect of individual
differences on firm related outcomes (e.g., hubris, overconfidence, and narcissism of CEOs)
(Olsen, Dworkis, & Young 2014; Chatterjee & Hambrick 2011, 2007; Malmendier & Tate
2005). Drawing upon these literatures, future work could examine additional personal
characteristics of sports franchise owners and how they affect managerial decision making. To
this end, I am currently gathering personal biographies of each owner and attempting to identify
commonalities and differences in personalities, attitudes, education, employment, and life
experiences across owners that could provide insight into their management styles and
investment decisions. Given the diversity, wealth, and visibility of sports franchise owners, these
individuals form an interesting group of people to observe and study.
Finally, results from this study highlight an important financial accounting situation
related to accounting for intangibles. This study shows that the brand value of a team can be a
significant source of economic profits as a driver of revenues and team values. However,
intangibles such as brand value have notoriously been difficult to value and account for
(Canibano, Garcia-Ayuso, & Sanchez 2000). Brand value is most likely to show up as part of
goodwill, calculated upon the sell of an entity. Yet, sports franchises are not frequently sold and
therefore team values are infrequently determined by a market mechanism. Forbes does provide
yearly estimates of values, but these are not part of a purchase transaction which would give rise
to a need to account for intangibles. Future research can continue developing valuation models
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
90
for firms like sports franchises that derive significant economic profits from their intangible
assets (Alexander & Kern 2004).
Sports Entertainment Future Directions
This study also highlights some future directions for research in sports entertainment. The
present study relies on team-level player compensation in its various empirical models, but does
not address the composition of the team. Teams can be built in a variety of ways including
drafting rookies, signing free agents, and player trades. Teams can also vary their strategy about
the composition of their team, meaning they could field their team focusing on superstar players,
veteran players, young and talented but unproven players, or a balanced mix. Future research
could examine how teams are built, the associated costs, financial performance, and on-the-court
performance (Spurr 2000). While this study does suggest that spending more on team-level
player compensation has a positive effect on on-the-court performance, other factors such as
team composition could be considered. This research may be able to speak to the optimality and
cost efficiencies of various strategic approaches to fielding a team.
Results from this study also highlight the significant effect that a team’s arena has on its
financial performance. As seen in Panel C of Table 7, arena age has a significant negative effect
on revenue, indicating that newer arenas are associated with higher revenue. Arenas are
generally not owned by the sports franchises themselves, but rather are developed in conjunction
with the cities they play in. Sports franchises effectively benefit from a subsidy from taxpayers in
the local community to provide a place for the games to be held. Furthermore, newer arenas
provide a significant increase in the ability to generate revenue. How sports franchises
collaborate with local governments to build or renovate their arenas along with the economic
impact of hosting games could be a fruitful avenue for future sports entertainment research.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
91
REFERENCES
Aaker, D. A., and R. Jacobson. (1994). The Financial Information Content of Perceived Quality. Journal
of Marketing Research, 31, 191–201.
Abernethy, M.A., Bouwens, J., and Van Lent, L. (2013). The Role of Performance Measures in the
Intertemporal Decisions of Business Unit Managers. Contemporary Accounting Research, 30,
925 – 961.
Alexander, D. L. (2001).Major league baseball: monopoly pricing and profit-maximizing behavior.
Journal of Sports Economics, 2, 341–355.
Alexander, D. L. and Kern, W. (2004). The Economic Determinants of Professional Sports Franchise
Values. Journal of Sports Economics, 5, 51 – 66.
Amir, E. and B. Lev. (1996). Value-relevance of nonfinancial information: The wireless communication
industry. Journal of Accounting and Economics, 22, 3–30.
Artz, M., Homburg, C., and Rajab, T. (2012). Performance-measure system design and functional
strategic decision influence: The role of performance-measure properties. Accounting,
Organizations and Society, 37, 445 – 460.
Asker, J., Farre-Mensa, J., and Ljungqvist, A. (2014). Corporate Investment and Stock-Market Listing:
A Puzzle? Working Paper.
Badenhausen, K. (2014). As Stern Says Goodbye, Knicks, Lakers Set Records As NBA’s Most Valuable
Teams. Forbes. January 22, 2014.
Baron, R. M., and D. A. Kenny. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173–1182.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
92
Barth, M.E., Clement, M.B., Foster, G., and Kasznik, R. (1998). Brand Values and Capital Market
Valuation. Review of Accounting Studies, 3, 41 – 68.
Benjamin, B. A., and Podolny, J. M. (1999). Status, quality, and social order in the California wine
industry. Administrative Science Quarterly, 44, 563–589.
Berman, S.L., Down, J., & Hill, C.W.L. (2002). Tacit knowledge as a source of competitive advantage
in the National Basketball Association. Academy of Management Journal, 45, 13-31.
Berri, D.J., Schmidt, M.B., and Brook, S. L. (2004). The Impact of Star Power on NBA Gate Revenues.
Journal of Sports Economics, 5, 33 – 50.
Berri, D.J., Schmidt, M.B., and Brook, S. L. (2006). The Wages of Wins: Taking Measure of the Many
Myths in Modern Sport. Stanford, CA: Stanford University Press.
Berri, D. J. and Schmidt, M.B. (2006). On the Road With the National Basketball Association’s
Superstar Externality. Journal of Sports Economics, 7, 347 – 358.
Bhojraj, S. and Libby, R. (2005). Capital Market Pressure, Disclosure Frequency-Induced
Earnings/Cash Flow Conflict, and Managerial Myopia. The Accounting Review, 80, 1-20.
Bloom, M. (1999). The performance effects of pay dispersion on individuals and organizations.
Academy of Management Journal, 42, 25-40.
Bryant, H. (2014). The NFL’s Great Disappearing Act. ESPN The Magazine. September 01.
Bryant, L., D. A. Jones, and S. K. Widener. (2004). Managing value creation within the firm: An
examination of multiple performance measures. Journal of Management Accounting Research, 16, 107-
131.
Bushee, B.J. (1998). The Influence of Institutional Investors on Myopic R&D Investment Behavior. The
Accounting Review, 73, 305 – 333.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
93
Canibano, L., Garcia-Ayuso, M., and Sanchez, P. (2000). Accounting for Intangibles: A Literature
Review. Journal of Accounting Literature, 19, 102 – 130.
Chandra, U. and Ro, B.T. (2008). The Role of Revenue in Firm Valuation. Accounting Horizons, 22,
199 – 222.
Chatterjee, A. and D.C. Hambrick. (2007). It’s all about me: narcissistic chief executive officers and
their effects on company strategy and performance. Administrative Science Quarterly, 52, 351 –
386.
Chatterjee, A. and D.C. Hambrick. (2011). Executive personality, capability cues, and risk taking: how
narcissistic CEOs react to their successes and stumbles. Administrative Science Quarterly, 56,
202 – 237.
Coon, L. (2012). NBA Salary Cap/Collective Bargaining Agreement FAQ. Accessed at
http://www.cbafaq.com/salarycap.htm
Cooper, J. C., and F. H. Selto. (1991). An experimental examination of the effects of SFAS No. 2 on
R&D investment decisions. Accounting, Organizations and Society, 16, 227–242.
Cyert, R. M., and March, J. G. (1963). A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice
Hall.
Dechow, P. (1994). Accounting earnings and cash flows as measures of firm performance: the role of
accounting accruals. Journal of Accounting and Economics, 18, 3 – 42.
Dechow, P. M. and Sloan, R.G. (1991). Executive incentives and the horizon problem: An empirical
investigation. Journal of Accounting and Economics, 14, 51 - 89.
Dietl, H.M., Grossmann, M., and Lang, M. (2011). Competitive Balance and Revenue Sharing in Sports
Leagues With Utility-Maximizing Teams. Journal of Sports Economics, 12, 284 - 308.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
94
Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. The Academy of Management
Review, 14, 57 – 74.
Ertug, G. and Castellucci, F. (2013). Getting what you need: How reputation and status affect team
performance, hiring, and salaries in the NBA. Academy of Management Journal, 56, 407 – 431.
Fama, E.F. and Jensen, M.C. (1983). Agency Problems and Residual Claims. Journal of Law and
Economics, 26, 327 –349.
Fama, E.F. and Jensen, M.C. (1985). Organizational Forms and Investment Decisions. Journal of
Financial Economics, 14, 101–119.
Feltham, G.A. and Xie, J. (1994). Performance Measure Congruity and Diversity in Multi-Task
Principal / Agent Relations. The Accounting Review, 69, 429 – 453.
Fort, R. and Quirk, J. (1995). Cross-subsidization, Incentives, and Outcomes in Professional Team
Sports Leagues. Journal of Economic Literature, 23, 1265 – 1299.
Fort, R. and Quirk, J. (2004). Owner objectives and competitive balance. Journal of Sports Economics,
5, 20 -32.
Giratikanon, T., Katz, J., Leonhardt, D., and Quealy, K. (2014). Which Team Do You Cheer For? An
N.B.A. Map. The Upshot in The New York Times,. May 12, 2014.
Gladden, J.M., Irwin, R.L., and Sutton, W.A. (2001). Building Brand Equity. Journal of Sport
Management, 5, 297 – 317.
Graham, J.R., Harvey, C. R., and Rajgopal, S. (2005). The economic implications of corporate financial
reporting. Journal of Accounting and Economics, 40, 3-73.
Harder, J.W. (1992). Play for pay: Effects of inequity in a pay-for-performance context.
Administrative Science Quarterly, 37, 321-335.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
95
Hausman, J.A., & Leonard, G.K. (1997). Superstars in the National Basketball Association: Economic
value and policy. Journal of Labor Economics, 15, 586-624.
Healy, P.M. (1985). The Effect of Bonus Schemes on Accounting Decisions. Journal of Accounting and
Economics, 7, 85 – 107.
HoopsHype. (2014). Owners. Accessed at http://hoopshype.com/owners.htm
Ittner, C. D., and D. F. Larcker. (1998). Are nonfinancial measures leading indicators of financial
performance? An analysis of customer satisfaction. Journal of Accounting Research, 36,
1–35.
Jensen, M.C. (2001). Value Maximization, Stakeholder Theory, and the Corporate Objective Function.
Journal of Applied Corporate Finance, 14, 8 – 21.
Jensen, M.C. and Meckling, W.H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and
Ownership Structure. Journal of Financial Economics, 3, 305 – 360.
Kaplan, R.S. and Norton, D.P. (1996). Using the Balanced Scorecard as a Strategic Management
System. Harvard Business Review. January – February, 75 – 85.
Keating, P. (2014). The Numbers. ESPN The Magazine. May 26, 2014.
Keidel, R.W. (1984). Baseball, Football, and Basketball: Models for Business. Organizational
Dynamics. Vol. 12, 5-18.
Keller, K. L. (1997). Strategic Brand Management. Upper Saddle River, New Jersey: Prentice Hall.
Kenny, D. A., D. A. Kashy, and N. Bolger. (1998). Data analysis in social psychology. In Handbook of
Social Psychology, Vol. 1, edited by Gilbert, D. T., S. T. Fiske, and A. Lindzey, 233–265. New
York, NY: McGraw-Hill.
Késenne, S. (2014). The Economic Theory of Professional Team Sports: An Analytical Treatment. Glos,
England: Edward Elgar Publishing Limited.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
96
Laverty, K. J. (1996). Economic “short-termism”: The debate, the unresolved issues, and the
implications for management practice and research. Academy of Management Review, 21, 825–
60.
Lee, C.M.C. (1999). Accounting-Based Valuation: Impact on Business Practices and Research.
Accounting Horizons, 13, 413 – 425.
Leeds, M. and von Allmen, P. (2013). The Economics of Sports. Upper Saddle River, NJ: Prentice Hall.
Lundstrum, L. L. (2002). Corporate investment myopia: a horserace of the theories. Journal of
Corporate Finance, 8, 353 – 371.
Madden, T.J., Fehle, F., and Fournier, S. (2006). Brands Matter: An Empirical Demonstration of the
Creation of Shareholder Value Through Branding. Journal of the Academy of Marketing Science,
34, 224 – 235.
Malmandier, U. and G. Tate. (2005). CEO overconfidence and corporate investment. Journal of
Finance, 60, 2661-2700.
Merchant, K.A. (1990). The effects of financial controls on data manipulation and management Myopia.
Accounting, Organizations and Society, 15, 297 – 313.
Merchant, K.A. and Van der Stede, W.A. (2012). Management Control Systems: Performance
Measurement, Evaluation and Incentives. Essex, England: Pearson Education Limited
Mizik, N. (2010). The Theory and Practice of Myopic Management. Journal of Marketing Research, 47,
594 – 611.
Olsen, K.J., K.K. Dworkis, and S.M. Young. (2014). CEO narcissism and accounting: a picture of
profits. Journal of Management Accounting Research, 26, 243-267.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
97
Pfeffer, J. and Davis-Blake, A. (1986). Administrative Succession and Organizational Performance:
How Administrator Experience Mediates the Succession Effect. Academy of Management
Journal, 29, 72 – 83.
Quirk, J. and Fort, R. (1992). Pay dirt: The business of professional team sports. Princeton, NJ:
Princeton U. Press.
Rishe, P. (2011). NBA Owners Win Big with New Collective Bargaining Agreement. Forbes.
November 28, 2011.
Rosner, S. R. and Shropshire, K.L. (2004). The Business of Sports. Sudbury, MA: Jones and Bartlett
Publishers.
Sedatole, K. L. (2003). The effect of measurement alternatives on a nonfinancial quality measure’s
forward-looking properties. The Accounting Review, 78, 555–80.
Settimi, C. (2014). The NBA’s Richest Local Television Deals. Forbes. January 22, 2014.
Seybert, N. (2010). R&D Capitalization and Reputation-Driven Real Earnings Management. The
Accounting Review, 85, 671 - 693.
Silver, N. (2014). Cabrera’s Millions and Baseball’s Billions. FiveThirtyEight. March 28, 2014.
Simmons, B. (2014a). The World’s Most Exclusive Club. Grantland. April 17, 2014.
Simmons, B. (2014b). The Worst Contracts in the NBA. Grantland. February 18, 2014.
Simmons, R. and Berri, D.J. (2011). Mixing the princes and the paupers: Pay and performance in the
National Basketball Association. Labour Economics, 381 – 388.
Simon, H.A. (1964). On the Concept of Organizational Goal. Administrative Science Quarterly, 9, 1 –
22.
Simon, C.J. and Sullivan, M.W. (1993). The Measurement and Determinants of Brand Equity: A
Financial Approach. Marketing Science, 1, 28 – 52.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
98
Spurr, S.J. (2000). The Baseball Draft: A Study of the Ability to Find Talent. Journal of Sports
Economics, 1, 66 – 85.
Staw, B.M., and Hoang, H. (1995). Sunk costs in the NBA: Why draft order affects playing time and
survival in professional basketball. Administrative Science Quarterly, 40, 474-494
Stein, J.C. (1989). Efficient capital markets, inefficient firms: A model of myopic corporate behavior.
The Quarterly Journal of Economics, 104, 655 - 669.
Stiroh, K.J. (2007). Playing for Keeps: Pay and Performance in the NBA. Economic Inquiry, 45, 145 –
161.
Taylor, B.A., & Trogdon, J.G. (2002). Losing to win: Tournament incentives in the National Basketball
Association. Journal of Labor Economics, 20, 23-41.
Van Riper, T. (2013). The NBA’s Billionaire Owners. Forbes. January 23, 2013.
Vrooman, J. (1995). A general theory of professional sports leagues. Southern Economic Journal, 61,
971–990.
Vuong, Q.H. (1989). Likelihood ratio tests for model selection and non-nested hypotheses.
Econometrica, 57, 307 – 333.
Wang, L. (2009). Money and fame: Vividness effects in the National Basketball Association. Journal of
Behavioral Decision Making, 22, 20-44.
Windhorst, B. (2013). Heat waive veteran Mike Miller. ESPN.COM. July 16.
Wise, M. (1998). Basketball; It’s Their Ball, and NBA Owners Call for Lockout. The New York Times.
June 30, 1998.
Wolfe, R.A., Weick, K.E., Usher, J.M., Terborg, J.R., Poppo, L., Murrell, A.J., Dukerich, J.M., Core,
D.C., Dickson, K.E., and Jourdan, J.S. (2005). Sport and organizational studies: Exploring
synergy. Journal of Management Inquiry, 14, 182-210.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
99
Zeithaml, V. (2000). Service quality, profitability, and the economic worth of customers: What we know
and what we need to learn. Journal of the Academy of Marketing Science, 28, 67–85.
Zimbalist, A. (2003). Sport as Business. Oxford Review of Economic Policy, 19, 503 – 511.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
100
APPENDIX
Variable Definitions
Variable Definition
All-Stars Number of All-Stars on a team for a season.
Arena Capacity Seating capacity of a team’s home arena.
Arena Age with Renovation Years since arena was either built or had major renovation
work done.
Average Player Age Average age of players on a team as of February 01 during
a season.
Conference An indicator variable set equal to one if the team plays in
the Eastern Conference, and equal to zero if the team plays
in the Western Conference.
Fan Cost Index The Fan Cost Index™ comprises the prices of four (4)
average-price tickets, two (2) small draft beers, four (4)
small soft drinks, four (4) regular-size hot dogs, parking for
one (1) car, two (2) game programs and two (2) least-
expensive, adult-size adjustable caps.
Home Attendance Average attendance for home games during the regular
season.
NBA Championships Number of NBA Championships won by a franchise.
Operating Income Net earnings before interest, taxes, depreciation, and
amortization.
Overall Return Change in team value plus operating income divided by
prior year team value.
Player Expenses Total team-level player compensation.
Playoffs An indicator variable set equal to one if a team made the
playoffs in the current year, and equal to zero otherwise.
Playoffs
t-1
An indicator variable set equal to one if a team made the
playoffs in the prior year, and equal to zero otherwise.
FOR THE LOVE OF THE GAME? OWNERSHIP AND CONTROL IN THE NBA
101
APPENDIX
Variable Definitions, Continued
Variable Definition
Population Metropolitan area population for team’s home city,
measured as of 2011.
Revenue Total revenue, net of revenue sharing and stadium debt
service.
Road Attendance Average attendance for road games during the regular
season.
Season Performance An ordered variable which takes a value of 1 if the team did
not qualify for the Playoffs, 2 if the team qualified for the
Playoffs, 3 if the team advanced to the Conference
Semifinals, 4 if the team advanced to the Conference
Finals, 5 if the team advanced to the NBA Finals, and 6 if
the team won the NBA Championship.
Team Value Estimated value of a team by Forbes. While Forbes does
not provide their valuation model for team values, they do
note four broad valuation factors they consider including
teams’ finances, market, arena arrangements, and brand.
Total Cost to Owner The sum of the total compensation paid to players (Players
Expenses) and the amount of luxury tax paid.
Value Change Change in team value from prior year.
Value Change Change in team value plus operating income
plus Operating Income
Value Return Change in team value divided by prior year team value.
Winning Percentage Regular season winning percentage calculated as the
number of wins divided by games played.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Enchanted algorithms: the quantification of organizational decision-making
PDF
Essays on supply chain management
PDF
Who's the physician in charge? Generalist and specialist jurisdictions in professional practice
PDF
The effects of a customer's comparative processing with positive and negative information on product choice
PDF
Two essays on the mutual fund industry and an application of the optimal risk allocation model in the real estate market
PDF
Topics in algorithms for new classes of non-cooperative games
PDF
Three essays on distance: examing the role of institutional distance on foreign firm entry, local isomorphism strategy and subsidiary performance
PDF
How product designs and presentations influence consumers’ post-acquisition decisions
PDF
Quango reforms and challenges in South Korea: social relations, informal networks, and hidden actions
PDF
Evolution of returns to scale and investor flows during the life cycle of active asset management
PDF
Understanding virality of YouTube video ads: dynamics, drivers, and effects
PDF
Empirical methods in control and optimization
PDF
Urban universities' campus expansion projects in the 21st century: a case study of the University of Southern Calfornia's "Village at USC" project and its potential economic and social impacts on...
PDF
The impacts of manufacturers' direct channels on competitive supply chains
PDF
Knowledge, motivation, and organizational influences in aligning learning to business performance results: a study of a promising practice
PDF
Modeling customer choice in assortment and transportation applications
PDF
Habits as friend and foe to self-control: the dual roles of routinized behavior
PDF
Essays on the role of entry strategy and quality strategy in market and consumer response
PDF
Large-scale multiple hypothesis testing and simultaneous inference: compound decision theory and data driven procedures
PDF
Essays on the economics of digital entertainment
Asset Metadata
Creator
Olsen, Kari Joseph
(author)
Core Title
For the love of the game? ownership and control in the NBA
School
Marshall School of Business
Degree
Doctor of Philosophy
Degree Program
Business Administration
Publication Date
03/05/2015
Defense Date
02/17/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
brand value,firm value,myopia,NBA,OAI-PMH Harvest,ownership objectives,performance measurement
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Young, S. Mark (
committee chair
), Allen, Eric (
committee member
), Fiss, Peer C. (
committee member
)
Creator Email
kari.olsen@usc.edu,kjolsen@outlook.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-537110
Unique identifier
UC11297784
Identifier
etd-OlsenKariJ-3219.pdf (filename),usctheses-c3-537110 (legacy record id)
Legacy Identifier
etd-OlsenKariJ-3219.pdf
Dmrecord
537110
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Olsen, Kari Joseph
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
brand value
firm value
myopia
NBA
ownership objectives
performance measurement