Academic journal article Contemporary Economic Policy

Revenue, Population, and Competitive Balance in Major League Baseball

Academic journal article Contemporary Economic Policy

Revenue, Population, and Competitive Balance in Major League Baseball

Article excerpt

I. INTRODUCTION

One common prediction in the literature on sports leagues is that market size will influence the distribution of wins. Both win-maximizing models of sports leagues and profit-maximizing models predict a positive impact of market size on team performance. Teams in large markets have a larger base of potential fans leading to higher attendance and larger media contracts for a given win percent. The result is greater revenue for large-market teams and the expectation of a higher win percent. The mechanism by which higher revenue leads to increased performance may be that owners pursue the sports goal of winning, so the larger revenue will be used to hire more talented players. Alternatively, if owners are assumed to pursue profits, large-market teams will exhibit a higher equilibrium win percent only if their marginal revenue (MR) of a win is greater than the MR of a win for small-market teams.

For many baseball fans, baseball executives, and baseball economists, this connection between market size and winning is intuitively obvious. But, the magnitude of this effect is an empirical question. A large home metropolitan area (consolidated metropolitan statistical area [CMSA]) will theoretically generate more revenue leading to competitive advantage, but the theory does not tell us if the effect is large or small. In analyzing the competitive balance of a sports league, the magnitude of the population effect is crucial.

This article investigates the impact of CMSA population on team revenue, the MR of a win, and team win percent (WP) for Major League Baseball (MLB) using a log-linear simultaneous equations model for the 1997-2001 seasons. An endogeneity test shows that revenue and win percent are interdependent indicating that the usual single-equation estimation of team revenue functions is inconsistent. Results from our model indicate that teams from larger cities do have higher revenue. Higher team revenue and higher MR of a win lead to increased payrolls, but the effort to hire more talent is not consistently rewarded with increased win percent. As a result, the impact of population on win percent is fairly small.

II. LITERATURE REVIEW

Various researchers modeling sports leagues have discussed the theoretical impact of CMSA population on WP (see Gustafson and Hadley, 1996; Marburger, 1997; Quirk and El-Hodiri, 1974; Quirk and Fort, 1992; Rottenberg, 1956). In this literature, it is universally agreed that the size of a team's CMSA should have a positive impact on WP. Since population itself does not directly impact win percent, we need to consider a causal chain of events that leads from population to winning. First, increased population must lead to higher revenue and then that higher revenue must be used to buy player talent that will lead to increased win percent.

There is an extensive empirical literature that supports the first link in the chain: population is an important determinant of attendance and revenue. Earlier articles focused on attendance, rather than revenue, because of the availability of these data. Some examples that document a strong positive effect of CMSA population on attendance include Demmert (1973), Noll (1974), Bruggink and Eaton (1996), and Gustafson, Hadley, and Ruggiero (1999). In addition to larger attendance revenues, large-market teams enjoy richer media contracts.

The profit-maximizing incentive for large-market teams to spend their additional revenue on talent is that the MR of a win may be greater in a larger market. Recent articles by Burger and Walters (2003), Krautmann (forthcoming), Solow and Johnson (2004), and Solow and Krautmann (2005) have conducted econometric analyses of MLB revenue functions in order to study how MR of a win varies. The Burger and Walters model allows MR of a win to vary with population and with win percent, while the Krautmann model allows it to vary only with population. The revenue function in the model of Burger and Walters is estimated using data on local revenue, while Krautmann uses both local revenue and total revenue. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.