Moneyball And Major League Teams Pay-Performance: A Case Of ValuationVW Staff
Moneyball And Major League Teams Pay-Performance: A Case Of Valuation Anomaly And Adaptive Market Efficiency?
University of Toronto – Rotman School of Management
March 1, 2016
An often discussed concept in accounting is the pay-performance relation, particularly with respect to executive compensation (known as pay-performance sensitivity, or PPS). This study explores the pay-performance relation in major league baseball teams, focusing on player payroll and its effects on teams’ performance. Moneyball (2003) exposed readers to the use of advanced analytics, or Sabermetrics, in baseball and how it improved the ‘bang for the buck’ in selecting baseball players and in managing games. It also offered the seductive idea that quantitative geeks could beat jocks in baseball personnel decisions. The study also contributes to the accounting literature on adaptive market efficiency and vanishing pricing anomalies by testing whether the use of Sabermetrics has indeed provided an unfair advantage to teams and general managers and, more importantly, whether the outing of these ideas has made the Moneyball effect go away.
Moneyball And Major League Teams Pay-Performance: A Case Of Valuation Anomaly And Adaptive Market Efficiency? – Introduction
There was but one question he left unasked, and it vibrated between his lines: if gross miscalculations of a person’s value could occur on a baseball field, before a live audience of thirty thousand, and a television audience of millions more, what did that say about the measurement of performance in other lines of work? If professional baseball players could be over- or under valued, who couldn’t?” (Michael Lewis, Moneyball: The Art of Winning an Unfair Game, 2003)
Pay-performance is regularly discussed in the accounting literature, particularly in the context of pay-performance sensitivity (PPS), which focuses on the change in the pay of chief executive officers and its association with the performance their companies (e.g., Dai, Jin, & Zhang, 2014, Hamm, Jung, & Wang, 2015, and Bushman, Zhonglan, & Weining, 2016). This study, in contrast with the PPS literature, examines the relationship between labour cost (albeit famous employees) and performance of Major League Baseball (MLB) teams. In addition, as discussed below, this study also contributes to the literature of adaptive capital market efficiency and mispricing anomalies.
Moneyball (2003) has changed sports culture, as demonstrated by the fact that the word “Moneyball” has become part of the baseball vocabulary and teams using Sabermetric thinking are described as playing “Moneyball” (Wikipedia, n.d.). The impact of the book and its popularity goes well beyond sports culture. It sold over a million copies and was made into a movie in 2011, starring Brad Pitt as Billy Beane, that was nominated for six Academy Awards, including Best Picture and Best Actor. The book also made it to the business world as shown by the Wall Street Journal article about it (Futterman, 2011). The entrance of the book into popular culture is evidenced by the fact that the Simpsons had an episode about it in 2010 called MoneyBart (Abbott, 2010). One reason that the book has captured the imagination of many Baseball fans could be because it has led to the romantic notion that quantitative geeks can best jocks in athletic personnel decisions, and, consequently, would make better general managers (GMs) for sports teams. This sentiment is evidence by a talk at the 2013 MIT Sloan Sports Analytics Conference whose title was “Revenge of the Nerds” (Van Riper, 2013).
Moneyball (2003) describes how the use of advanced baseball analytics, also known as Sabermetrics1, has led the Oakland Athletics to superior performance despite the low payroll of the team. As such, the question that the book, and the theory behind it, raises is whether there is a superior pay-performance relation for major league baseball (MLB) that use Sabermetrics than other teams, and whether this advantage, if any, remained over time. To date, very few academic papers have attempted to examine this directly. Baseball is appropriate for the study of these effects because it lends itself to the ideas of statistical analysis since it is less team-oriented than other major league sports (e.g., the NFL, NBA, or the NHL) and basically has two-player interaction (a pitcher confronting a batter) vs. other major league sports where there is a greater intra-team interaction, in turn, making the statistical analysis much more complex. Another reason that makes MLB a natural candidate for studying the pay-performance relationship is that, as opposed to other major league sports, it has no salary cap2. The few academic papers on the topic (e.g., Demmink, 2010, Deli, 2013) deal mostly with specific aspects of the game and their value instead of the overall pay-performance concept. An early study that examined the payperformance relation in MLB is Scully (1974), but it did so at the individual player level, rather than team level, and, moreover, due to the period that it examines, it did not look at the effects of Moneyball. The study also complements and extends Hakes and Sauer (2006) by updating and extending their sample and by looking at GM specific effects, which are found to be important in explaining the pay-performance relation in MLB teams. Furthermore, one of the main contributions of this paper lies, as will be discussed next, in its linkage to adaptive market efficiency and the mispricing anomalies literature, which was not addressed at all by Hakes and Sauer (2006).
Lo (2004) proposed the theory of adaptive market efficiency in capital markets. Adaptive efficiency, which is consistent with the ideas of Grossman and Stiglitz (1980), argues that markets efficiency is dynamic and context-dependent. Furthermore, this notion reconciles market efficiency with documented behavioral biases by investors, as suggested by behavioural finance (Thaler, 1993). Consequently, one can argue that the studies of Schwert (2003), Green, Hand and Soliman (2011), Radhakrishnan and Wu (2014) and Mohanram (2014), who report on the disappearance over time of mispricing anomalies, are essentially an application of the notion of adaptive market efficiency. As such, this study provides additional evidence on the question of whether markets are indeed adaptively efficient. Using major league baseball as the market provides a relatively simple way to measure intrinsic value and to test whether markets quickly adapt to valuation anomalies. Specifically, this study looks at is whether there was indeed, as Moneyball (2003) claims, the use of Sabermetrics has provided an unfair advantage and whether it vanished after 2003 (the release of the book).
The first research question addressed in this study is whether there were any Moneyball effects at the team level using positive analysis of pay-performance efficiencies and whether such anomalies, if any, disappear, as the adaptive market efficiency theory predicts, after the release of the book. The results show that very few teams showed a statistically significant (or even a mildly significant) superior pay-performance relation over time and, moreover, any such effects, consistent with the ideas of adaptive market efficiency, disappeared after the book release.
The second research question explored in this study is whether the pay-performance relationship is better for teams that are perceived to be playing Moneyball than other teams. The results of this analysis show that perceptions on ‘Moneyball’ teams have very little to do with actual team performances.
The third research question in this study is whether the Moneyball effects should be examined at the GM level, rather than the team level, and whether such effects, if any, disappeared, as the adaptive market efficiency theory predicts, after the book release. This question was examined by looking at Sabermetric GMs as a group, and as individuals, compared to the rest of MLB. Individual GM effects are trickier than team effects as GMs often move between teams. One result is that all GMs, whether Sabermetric or not, show a significant and positive pay-performance relationship, something which should not be taken for granted. Another interesting finding is that the hero of Moneyball (2003), Billy Beane, had a significant advantage over non-Sabermetric GMs at one time but it disappeared once the book was released. Interestingly, Billy Beane’s mentor, Sandy Alderson, who came up with the idea to use Sabermetric analysis, was just at par with the rest of MLB before 2002 but did significantly worse than it afterwards. Another interesting result is that Sabermetric GMs that were hired after 2002 (Andrew Friedman, Keith Woolner and Theo Epstein) have a statistically significant better pay-performance than all other baseball GMs. Overall, the results of this analysis provided interesting insights well beyond the team level analysis and support the ideas of adaptive market efficiency.
See full PDF below.