Economists are notoriously vague as to what a sports team produces. The most popular measure of output is winning percentage, but winning percentage is not the only candidate for measuring sporting contest output. Other suggestions include attendance, the ratio of team i's to team j's scoring, and games produced jointly. Michael Leeds and Peter yon Allmen (2005) have briefly discussed the definition of output and mentioned that the definition of output can depend on the focus of the market. This is true. But do the metrics supplied in the literature make sense measuring what a sports team produces?
A sports team employs capital (a stadium or ballpark, and sporting equipment) and labor (players, coaches, trainers, and front office staff) to produce the sporting contest. Thus, according to neoclassical economic theory, the inputs hired by the team should relate to team output. Yet, as will be demonstrated, some of the metrics suggested for sporting contest output will fail this idea and thus ultimately must be rejected as an appropriate measure of production.
In contrast, institutional economists such as Thorstein Veblen (1904) have stated that output is the product of a joint stock of knowledge. Therefore, the individual productivity of capital and labor is impossible to attribute to output. Clarence Ayres elaborated on Veblen's idea by stating that "[e]conomic science has no technique of independent measurement of any of these entities, utility, productivity, or value" (1978, 80). Rodney Stevenson summed up the institutionalist theory of production by stating that "the process of production is defined by and implemented through the institutional structure of society" (1988, 70).
This paper is organized by looking at previous definitions of sporting contest output (attendance, ratio of team i's scoring to team j's, winning percentage or wins, and games) in the literature and addressing any deficiencies in these definitions. This will be followed by a brief discussion of the implications of various output definitions and, finally, some conclusions.
Is Team Output Attendance?
One candidate is attendance (Rottenberg 1956, 255). In fact, Donald Alexander used attendance as a measure of output to examine Major League Baseball (MLB) pricing behavior (2001). Assuming that the stadium is not at capacity, J. Cairns, N. Jennett, and P. J. Sloane argued that this measure "would be misleading since the production of a match watched by 30,000 spectators requires few additional resources compared to one watched by 15,000" (1986, 11).
For example, on November 22, 2003, the National Hockey League held an outdoor contest in which 57,167 attended the sporting event between the Montreal Canadians and the Edmonton Oilers. A few weeks earlier (November 4, 2003), the same two teams played in Montreal, but the attendance was 18,218. If team output is attendance, then the sporting contest in Edmonton was more than three times the output of the sporting contest played a few weeks earlier between the same two teams. No doubt gate (ticket) revenue from the sporting contest is greater with more in attendance, but this is a revenue measure, not a production measure. The inputs employed by the team do not directly produce a sporting contest if attendance is used as the metric in measuring sporting contest output, which violates the neoclassical theory of production.