1. Introduction Walter C. Neale’s (1964) exposition of ‘the peculiar economics of sports’ helped to focus attention upon two important and closely related themes in the economic analysis of professional team sports. The first theme concerns an adequate economic conception of sporting leagues and the second concerns a clear statement of the ‘uncertainty of outcome hypothesis’. More specifically, Neale made the first attempt to define a sporting league in economic terms and argued that leagues were examples of multi-plant monopolies being singly responsible for the administration of sports. For example, a team cannot determine its own output level (as measured by the number of games). That depends on how many teams the league admits and on how many times the league decrees teams play one another during the season. For Neale (1964), a central reason why leagues emerged as monopolies was the ‘Louis-Schmelling Paradox’. Along with Rottenberg (1956), this was one of the first formal statements of the uncertainty of outcome hypothesis, which states that sports-fan interest is greatest when sporting competition is at its most intense. While each team strives to outperform its competitors, both on the field and financially, the league’s survival depends on the maintenance of a competitive balance. In sports, every act of production requires the presence of an opponent – hence the paradox. In contrast to Neale, Sloane (1971), in an analysis of European football, argued that a sporting league and its constituent teams may be more accurately viewed as a cartel than as a multi-plant firm. Teams in most sports take decisions about investment, about whether to produce at all, and they usually have substantial control over ticket prices. They also control merchandising which is now a major source of revenue. One of the most important aspects of Sloane’s definition of sporting leagues is that it provides a more plausible framework within which to explore the potential problems associated with managing the mutual interdependence between teams. Neale’s definition could imply that centralised control coordinates passive teams. The problem of mutual interdependence can be illustrated in the context of league management policies. Many policies of cross-subsidisation have been justified on the basis of the uncertainty of outcome hypothesis. Leagues have attempted to transfer resources from stronger, more successful teams, to smaller less successful teams to try to initiate a ‘virtuous circle’ of overall benefit to the league that can be actively promoted and sustained by cross-subsidisation policies. The outcome of these policies has been weak. Thus, theoretical attempts to examine the effectiveness of cross-subsidisation policies in improving competitive balance have been developed. They have their origin in a cartel 5 model of sporting leagues developed by El Hodiri and Quirk (1971)1 . These models take as their starting point a situation of competitive imbalance and examine the consequences of changes in league policies. Based on predictions developed from this model, Fort and Quirk (1995) provide some statistical inferences to support the view that these policies have been ineffective in the US. Quirk and Fort (1992) also provide detailed descriptive historical statistics to suggest that US Leagues tend to conform to long-run domination. In the UK Dobson and Goddard (1995) provide similar findings. Though the literature is somewhat scant this evidence can be seen as offering support for the ‘invariance proposition’. This was originally suggested by Rottenberg (1956). It states that a league’s competitive balance (talent distribution) is independent of the allocation of the ownership right to sell players services. It is a direct application of Coase’s theorem. A common characteristic of the modeling approach in this literature is the assumption that talent is homogeneous and perfectly divisible and, therefore, well suited for marginal analysis. An immediate consequence of this approach is that a team is identified, apart from some exogenous characteristics, by the total amount of talent it possesses. Furthermore, the price of talent (salaries) is determined in a competitive market2 .