### Abstract

This paper explores the playoff bonus system used by Major League Baseball. The unique incentive structure used is positively related to organizational success, and is studied using a Grounded Theory methodology. Exploration and analyses found that World Series winning teams prospectively distributed ten percent (10%) more shares than losing teams and repeat winners distributed still more. By developing an incentive structure as an intermediate step in the performance environment and linking the bonus’s value to future performance, decision makers may be able to positively impact organizational outcomes – particularly over repeated periods.

**Key words:** Major League Baseball, bonus system, payment, incentives

### Introduction

In October of 2007, the Colorado Rockies baseball team made the news for two reasons. First, they won 20 games in the last month of the season to wipe out a six game deficit and make the playoffs. The team became media favorites as announcers told and retold the story of how a franchise with one of the lowest payrolls in the game eventually became World Series champions. Moreover, reporters lauded the fact that success came not through the actions of one or two superstars, but by working together. This second reason for the media’s attention was the Rockies players’ decision to award a full playoff share to the family of Mike Coolbaugh.

Mike Coolbaugh was a minor league coach in the Rockies organization whose tragic and untimely death occurred on July 22, 2007 when he was struck by a foul ball during a Tulsa Drillers baseball game. At the time, he was working as a first base coach for the Drillers: a Double-A affiliate of the Colorado Rockies. The value of the playoff share eventually grew to $233,505.18 and earned the team a ‘Sportsman of the Year Award’ nomination from Sports Illustrated (38).

The purpose of this paper is to explore and explain the historical dynamics of the Major League Baseball’s (MLB’s) playoff bonus distribution system. That high performing teams’ members would be willing to sacrifice some of their own potential earnings to support their organizations’ staff members is, on the face of it, contrary to the ‘self-interests explanations’ (SIE; 14) commonly used in management, economic, and sociology theories’ literatures. Instead, it is more akin to a transformational leadership style commensurate with higher levels of organizational identification (30).

A Grounded Theory approach is used and a three-phase of the research methodology outlined by Schollhammer (43) – a relationship between share distribution and outcomes were hypothesized, an assessment of available data in the historic record was conducted, and the outcomes were evaluated based on current theories – are described. Both working managers and academic researchers can use the approach used herein to study the organizational phenomenon that commonly arise in Sport.

#### New Contribution

The paper makes two contributions managers can use to study and improve their organization’s performance. First, we present the MLB system as a case for exploring alternative frameworks for financial incentive structures. In particular, the roles of players’ decisions, rather than managers’ contributions (44), in influencing organizational success. Second, we analyze historic data, testing hypotheses developed using the grounded theory approach, to assess how patterns in share distribution practices are related to organizational performance outcomes.

For management researchers, MLB has offered a fertile empirical test bed. With the availability of data sets sometimes spanning over 100 years, baseball provides a natural experiment through which organizational behavior can be studied. In this case, we analyze the bonus system framework that can be explored in relationship to other more widely used approaches using controlled experimental designs that rely on the availability of data (44). Once completed, such research can provide decision-makers the ability to engage in a strategic process of setting appropriate performance indicators similar to those used in the “Moneyball” approach to player selection (49).

The background presented here follows an inductive/deductive processes used to study phenomenon where event timing is a critical element in the outcomes of interest. First, a brief description of the analytic strategy is given. Second, the historical context of MLB’s playoff bonus system is described. Next, data drawn from the MLB’s records is analyzed. Third, a theory is generated to explain the bonus system employed in baseball. Finally, we discuss the implications of timing in the development of incentive systems.

#### Background

Suddaby (48) both defined and outlined the use of Grounded Theory. Grounded Theory is as an approach positions phenomena as embedded in context that provides meaning that is often missing in traditional research. In other words, in the traditional model a researcher builds a hypothesis, constructs or identifies sources of data to test that hypothesis, and then tests that data to either support or reject the proposition developed. Suddaby argued that to do so ignores the perspective of the “actors” whose behaviors the data presupposes to study. Without exploring the intent of the actors, findings may be misrepresentations of the data. Further, to ignore the perspective of the actors is to ignore potential new theoretical lines of inquiry that emerge from their explanations.

The use of Grounded Theory in any context should then be defended as appropriate. Grounded Theory is most suited to efforts to understand interdependent processes, one where actors engage in an inter-subjective experience (48). The methodology is more effective when it considers extended time periods and when the relationships observed in the initial instance of the phenomenon are found to exist in other times or settings (16). Finally, the use of Grounded Theory is focused on how subjective experiences can be abstracted into theoretical statements about causal relations between actors.

Finally, it should be noted that individuals engaged in Grounded Theory research have an especial duty to explain the qualitative motivations in the research. As Grounded Theory should begin with a discussion of the data and then move to an exploration of the qualitative factors that directed inquiry, such an approach is often subjugated to the needs for traditional presentation forms for the sake of clarity – involving the presentation of a theoretical overview. In this paper, a three-step approach to exploring the phenomenon’s key elements and their relationships is employed, based on the strategies for sensemaking proposed by Langley (29). Collectively, the three phases of research used to understand the observed phenomenon – Describe, Analyze and Theorize – are referred to as the ‘DAT’ methodology. This method is particularly useful when data is temporally embedded in events rather than simple variables.

In the first phase, Define, a common narrative is developed from the process data based on unique instances of the phenomenon. The Define phase seeks to outline the context in sufficient detail from the actors’ perspective. Such an approach is requisite to assist in the narrowing of the theoretical frame for the study. Put another way, while many explanations are possible, they can be further limited by the context of the actors in situ.

The second phase, Analyze, focuses on the construction and execution of an analytic approach through which data on the dynamics identified in the Define phase are both gathered and processed in a manner that allows for simultaneous classification, to reduce complexity, and hypothesis driven statistical examination. These first two approaches are used to envelop and specify the phenomenon’s characteristics from both ends of the inductive research spectrum.

The final phase provides for the opportunity for Theory building based on an understanding of context and an assessment of data on the phenomenon being studied. Here, an alternative approach to sensemaking is used where different theoretical perspectives are applied in a deductive manner to determine how the phenomenon is best explained. The remainder of the paper follows the DAT format for investigating the incentive system used by MLB.

#### Phase 1: Describing the Phenomenon with Narrative

Groves argues that team dynamics are at play when “the decision makers base their decision choices on different information, yet are motivated by a common goal (19).” Further, Groves also notes that employee behavior may be more accurately analyzed in terms of the compensation they receive from the organizations for their participation in it. There are three features of the MLB Playoff bonus system that depart from most incentive and reward programs used by professional firms – the process of valuation, control and criteria.

Valuation addresses the resources set aside for the purposes of creating an incentive structure. In the broader business context, valuation is often a function of organizational history. The amount of resources set aside for the purpose of incentive distribution is not uniform as organizations specify the parameters on a case-by-case basis. Valuation can be understood not only in terms of how much, but also in terms of when the establishment of the specific value takes place. Further, valuation speaks not only to the pool of available resources for such incentives, but whether that pools is absolute or relative to firm performance.

For instance, an organization may have developed a practice of setting aside 10 percent of pre-tax profits for the purpose of distribution by management. For instance, an organization may decide that it sets aside 25 percent of any profit for distribution whereas another might only set aside that 25 percent if the organization exceeds last year’s profit levels. Additionally, organizations may simply allocate a fixed sum once a specific objective or subjective threshold is accomplished, for example, when the organization’s profits exceed 10 million dollars, one million is put into the profit sharing pool.

The second feature of bonus financial incentive systems is control. In this context, control speaks to the authority that allocates the bonus dollars. In most cases it is facilitated through the management structure. Control may be facilitated both directly and indirectly. Direct control is exercised when management evaluates individual performance to determine the degree to which an individual should be rewarded using the incentive pool. Indirect control is exercised when the organization codifies in contracts that specify the conditions under which an individual has a right to access the pool. For the most part, this process is considered a management responsibility.

Finally, the issue of criteria addresses the constraints and rules associated with participation in the pool. Organizations may, for instance, develop a pool for the sales staff. Such a pool is often aligned with compensation schemes to recognize high performers and represent constraints on the criteria for participation. Other organizations may create committees whose task it is to determine which employees meet the criteria for inclusion in the pool.

##### The Players’ Bonus in MLB

The bonus incentive model in MLB has an unusual scheme in that the system of valuation and control are far from the norm. Additionally, the criteria for distribution diverge from many bonus systems. To understand the formalization of this incentive structure, it is necessary to describe how the value of a share is calculated. We shall do so here in the order in which events happen so as to make specific points about the nature of what is both known and unknown at that point in the process.

##### Identifying the phenomenon’s key constructs

In 1903 the National Agreement united the American and National Leagues. This agreement created the World Series and stipulated that a portion of the playoff series gate receipts go directly to the participating teams’ players. At the time, prior to major radio or television contracts, the gate represented a significant portion of most teams’ total revenue. Therefore, the bonus system was a major revenue sharing scheme well ahead of its time (40). Further, the bonus system’s design and longevity are both rarities in the modern industrial era.

MLB has a playoff reward program and players on winning teams can earn substantial bonuses. Under the MLB playoff structure implemented in 1995 (the current system allows for three divisional champions and one ‘wild-card’ in each league), the players pool (P*) is comprised of 60 percent of the gate receipts from the first three rounds of the Division Series and 60 percent of the gate receipts from the first four rounds from the League Championship Series and World Series across the entire playoff contest. By only using the gate from the initial games, the formula presented in equation 1 does not incentivize the extension of the series. These receipts are divided as team shares among 12 clubs using a performance multiplier (M^T) that escalates as the team progresses further in the playoffs, described in Table 1. The team share (P^T) is then the fraction of the players’ pool that is available to divide among players.

**equation goes here**

This dynamic represents the first divergence from traditional bonus models. In MLB, the valuation process is prescribed but disjointed. Specifically, equation 1 specifies the value of the MLB players’ pool (P*), making it clear which teams are eligible for participation at the front, and allowing teams to estimate, based on historic trends the value associated with different performance outcomes. Individual team’s payouts escalate as they progress further in the playoffs with the World Series Champion getting the largest sum as prescribed by both Table 1 and equation 2. In 2006, the Players’ pool was $55.06 million and the St. Louis Cardinals took the lion’s share of $20.2 million (36 percent) by winning the World Series. That equates to $300,000 per player on the World Series winning team.5 (See Table 1)

At this point, a second divergence occurs from traditional models along the issue of control. In most cases the control of bonus dollars resides in management, however in the case of MLB, only players may allocate a portion of the players pool and they may do so in any manner they wish. This dynamic is significant in two ways. First, management has no ability to control the distribution of shares. Second, the distribution of shares occurs before the beginning of play in the post season. In other words, the employees create a system to allocate future performance-based earnings using a fractional system based on the concept of a “share” (V_s) as described in equation 3.

_Sporting News_ writer Todd Jones, a former MLB pitcher, notes that only players who have been on the roster the entire season are given a say in the allocation of shares. Players taken off the rosters for an extended period of time are generally not eligible to vote, and all players eligible to vote are all given a single share. The meeting then does not focus on the shares allocated to those in the room, rather it focuses on the shares to be allocated to those not in the room, and several players note that these meetings can become quite heated. Players often create “fractional shares” or simple “cash awards” (A) to recognize the work of auxiliary personnel. Because the allocation allotted to the team is fixed, every additional share authorized reduces the relative value of each player’s share. Additional discussion about the meaning of shares will be addressed in the next section. (See Table 2)

In summary, the MLB system of playoff shares provides a unique perspective into a system with a long history that is framed in such a way that empowers plays by giving them both control and allowing the players to determine the criteria. These decisions have an effect on valuation that is related to both performance and the share distribution decisions of the organization’s players. Each additional share distributed by the players reduces the value of the share each rostered player receives.

##### The meaning of a share

The authors were unable to secure interviews through the Major League Baseball Players Association. As a result, the need to explore what these shares meant to the players was critical to our DAT approach. Using Lexus-Nexus, the authors retrieved all news stories available on the service since 1980 using the search terms “playoff shares” and “baseball.” These 691 articles were then reviewed for the purpose of identifying what a share meant to the players. From those articles, duplicates were removed and articles outside of scope, as in the case of stories focusing solely on the value of the share as determined after the World Series or commentary provided by the writer. The remaining data was further segmented into direct quotes and evidence-based commentary whose origin was the players and their share deliberations.

Clearly players would forgo the value of a playoff share for the World Series title, but the underlying message was that the bonus is considered to be very significant and given in recognition of time served on the roster rather than performance of a specific player. For instance Astros General Manager Tim Purpura was quoted as saying, “The players that are on the roster as of June 1 typically get a full share… any players who were (on the roster) part of the season, they get voted on (by the players) (32).” A similar sentiment was made by the players of the Boston Red Sox who gave partial shares or cash rewards to anyone who wore the uniform, as well as a number of front office staff (24).

However, because the players make the decision according to their own preferences, time is no guarantee of a share. The players of the Toronto Blue Jays awarded former Chiefs outfielder Rob Ducey a small fraction of a share (16%), which was far less than others like Tom Lawless (59%), Rob MacDonald (25%), Mike Flanagan (20%) and Tom Quinlan (17%) even though Ducey was on the team longer than the other players (33).

Players have seen the exercise of their rights over the control of the playoff shares as a form of power. In 2002 the hitters in the San Diego Padres were so impressed by the work of one of the minor league coaches that they were able to “finagle” him onto the major league roster. He quit his job as a minor league coach but replaced it not only with a six-figure salary but a full playoff share after the Padres reached the World Series (7). Additionally, in 1996 the playoff shares were at times withheld from those crossed the picket lines. Mike Busch failed to receive a playoff share from the rostered players, despite a direct plea by then manager Tommy LaSorda. “What matters is the name on the front of your shirts, not the name on the back,” (17).

The dynamics inside these team meetings were noted as potentially contentious. Baltimore manager Johnny Oates, a former player, commented that he never liked team meetings but he did enjoy splitting up the playoff shares at the end of the year (22). A day after clinching the American League Western Division title, California Angels’s Rod Carew stormed out of a Sunday meeting in which the Angels decided how to divide playoff shares and World Series money among themselves. Carew would say only, “Money does strange things to people.” Carew, a native of Panama, apparently interpreted the decision to offer two Latino pitchers a half-share as discriminatory and angrily left the meeting. Reggie Jackson and Doug DeCinces were reported to have tried to restore the peace, but were unsuccessful (45).

Interestingly, the 1991 story of Dave Pavlas, Matt Howard and Dale Polley stood out in that it was the single instance where management, in this case George Steinbrenner of the New York Yankees, gave individuals each a check for $25,000 and a 1996 World Series ring, when the players voted to give them nothing. Steinbrenner also traded the primary force behind the decision to withhold the share, Jim Leyritz, the following day (4, 42).

Another ongoing commentary by the players is the financial value of a share for younger players. A few players identified the impact of a share on people who worked to create a winning team. In 1991 Norm Charlton of the Cincinnati Reds said, “You get two things for winning the World Series – a ring and our playoff shares.” He pointed out that for many of the younger players, the value of a share was the only recognition they received for winning a division or league championship. Further, he argued that the amount given was substantial relative to the salaries earned by some rookies (6).

Additionally, in organizations that traded players midseason, the players used shares to recognize these individuals’ contributions, even after having left to work for another competing organization. In 2004, the Boston Red Sox voted Nomar Garciaparra to receive a full share even though he was traded to the Chicago Cubs (18). When infielder Jeff Cirillo was released from the Minnesota Twins to the Arizona Diamondbacks, his final message was, “Good luck, and don’t forget about the old guy in the playoff share meeting” (34). This was again the case when the Florida Marlins players voted to give Pat Rapp of the San Francisco Giants a full playoff share for his contribution to the team before his trade (1).

In 2009, Major League Baseball (MLB) reported that the players for the Milwaukee Brewers offered former manager, Ned Yost, one of the Brewers 48 full postseason shares. The firing of Ned Yost marked the first time in major-league history — except the strike-split 1981 season — that a manager was fired in August or later with his team in playoff position. In their press release, MLB quoted an unnamed player who said “There are unwritten rules about how to do things, and it was the right thing to do. If you’re there more than 50 percent of the season, you’re pretty much getting a full share.” That same year, Nick Adenhart was killed in a car accident hours after starting as a pitcher for the Angels franchise. The Angels players voted to provide the late pitcher’s family a full share.

In summary, we found that players expressed a clear understanding of the diminishing return of the effect of the issuance of additional shares, that issues of equity, fairness and a sense of team spirit was a consistent theme, and that shares were also an expression of power by the players that could be exercised to make individuals feel either included or excluded. The series of events and decisions made by the Colorado Rockies described at the paper’s outset served as the impetus for the study at hand. However, it is clear that players use shares to recognize efforts of non-players as well as remediate perceived slights by management. While other teams have engaged in similar behaviors, none received the widespread coverage that the Rockies’ gesture did, in no small part because of their successful run to the World Series. The authors speculated that differences in share distribution patterns between teams might predict playoff series’ outcomes.

##### Shares and their relationship to outcomes

Emergent in this discussion is the idea that playoff shares can serve as an incentive for teams. Further, the structure involves an intricate timing that alters the manner in which the share itself is perceived. Table 1 speaks to the steps associated with the valuation, control and criteria used in share calculations. An initial component of evaluation is assigned in the securing of a place in the playoffs. At that point, team members know they are eligible for a share, however the specific value of that share remains unknown as it is based on future team performance. Rather than making the decision after the fact, the meeting that sets the criteria occurs place before any of the playoff series are played. This means that eligibility for a share is established before the value of that share in absolute terms is known. What is known is the value in relative terms.

The baseball system creates certainty around a payout whose value is unknown. While this may be a small point, it is a significant one. The implication of increased information provides short-term certainty other recognition systems often fail to provide. In 2008, 21 percent of all MLB players were paid less than $400,000, and the median payroll for a player was $1 million dollars. A playoff share will constitute a significant bonus for most of MLB players. For non-uniformed players, a share can triple or quadruple an annual salary.

The playoff shares system is distributed based on a criterion of past performance, where the control resides at the professional level to recognize peer performance, and creates a value that is based on future performance. Therefore, the major hypotheses are designed to explore share distribution and winning the World Series.

Hypothesis 1: Winners of the World Series will distribute significantly more playoff shares than teams that lose the World Series.

The expansion of the playoffs in 1995 provides a second period with potentially greater variability in the share distribution–playoff outcome relationship. With more teams in the playoffs over more rounds, the inference that distributing more shares is related to performance can be tested more rigorously. Therefore, the additional hypothesis is:

Hypothesis 2: Since 1995, Winners of the World Series will distribute significantly more shares than teams that lose the World Series.

It is posited that winning teams will distribute more shares. It stands to reason that players across the league will either implicitly or explicitly internalize this information. As a result, the number of shares distributed on average should increase over time. Therefore, another hypothesis arises:

Hypothesis 3: Baseball teams making the playoffs will distribute more shares over time.

It is possible to further explore the learning phenomenon by focusing on the teams that repeatedly make the playoffs. Having the firsthand experience assessing the playoff bonus – performance outcome relationship, teams that have won previously will seek to ensure their competitive advantage by increasing their distribution disproportionately. Therefore, the following hypothesis is made:

Hypothesis 4: World Series winning teams will distribute disproportionately more shares over time.

Based on this series of hypotheses, the Analytic Phase of the DAT process was undertaken.

### Methods

Baseball has been the subject of many empirical studies. One reason for the large number of studies is the availability of reliable and valid data on player demographics, performance outcomes, pay rates and incentives. Rottenberg (41) explored the market factors of the baseball labor market . A second reason for the interest in baseball’s statistics is the stability of the sport over time allowing players from various eras to be compared with some fidelity – the ‘dead ball’ and ‘steroid’ eras not withstanding.

With respect to compensation, Kahn (26) explored the relationship between managerial quality and salary. However, most of the studies to date have generally focused on baseball as a market phenomenon (5, 12, 13, 28, 35, 39). There have, to date, been few, if any, studies that have looked at the playoff shares incentive structure as described herein.

Using data publicly available on the number and value of World Series shares issued going back to 1903, the researchers tested the hypothesis that teams that were more egalitarian in the distribution of shares to non-uniformed players would be more successful. Because, within a year the size of teams should be relatively comparable, a simple different t-test was employed on the number of shares winning teams distributed prior to the start of the playoffs and the number of shares that the losing teams distributed. As only World Series shares were available from MLB, this hypothesis could only focus on winners and losers of the final series, and not on any of the championship and pennant races that allowed a team to compete in the final series. Based on the meaning behind shares, we argue that the differences within a single season can be attributed to shares allocated to players that played an incomplete season and auxiliary staff.

A second data gathering exercise focused on the playoffs that occurred after the introduction of the wild card in 1995. Wild card teams are ineligible for shares unless they win their division. Because this change to the system created new dynamics, we focused on the almost 30 years of data that emerged since that time. Again, data on shares were gathered from publically available data sources, including but not limited to MLB, _U.S. News and World Report_, and the Associated Press.

Analysis of the data was conducted using SPSS 17. Hypotheses 1-4 were tested using a t-test. This research involved the exclusive use of secondary data and as such was exempt research study per university institutional research board regulations.

### Results And Discussion

Phase 2: Analyses of the Available Data

Analyses of the relationship between World Series performance and playoff share distribution supports the assertion that winning teams are more egalitarian in their distribution of shares, authorizing more than their losing competition. Further, the dynamics of share distribution have been altered as the rules of the game have changed, most significantly with the initiation of free agency and the alteration to the playoff series’ format following expansions (23). Table 2 outlines the pattern of share distribution by franchise. (See Table 3)

Winning teams have historically been more generous (Hypotheses 1 is supported)

Over the 102 years of the World Series through 2006, we first calculated the difference in shares offered between winning and losing teams. Over that time period, winners distributed 88.93 more shares than losers, resulting in an average differential distribution of 0.872 shares for winning teams over losing teams (s = 3.65). The data supports the proposition that, within each year, winning teams offered more shares than losing teams (t = 2.407, p < 0.001). (See Figure 1)

Another way to look at the problem is to study the winner’s premium and compare it with the winner’s counter-premium. The winner’s premium is the number of shares issued by the winning team above what the losing team offered. For instance, when the winning and losing teams both issue 10 shares, an identical number of shares, neither is offered a premium. However, when the winning team issues 15 shares and the losing team issues 10 shares, the winner’s premium is said to be 5. The converse, the winner’s counter-premium exists when the winning team offers less shares that the losing team. The use of the term premium identifies the magnitude of the difference between winning teams that more distributive and those that are less relative to their competition. Figure 1 presents the winner’s premium/counter-premium in the form of a control chart where the y-axis is keyed to the standard deviation. Most statisticians would argue that the process only moves out of control in recent years, interestingly enough in a manner that coincides with the inauguration of the wild card rules of 1995. There is an additional anomaly that coincides with the expansion of MLB in 1965. (See Figure 2)

In the historic case, the winners’ premium has been 3.26 shares, based on 195.8 shares over 60 instances where the winner of the World Series is the team that has offered more shares. The winner’s counter premium has been 2.89, based on 102.2 shares over 38 instances where the winner of the World Series is the team that has offered more shares. Further, winners were 1.6 times likelier to have offered more shares than losers. When more generous teams won, they generally authorized 3.26 shares more than their losing competition. When the less generous team won, the difference between the winners and losers was actually smaller (2.7 shares).

Additionally, it should be noted that repeated appearance in the playoffs is related to salary. An analysis of shares authorized correlates to the payroll of the team (r2 = 0.26). However, it should be noted that making it to the playoffs multiple times does not necessarily correlate with higher salaries. A similar analysis was conducted by adding the frequency the team appeared in the playoffs for that same time period. In that case, the quality of the correlation dropped (r²_adj=0.22).

Therefore, a minimum share distribution spread may exist before differential performance is realized. Therefore, we find that Hypothesis 1 is supported in both analyses.

Additionally, we focused on the distribution parameters since the baseball playoff series was restructured in 1995 through the most recent data available in 2008. The results in this case were not significant. Over that time period, winners distributed 0.74 more shares than losers, resulting in a negligible differential distribution of 0.05 shares for winning teams over losing teams (s = 5.53). In those same years, the winning team offered more shares as often as the losing team – six of twelve times the World Series were won by the organization offering the winner’s premium. However, further analysis of the data shows that the winner’s premium of these years was 6.05 and the winner’s counter-premium was 4.33. This provides some evidence that effect is relatively weak and that limiting the time to twelve years may have undermined the power of the analysis. Therefore, while Hypothesis 2 is not supported, there is counter evidence that may need further study.

Overall, in the historical model, winners were 1.6 times likelier to have offered more shares than losers. Further, when more generous teams won, they generally authorized 3.26 shares more than their losing competition. When the less generous team won, the difference between the winners and losers was actually smaller (2.7 shares). Therefore, a minimum share distribution spread may exist before differential performance is realized. This analysis limited to the World Series between the years of 1995 to 2008 found winners were as likely to have offered more shares than losers. Further, when more generous teams won, they generally authorized 6.05 shares more than their losing competition. When the less generous team won, the difference between the winners and losers was actually smaller (4.33 shares).

Teams have become more generous over time (Hypothesis 3 is supported). There has been a general trend to increase the number of playoff shares. On an annual basis, teams in the World Series have increased the average number of shares distributed annually by 0.31 (r2 = 0.78). Since 1995, the average number of shares authorized by teams in the World Series has increased 1.03 shares annually, with an Adjusted R-squared of 0.330. When that analysis is expanded to include the divisional champions for both the American and National League, the distributional structure does not change (1.10 v. 1.03 shares annually), but the explanatory factor is increased significantly (Adjusted R-squared = 0.61), suggesting that the increase is foundational and the diminished explanatory power can be attributed to the relatively small data set in the subset analysis. Therefore, Hypothesis 3 is supported in both the full sample and the period from 1995 to 2008.

Repeat winners are still more generous (Hypothesis 4 is supported). Among World Series winning teams, there has been a consistent increase in the number of shares distributed on average. For every additional World Series a team has won in the past 1995 to 2008 period, teams offered an additional half-share (r2 = 0.32). Winning teams have developed a winning formula with respect to share distribution. Further, that players learn to forgo the initial temptation for engaging in SIE behaviors at the outset is important, but not predicted in most of the theories employed by management, economic and sociology researchers. Therefore, it would be beneficial to have a set of guiding principles prior to implementing such an incentive system; hence, the need for a theoretic exploration of MLB’s bonus system.

#### Phase 3: Theoretic Assessment of MLB’s Bonus System

The narrative and empirical evidence offer a new opportunity to explore the theoretic paradigms used to explain and create bonus systems that promote organizational missions. Issues of power, equity and pro-social behavior, and performance forecasting characterize playoff share distribution decisions and their positive impact on organizational outcomes. They also constitute decision-making in an environment where the motivation can range from purely self-serving (e.g., Self-Interested Economics (SIE) explanations) to completely egalitarian. With respect to SIE motivations, Principal-Agency is the most widely used model for exploring compensation issues in organizations – Principal-Agency Theory (8). At the other end of the spectrum is Equity Theory that predicts the remuneration system will distribute shares across all participants at equal levels. The range of theories is explored from most self-interested to most egalitarian.

##### Alternative theory one – principal-agent theory

Bonus systems are often used as an extension of governance policies to promote congruence between principals’ and agents’ goals. Therefore, the topic of incentive-reward systems often arises in the economics’ literature as a search of Google Scholar using the term “bonus system theory” reveals. In particular, economics views such systems through the lens of contractual arrangements where an agent’s rationale behavior is to maximize their utility by fulfilling the contract and doing nothing more (50).

The limits and boundaries of agency theory lie in its model of human motivation. In the case of baseball, economic models based on the ‘rational’ or ‘economic man’ would suggest that the optimum distribution for the individuals (agents) acting on their own behalf would be to distribute a maximum of 40 full shares (47). Given that few if any teams go a full season without roster changes, the number of full share equivalents distributed should be less than 40 with only fractions given to individuals who have been on the roster for partial period.

The near-term financial incentives are for the empowered players to act in their own interest without consideration for non-uniformed stakeholders or players no longer on the roster. The reality is far from this. In 2006, players on each of the teams in the playoffs authorized the creation of 52.86 shares, on average. Therefore, the prima fascia evidence indicates that one of Agency Theory’s main tenets is violated by the bonus system used by MLB. Beyond the shortcomings of economic theories identified herein, other researchers have suggested that other organizational theories may have greater explanatory power (2, 50). In particular, they point to Stewardship Theory for addressing the egalitarian features of the bonus system used by MLB instead of Agency Theory (9, 10, 15, 50).

##### Alternative theory two – stewardship theory

Stewardship Theory describes a phenomenon where an agent’s goals are subordinated to a broader organizational or societal aims. Stewardship Theory assumes an individual’s decision model is ordered such that pro-organizational, collectivistic behaviors have higher utility functions than individualistic, self-serving behaviors. A steward protects and maximizes the organization’s collective wealth through superior firm performance, because, by so doing, the steward’s utility functions are maximized. Alternatively, the steward may calculate that working toward organizational, collective ends is the best trade-off between personal needs and organizational objectives. Hence, the utility gained from pro-organizational behavior is higher than the utility that can be gained through individualistic, self-serving behavior.

The main problem with Stewardship Theory, as it applies to MLB’s bonus system, is that incentive pay is antithetical to the primary model of behavior anticipated. From a Stewardship Theory perspective, how bonuses are used or distributed within organizations would have little influence on employees’ individual motivations to work hard toward organizational goals. Given players are earning a living wage, which is assured with the minimum pay agreed to through the Collective Bargaining Agreement, no other motivation than those intrinsic to the organization’s aims should be needed.

There is also objective evidence that Stewardship Theory is not applicable to MLB. In addition to the 1919 Black Sox scandal, a reference to game fixing by the Chicago White Sox, the widespread use of steroids more recently provide both specific and general examples that players are not fully vested stewards protecting the integrity of the game’s values and traditions. With respect to the latter, steroid use, not only have players cheated and skewed the on-field records, they have also lied repeatedly to cover their misdeeds, and have engaged in this behavior in what seems to be a collective mind rather than individual exception (37). The players were able to engage in this behavior for an extended period of time under the protection of their union’s collective bargaining agreement (46). While steroid use was widespread, and alleged to have been endemic to some teams, the difference between individual versus collective decisions may be explained using another pair of theories.

##### Alternative theory three – expectancy and equity theories

Psychological paradigms such as Expectancy and Equity Theories have been applied to organizational compensation questions and baseball in particular. These theories explore the positive versus negative aspects of individuals’ behaviors under pay-for-performance situations. There has been significant work in the area of equity and expectancy theories in baseball. Lord and Hohenfeld (31), Duchon and Jago (11), Hauenstein and Lord (21), and Harder (20) studied the performance of baseball players who were either free agents or had been through salary arbitration. This line of research has looked at the pay-for-performance question based on salaries, not collectively distributed bonuses. Further, a major reason for implementing a bonus system is to avoid the managerial transaction costs of micro-managing employees’ individual expectations and trying to equitably fulfill them. Indeed this is a major benefit of creating an incentive system controlled directly by the employees.

The teams’ distributions of playoff shares and performance outcomes allow for some inferences to be made with respect to Expectancy and Equity. In particular, the distribution of more shares could be equated to greater perceived equity, which in turn leads to improved performance. However, the distribution of more shares could be defined as egalitarian without any regard to merit with equal veracity. Therefore, expectancy and equity theory are problematic under the given circumstance for a manager wishing to take advantage of a program with a proven record of success in promoting organizational goals.

There are other theories that can be applied to MLB’s bonus system and other potential propositions arise from the results herein. Institutional Theory could be used to explain the increasing amount of shares being created as teams engage in a mimetic process based on the winners’ behaviors. Alternatively, human resources researchers might propose that there will be greater player retention among more egalitarian squads. Both of these lines of inquiry will require additional information that is best gained by going directly into the process.

### Conclusion

MLB’s playoff bonus system exhibits many desirable characteristics, but it does not conform to the most commonly used found in the design of such systems. It is important to reiterate the unusual features of the system. Shares have nominal value at the point they are issued, and gain in value as a team progresses through the playoff series by performing competitively. Allocations generally include some aspect of retrospective acknowledgement – shares are allocated to the people that the core players believe have worked hard to get their team to the playoffs, paid against a value defined by future performance. These people can be players, support staff, or managers. As a result, beneficiaries of shares have full information about the distribution of shares without knowledge of the value of those shares. However, because these shares are allocated prospectively, share recipients can alter behaviors in ways that maximize the value of those shares. In this respect, the system constitutes an incentive structure on a pay-for-performance basis.

The link between performance and reward distribution is one mechanism that management scholars have long advocated as an important motivational or demotivational tool (25). In particular, the use of group incentive plans are recommended when the group is small, team members are engaged in the same kind of work, the payment is clearly linked to the performance, the output depends on the workers, the operational cycle is fairly short, the bonus is substantial relative to standard pay, and there is an atmosphere of mutual trust between management and workers (3, 36). The MLB playoff bonus system is such an incentive plan and has been in existence since 1903. Further, it has an unusual feature in that the players determine playoff bonus share distribution, not only amongst themselves, but also to managers, trainers and other staff members in the organization.

The data from this analysis supports a bonus system that creates incentives for individuals that linked performance to outcomes in ways determined by the principal actors, but not actively managed by owners may support performance in substantive ways. Further, to the extent that such efforts are seen as egalitarian and recognizes the broader contributions of individuals on the margin, such distributions seem to support overall organizational success. Additionally, and maybe more significantly, this system defines the reward structures prospectively. While most players would consider winning the World Series the ideal outcome, this system creates process incentives in addition to outcome incentives. This represents a significantly different method for incentivizing outcomes. Most bonus systems are designed and allocated after the fact with little to no information given to the recipients of the reward.

MLB players have elected to distribute progressively more playoff bonus shares as time progresses. Therefore, there is some form of institutional knowledge accruing. Further, that teams with more experience distribute still more shares than their competitors is an indication that the feedback mechanism functions in a positive manner (27). Collectively, the hypotheses demonstrate the efficacy of MLB’s bonus system.

### Applications In Sport

The distribution of annual bonuses by professionals to themselves and other organizational members is a common feature of law practices, investment banking firms, consulting practices and some other closely held businesses. Bass suggests that group incentive plans are recommended under specific instances. In practice, these distributions are typically allocated retrospectively when the bonus pool’s amount is known using performance standards determined ex post facto. Generally, the professionals (e.g., partners) determining the distribution are the firm’s management.

Studying the distribution of World Series playoff shares provides a unique glimpse into an incentive scheme. The MLB Playoff Shares system uses a valuation system where non-managerial professionals determined share allocation a priori and the bonus amount is not assured, and the performance measure – winning the World Series – is absolute. The baseball model is a system where a bonus is structurally defined, controlled by the agents of the organization with the greatest direct ability to impact outcomes, and is retrospectively allocated but prospectively incentivizing. The incentive structure created is therefore both prospective and retrospective, and moreover, the timeline presents a short horizon to gain, addressing the common pool resource problem of long-term versus short-term feedback. Therefore, if the incentive scheme is an efficacious one, it may warrant broader adoption amongst other professionally led organizations seeking to improve their performance.

### Tables

#### Table 1
Performance Multiplier

Title	Number	Performance Multiplier
World Series Champion	1	0.36
League Champion (World Series Loser)	1	0.24
League Champion Series Runners Up	2	0.12
Division Series Runners Up	4	0.03
Second Place Finishers (Non-Wild Card Clubs)	4	0.01

#### Table 2
MLB’s Bonus System’s Design

Construct	Definition
Between Team Pool Determination	Formula-driven and invariant from year-to-year. System rewards teams, not individuals. Team’s payout varies based on how they finish in the playoffs.
Allocation Timing	Share distribution is determined prior to share valuation.
Within Team (Work Group) Control	Front-line employees – players – determine their own and staff members’ shares in an informal process designed to incentivize team aims.
Allocation Criteria	Measurement models unknown. Only players that were on the roster the entire season have control of the bonus allocation. Rosters are expanded for the post season; therefore, some aspect of tenure likely plays a role. In addition, the large wage disparity between ‘star’ players and other organizational members may result in a premium on altruistic distribution schemes.
Valuation	A share’s realized value is not known until the team’s season ends. Teams that perform better vis-à-vis other units receive progressively greater rewards.

#### Table 3
Share distribution by winning franchise, sorted by Average Share Premium (1903 to 2008)

World Series Championship Franchise	Number of Wins	Average Share Premium
Chicago White Sox	3	-5.83
Florida Marlins	2	-5.55
Minnesota Twins	2	-2.00
Atlanta/Milwaukee/Boston Braves	3	-1.87
Cleveland Indians	2	-1.20
Toronto Blue Jays	2	-1.10
St. Louis Cardinals	10	-0.88
Washington Nationals	1	-0.80
SF/NY Giants	5	-0.44
LA/Brooklyn Dodgers	6	0.37
Detroit Tigers	4	0.70
Oakland/Philadelphia Athletics	9	0.76
Anaheim Angels	1	1.20
Chicago Cubs	2	1.20
Cincinnati Reds	5	1.40
Pittsburgh Pirates	5	1.50
Baltimore Orioles	3	1.53
Boston Red Sox	5	1.95
Kansas City Royals	1	2.70
New York Yankees	26	2.73
Arizona Diamondbacks	1	3.00
Boston Americans	1	4.00
Philadelphia Phillies	1	4.45
New York Mets	2	6.20

### Figures

#### Figure 1
Premium/Counter-premium Control Chart (World Series: 1903 – 2008, y-units are 1)
”)

#### Figure 2
Share distribution of Playoff Teams by Frequency of Participation (1995 through 2008)
”)

### References

1. Armstrong, J. (1997, October 15). Jock-Ularity, The Denver Post, p. D2.
2. Baker, G. P., Jensen, M. C., & Murphy, K. J. (1988). Compensation and incentives: Practice versus theory. The Journal of Finance, 43(3), 593-616.
3. Bass, B. M. (1965). Organizational Psychology. Boston: Allyn and Bacon.
4. Borzi, P. (1997, May 13). Yankees Notes, The Star-Ledger.
5. Bovarsson, O. B., & King, B. (1998). The Value of Arbitration Right in Major League Baseball: Implications for Salaries and Discrimination. Quarterly Journal of Business and Economics, 37, 65-80.
6. Chass, M. (1991, March 17). Piniella’s formula: Will and Intensity, Sports Desk, p. 1.
7. Close… Closer… even Closer; See a different game; So you think you know baseball ? Then what makes a good batting practice pitcher? Where the heck did the fungo originate? How much meal money are major leaguers given? Baseball’s big picture is defined by the snapshots that provide the game’s intricate details. (2002, July 29, 2002). Editorial, The Sporting News.
8. Conyon, M. J. (2006). Executive compensation and incentives. The Academy of Management Perspectives, 20(1), 25.
9. Davis, J. H., Schoorman, F. D., & Donaldson, L. (1997). Toward a stewardship theory of management. Academy of Management Review, 22(1), 20-47.
10. Donaldson, L. (1990). A rational basis for criticisms of organizational economics – a reply to barney. Academy of Management Review, 15(3), 394-401.
11. Duchon, D., & Jago, A. G. (1981). Equity and the performance of major league baseball players: An extension of Lord and Hohenfeld. Journal of Applied Psychology, 66(6), 728-732.
12. Ferguson, D. G., Jones, J. C. H., & Stewart, K. G. (2000). Competition within a Cartel: League Conduct and Team Conduct in the Market for Baseball Player Services Review of Economics and Statistics, 82, 255-265.
13. Fizel, J. (1996). Bias in Salary Arbitration: The Case of Major League Baseball. Applied Economics, 28, 255-265.
14. Folger, R., & Salvador, R. (2008). Is management theory too “Self-ish”. Journal of Management, 34(6), 1127-1151.
15. Ghoshal, S. (2005). Bad management theories are destroying good management practices. Academy of Management Journal Learning and Education, 4(1), 75-91.
16. Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine Publishing Company.
17. Gonzales, M. (1996, February 8). Bourgeois is just Asking for a Chance Talent Aside, Bourgeois Hopes to be Welcomed, The Chicago Tribune, p. 1D.
18. Goold, D. (2009, November 30). Cardinal Beat, St. Louis Post-Dispatch.
19. Groves, T. (1973). Incentives in Teams. Econometrica, 41(4), 617-631.
20. Harder, J. W. (1992). Play for Pay: Effects of Inequity in a Pay-for-Performance Context. Administrative Science Quarterly, 37(2), 321-335.
21. Hauenstein, N. M. A., & Lord, R. G. (1989). The Effects of Final-Offer Arbitration on the Performance of Major League Baseball Players: A Test of Equity Theory. Human Performance, 2(3), 147.
22. Hohlfeld, N. (1993, May 2). Bonds on his success: It’s only April, dude, Houston Chronicle, p. 7.
23. Holbrook, M. B. (2003). Adventures in Complexity: Write An Essay on Dynamic Open Complex Adaptive Systems, Butterfly Effects, Self-Organizing Order, Coevolution, the Ecological Perspective, Fitness Landscapes, Market Spaces, Emergent Beauty at the Edge of Chaos, and All That Jazz. Academy of Marketing Science Review, 6.
24. Horrigan, J. (2005, September 29). Tito: Ortiz ploy a sacrifice for team, The Boston Herald, p. 98.
25. Kanter, R. M. (1987). Pay: From status to contributions: Some organizational implications. Personnel, 64, 25.
26. Khan, L. M. (1993). Managerial Quality, Team Success, and Individual Player Performance in Major League Baseball. Industrial and Labor Relations Review, 46(3), 531-547.
27. King, C., & Delurey, D. (2005). Twins, Siblings or Cousins? Analyzing the Conservation Effects of Demand Response Programs. Public Utilities Fortnightly, 143, 54-61.
28. Krautman, A. C. (1999). What’s Wrong With Scully-Estimates of a Player’s Marginal Revenue Product? Economic Inquiry, 37, 369-381.
29. Langley, A. (1999). Strategies for theorizing from process data. Academy of Management. The Academy of Management Review, 24(4), 691.
30. Lim, J., & Cromartie, F. (2001). Transformational leadership, organizational culture and organizational effectiveness in sport organizations. Sport Journal, 4(2).
31. Lord, R. G., & Hohenfeld, J. A. (1979). Longitudinal field assessment of equity effects on the performance of major league baseball players. Journal of Applied Psychology, 64(1), 19-26.
32. Mctaggart, B. (2005, November 23). Astros reaping Series reward; Most players get playoff share of nearly $192,000, The Houston Chronicle, pp. Sports, 1.
33. Michael, M. (1990, October 25). Chiefs Scour Bushes for Fruit of the Farms, Post-Standard, p. E1.
34. Miller, P. (2007, August 4). Departing Cirillo goes to bat for fielding: Diamondbacks claim second baseman, St, Phil Miller Pioneer Press.
35. Miller, P. A. (2000). A Theoretical and Empirical Comparison of Free Agent and Arbitration-Eligible Salaries Negotiated in Major League Baseball. Southern Economic Journal, 67, 87-104.
36. Mirabile, M. (2007). The NFL rookie cap: An empirical analysis of one of the NFL’s most closely guarded secrets. Sport Journal, 10(3).
37. Mitchell, G. (2007). Report to the Commissioner of Baseball of an Independent Investigation into the Illegal Use of Steroids and other Performance Enhancing Substances by Players in Major League Baseball. New York: Major League Baseball.
38. Price, S. L. (2007, Nov. 29, 2007). My Sportsman: The Colorado Rockies, Sports Illustrated. Retrieved from http://sportsillustrated.cnn.com/2007/magazine/specials/sportsman/2007/11/28/price.sportsman/index.html?section=si_latest
39. Richards, D. G., & Guell, R. C. (1998). Baseball Success and the Structure of Salaries. Applied Economic Letters, 5, 292-296.
40. Rockerbie, D. (2009). Free agent auctions and revenue sharing: A simple exposition. Journal of Sports Management, 23(1), 87-98.
41. Rottenberg, S. (1956). The Baseball Players’ Labor Market. The Journal of Political Economy, 64(3), 242-258.
42. Rozner, B. (1998, March 27). Cubs’ Hendry could be in line for major-league GM job, Chicago Daily Herald.
43. Schöllhammer, H. (1973). Strategies and methodologies in international business and comparative management research. Management International Review, 13(6), 17-32.
44. Smart, D. L., Winfree, J. A., & Wolfe, R. A. (2008). Major League Baseball managers: Do they matter? Journal of Sports Management, 22(3), 303-321.
45. Sports briefs. (1982, October 4). St. Louis Post-Dispatch.
46. Staudohar, P. D. (2005). Performance-enhancing drugs in baseball. Labor Law Journal, 56(2), 139.
47. Stroh, L. K., Brett, J. M., Baumann, J. P., & Reilly, A. H. (1996). Agency theory and variable pay compensation strategies. Academy of Management Journal, 39(3), 751.
48. Suddaby, R. (2006). From the Editors: What Grounded Theory is Not. Academy of Management Journal, 49(4), 633-642.
49. Wassermann, E., Czech, D. R., Wilson, M. J., & Joyner, A. B. (2005). An examination of the Moneyball Theory: A baseball statistical analysis. Sport Journal, 8(1).
50. Williamson, O. E. (2003). Examining economic organization through the lens of contract. Industrial and Corporate Change, 12(4), 917.

### Corresponding Author

Eric W. Ford, MPH, Ph.D.
Forsyth Medical Center Distinguished Professor of Management
The University of North Carolina Greensboro
PO Box 26165
Greensboro, NC 27402
<ewford@uncg.edu>
Phone: 806-787-3267
Fax: 336-334-5580