Contract Options and Performance: The Case of Major League Baseball.
Labor contracts, especially those in sports at the professional level, can have a variety of stipulations that seek to balance the needs of employer and employee. In professional sports, such as Major League Baseball (MLB), employment contracts vary by salary and length, and can include various performance incentives. Contracts in professional sports may also be renegotiated and extended prior to the contract's end. In addition to guaranteed contract years, contracts may also contain option years, which may or may not become additional years of a contract. How this year (or years) becomes guaranteed depends on the option type (Baseball Prospectus 2009-2018a). This study focused on the contract option feature of MLB player contracts.
The literature on contracts in professional sports, primarily MLB, have focused, with a few recent exceptions, on contract length. Numerous studies have looked for shirking behavior by MLB players in the early years of multi-year guaranteed contracts (Hakes and Turner 2008; Lehn 1982; Scoggins 1993; Krautmann 1993; Krautmann and Donley 2009; Krautmann 1990; Krautmann and Solow 2009). Bern and Krautmann (2006) have also looked for this behavior in the context of the National Basketball Association. Studies have looked for opportunistic behavior by players in the last year of a contract (O'Neill 2013; Harder 1991; Dinerstein 2007; Martin et al. 2011; Maxcy et al. 2002). Frick (2011) looked at opportunistic behavior in the final year of a contract for German soccer players.
Literature has recently begun to form surrounding aspects of MLB player contracts other than contract length. Krautmann (2018) looked at contract extensions in MLB. Contract extensions occur when a team and player renegotiate a new contract for the player prior to the end of the player's current contract. Krautmann (2018) found that these extensions take place most commonly in the last year of a player's contract, and more commonly occur for players with between 3 to 6 years of MLB service, those that are arbitration eligible. Another feature of MLB player contracts that has begun to receive attention is the contract option. Contract options are additional years of contract at the end of a player contract which are not guaranteed but can become guaranteed at the discretion of the player, team, both, or if certain conditions are met. Two papers in the literature have used principles of option theory in finance to assess whether teams rationally purchase and exercise contract options (Gross and Link 2017; Dinerstein 2007). Contract options that give the team the right to keep the player for additional years are called club options. Dinerstein (2007) looked to understand how teams go about deciding whether to exercise a club contract option. He found no pattern in team decision making regarding exercising options. Gross and Link (2017) looked to test whether teams follow option theory when purchasing and exercising contract options. Unlike Dinerstein (2007), they did find evidence that teams behave as expected. Players with greater performance inconsistency were more likely to be offered an option, and teams were more likely to exercise an option if the player's expected performance exceeds the value of the option.
While past literature looked at options in MLB from the perspective of the team, this study looked at options from the perspective of the player. More specifically, this study looked to understand how options affect incentives for player performance. As Gross and Link (2017) have found, and option theory would predict, teams will exercise a club option if the expected contribution of the player exceeds the player's option salary, minus the contract buyout. From the player's perspective, this should disincentivize effort as the player will not see any return on effort that would make his expected contribution in the option year exceed his salary minus the buyout. While about two-thirds of options in the sample were club options, all four option types are discussed in the next section. Different option types may incentivize effort in different ways.
Contract options are a feature of MLB player contracts where terms of a guaranteed player contract specify terms of potential optional years of the contract that come to fruition at the discretion of one or both parties to the transaction. The sample considered in this study consisted of player-year observations for contracts from the 2009 to 2017 MLB seasons. Only non-pitchers with at least 3 years of service were included in the sample, as no players with less than 3 years of service have contracts with options. About 13% of new contracts awarded during the sample period contained options. Most contracts with options included provisions for one optional year. While the whole sample contained new contracts with options at a rate of about 13%, this rate was about 20% for contracts signed in free agency, compared to only about 7% for those signed during arbitration eligible years.
Characteristics of players signing new contracts with and without options are shown in Table 1. The sample consisted of 1269 new player contracts, for 536 unique players. The difference in contract length between contracts with and without options was over 1 year. This indicated that contract options were more common in multi-year contracts than in one-year contracts. Not surprisingly, salaries were higher for contracts with options than those without. Overall, options were a more common feature of contracts that were greater in length and for higher salary. As mentioned previously, options were much more common in free agent contracts than in contracts given to arbitration-eligible players. Positions for players with contracts with and without options were generally similar. The only marginally significant difference was for center field.
Contract options in MLB can generally be classified into one of four types: club options, player options, mutual options, and vesting options. Todd (2014) gave detailed descriptions of the four option types. Club options give the team the right to extend the player's contract at the end of the contract, often by 1 year. If the club decides not to pick up the option, the player is paid a buyout, which is usually an amount far less than the full salary for the option year. Club options are by far the most common type of option in practice. In the sample period, nearly two-thirds of all contract options were club options. Player options give the player the ability to extend his contract at the end of the contract, again typically by 1 year. There is not a buyout associated with player options. These options are very uncommon in practice. In the sample period, just over 3 % (N= 5) of options were player options. Mutual options are a combination of player and team options that require both the player and the team to agree to extend the contract. If the team declines the option, the player is often paid a buyout. However, if the player declines the option, usually most of the buyout is forfeited. Outside of the buyout and fixed salary, having a mutual option is like having no option at all. About 20% of the options in the sample period were mutual options. The last type of option to consider is the vesting option. Vesting options could be considered a subset of club options. With vesting options, the option year becomes guaranteed if some threshold is surpassed by the player. This threshold usually involves the player reaching some level of games played, plate appearances, or innings pitched during the contract. If the threshold is not passed, the team then has the choice to pick up or decline the option.
How an option was likely to affect player incentives for exerting effort and subsequently performance would depend on the type of option. Club options limit the potential reward players could receive from exerting high effort and performing at a high level. The team can exercise an option for the pre-specified option year(s) at a fixed salary. The club option also contained a buyout that was paid to the player if it was not exercised, and the player could then sign with another team. These two forces should have disincentivized effort by players who were under contracts with options. As vesting options can be thought of as a subset of club options, the impact on player behavior should have been similar. Player options may have incentivized or disincentivized effort. As the player could choose not to exercise the option year, the player could reap the full benefits from exerting high effort and signing a new contract at a higher salary following the end of the guaranteed contract years. Alternatively, the player could choose to exert low effort and exercise the option receiving the pre-specified salary. As mutual options required both the player and the team to exercise the option, mutual options were like having no option, so the incentive effects were likely to be like that of having no option.
Data and Methodology
The sample considered in this study covers player-year observations from 2009 to 2017. Contract data came primarily from the Cot's Baseball Contracts website. This website had player contract data available starting in 2009 (Baseball Prospectus 2009-2018b). While this website contained information about contract length, salary, and whether a contract has an option, about 30 % of contracts with options do not specify the option type. Missing option types were searched and hand-entered into the dataset. This option type information came from media articles about the contracts, primarily from websites like ESPN.com (2018) and MLB.com (2018). Remaining data, primarily player performance data, came from the Baseball-Databank and Baseball-Reference websites (Lahman 2018; Baseball-Reference.com 2018).
As is standard in the literature, the sample was restricted to batters only. The sample was further restricted to only those players with at least 3 years of service. (1) This left a final sample for analysis of 2143 player-year observations.
Sample Restrictions and Descriptive Statistics
Table 2 shows descriptive statistics for the sample to be used in the analysis. As shown in the previous table, options were most commonly included in contracts that were longer and with higher salary. Players under contract with option had higher values of Wins Above Replacement, WINS. WINS was an advanced performance metric that looked to quantify player productivity into WINS added to his team over a replacement level player, typically a major league bench player or strong AAA player. BaseballReference.com (2018), the primary site which measures WINS. WINS was based on hitting, fielding and baserunning, and is a function of games played. This positive difference was likely due to positive selection. Better players were given longer contracts at a higher salary, and these contracts were more likely to contain options. About a quarter of the observations in the sample were contracts with options. This number was higher than in Table 1 because all player-year observations were considered here, and multi-year contracts were more likely to contain options.
The theoretical model gave predictions for how the player would perform when his contract had an option. Having a club (or vesting) option was predicted to disincentivize player effort. Mutual options were like having no option at all, and as such were grouped with contracts having no option. Player options could have a positive or negative impact. To test these predictions, regressions were run of the form
[mathematical expression not reproducible],
where HasChibOption is a dummy variable indicating that the contract has either a club or a vesting option, HasPlayerOption is a dummy variable indicating that the contract has a player option, ContractYearsLeft is the number of guaranteed years remaining on the contract, [[alpha].sub.i] is a player fixed effect, and [[lambda].sub.t] is a year fixed effect. In addition to allowing the relationship between contract years remaining and performance to be linear, three other functional forms were considered. A specification was considered where contract years remaining entered with dummy variables for 2 years left, 3 years left, and four or more years left. This allowed for the possibility that the relationship was non-linear. Another specification included just a dummy variable indicating four or more years remaining. This specification was like that of Krautmann and Donley (2009) which looked at long-term contracts. Last, a specification was used which included just a dummy variable indicating the final year of a contract. This was the form considered in the opportunistic behavior literature, like that of O'Neill (2013).
To further ensure the robustness of the presented results, additional control variables were included in the regressions. Position dummy variables were included to indicate fielding position. To address the possibility that team-related factors in a given year, such as managerial decisions, team standing, and teammate complementarities, may affect player performance, team-year fixed effects were included. The estimating equation for these regressions is
[mathematical expression not reproducible],
where [[rho].sub.it] are dummy variables indicating position, and [[theta].sub.it] is the set of team-year fixed effects. As with the other specification, contract years remaining were allowed to enter in multiple ways.
To understand how contract options affect player performance, regression results are presented. Table 3 shows results for regressions run with WINS as the outcome variable, and key independent variables indicating having a club option or having a player option. The results indicated that having a club option had a negative impact on player performance, as predicted. In two specifications, this coefficient had two-tailed significance only at the ten-percent level, and in the other two it only had one-tailed significance at the ten-percent level. For player options, the coefficients were small and positive, but insignificant. This was not surprising as the sample with player options was small, and the prediction of the impact of a player option was unclear.
Table 4 included additional control variables in the regressions presented. Specifically, fielding-position dummy variables have been added, as have team-year fixed effects. The inclusion of these additional control variables led to an increase in the magnitude of the coefficients reflecting a club or vesting option. This coefficient in each regression was statistically significant at the five-percent level. These results provided further support for a negative relationship between having a club or vesting option and player performance.
This study looked to understand MLB player performance when contracts had option years. About 13% of new contracts awarded to batters during the sample period contained at least one option year. Of contracts that had option years, the most common option type was the club option, although about a third of the contracts had either player, mutual, or vesting options. The expected impact of various option types on effort and performance was discussed, and notably it was predicted that club and vesting options would disincentivize effort. These results showed that when controlling for the number of years remaining on a contract, having a club option led the player to perform at a lower level. While the results of Gross and Link (2017) suggested that teams take out options to protect themselves from players with highly variable output, the results of this study suggested that teams should fear these options may lead to lower player effort.
Having a contract option is not much unlike the employment situation of many working adults. Outside of academia and professional sports, having a guaranteed contract or a high sense of job security is not very common. Employment situations where compensation is fixed at a set level in some way are comparable to contracts with a club option. This is likely the case with many government jobs. Employees are employed at will by their employers who continually choose whether to continue the employment relationship. Because most employees are not under long-term contract, there is some incentive to perform at a high level to keep a job. Because the employee feels some sense of job security and the salary is set at some level, effort levels are likely not as high as they would be if the employee was constantly fighting for a new job or could renegotiate higher salaries.
There is still much work to be done in trying to understand the use of options in MLB and the impact these options have on player effort. One drawback of this existing work is the small sample size. Because options make up a relatively small percentage of contracts in MLB, additional years of data could help to improve the precision of the presented results. Increased data availability could also help in identifying differential impacts on player effort by type of option. Studying options in other sports could also shed light on this issue. While player options are uncommon in MLB, they are more common in the National Basketball Association. The literature would also benefit from further work on other non-contract length and non-salary components of contracts, such as extensions and performance incentives.
Published online: 13 December 2018
Acknowledgements I would like to thank Dr. Mindy Marks for her extensive feedback on this work. This work also benefitted from comments from seminar attendees at Northeastern University.
Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Baseball Prospectus. (2009-2018a). Cot's Baseball Contracts. "2017-2021 Collective Bargaining Agreement", Retrieved from: http://legacy.baseballprospectus.com/compensation/cots/ Accessed 18 August 2017.
Baseball Prospectus. (2009-2018b). Cot's Baseball Contracts. "MLB Contracts, By Team" Retrieved from: http://legacy.baseballprospectus.com/compensation/cots/ Accessed 1 November 2017.
Baseball-Reference.com. (2018). Index of /data/. Retrieved from: http://www.baseball-reference.com/data/ Accessed 12 February 2018.
Berri, D. J., & Krautmann, A. C. (2006). Shirking on the court: Testing for the incentive effects of guaranteed pay. Economic Inquiry, 44(3), 536-546.
Dinerstein, M. (2007). Free agency and contract options: How major league baseball teams value players. PhD diss., Stanford University.
ESPN.com. (2018) MLB - Major League Baseball teams, scores, stats, news, standings, rumors. Retrieved from http://www.espn.com/mlb/ Accessed 21 March 2018.
Frick, B. (2011). Performance, salaries, and contract length: Empirical evidence from German soccer. International Journal of Sport Finance, 6, 87-118.
Gross, A., & Link, C. (2017). Does option theory hold for Major League Baseball contracts? EconomicInquiry, 55(1), 425-433.
Hakes, J. K., & Turner. C. (2008). Long-term contracts in Major League Baseball. In IASE/NAASE Working Paper Series, No. 08-31 (pp. 1-43).
Harder, J. W. (1991). Equity theory versus expectation theory: the case of Major League Baseball free agents. Journal of Applied Psychology, 76(3), 458-464.
Krautmann. A. C. (1990). Shirking or stochastic productivity in Major League Baseball? Southern Economic Journal, 56(4), 961-968.
Krautmann. A. C. (1993). Shirking or stochastic productivity in Major League Baseball: reply. Southern Economic Journal, 60(1), 241-243.
Krautmann. A. C. (2018). Contract extensions: the case of Major League Baseball. Journal of Sports Economics, 79(3), 299-314.
Krautmann. A. C., & Donley. T. D. (2009). Shirking in Major League Baseball revisited. Journal of Sports Economics, 10(3). 292-304.
Krautmann, A. C., & Solow, J. L. (2009). The dynamics of performance over the duration of Major League Baseball long-term contracts. Journal of Sports Economics, 10(1), 6-22.
Lahman, S. (1996-2018). Download Lahman's baseball Database. Retrieved from: http://www.seanlahman. com/baseball-archive/statistics/ Accessed 7 July 2018.
Lehn, K. (1982). Property rights, risk sharing, and player disability in Major League Baseball. The Journal of Law & Economics, 25(2), 343-366.
Martin, J. A., Eggleston, T. M., Seymour, V. A., & Lecrom, C. W. (2011). One-hit wonders: A study of contract-year performance among impending free agents in Major League Baseball. NINE: A Journal of Baseball History and Culture, 20(1), 11-26.
Maxcy, J. G., Fort, R. D., & Krautmann, A. C. (2002). The effectiveness of incentive mechanisms in Major League Baseball. Journal of Sports Economics, 3(3), 246-255.
MLB.com. (2018). The official site of Major League Baseball. Retrieved from https://www.mlb.com/ Accessed 21 March 2018.
O'Neill, H. M. (2013). Do Major League Baseball hitters engage in opportunistic behavior? International Advances in Economic Research, 19, 215-232.
Scoggins, J. F. (1993). Shirking or stochastic productivity in Major League Baseball: comment. Southern Economic Journal, 60(1), 239-240.
Todd. J. (2014). Options in MLB contracts: primary option types. Retrieved from: https://www. mlbtraderumors.com/2014/01/options-in-mlb-contracts-primary-option-types.html Accessed 23 March 2018.
(1) Players can generally be divided into three groups based on MLB service, the term given to professional experience. Players with less than 3 years of service can only receive contract offers from their current team, typically the team that selects the player in the MLB draft, and the player cannot negotiate for a higher offer. These players are paid at or near the league minimum and typically receive one-year contracts. No players with less than 3 years of service from 2009 to 2017 have a contract with an option. Players with between 3 and 6 years of service are arbitration eligible. While these players cannot receive outside offers, if the player feels his offer is not fair, he and the team can enter into arbitration overseen by a third party. Players with greater than 6 years of service are eligible for free agency, where all teams can make offers to the player (Baseball Prospectus 2009-2018a).
Richard J. Paulsen (1) (iD)
[mail] Richard J. Paulsen
(1) Northeastern University, 301 Lake Hall, 360 Huntington Avenue, Boston, MA 02115, USA
Table 1 Player characteristics by option status Variable All No option With option Difference Length 1.55 1.41 2.47 1.06 *** Salary 3.89 M 3.65 M 5.53 M 1.87 M *** Age 31.1 31.0 31.7 Q 79 *** Free Agent 0.45 0.41 0.69 0.28 ** Catcher 0.17 0.17 0.21 0.04 First Base 0.14 0.13 0.15 0.01 Second Base 0.14 0.14 0.13 -0.00 Third Base 0.12 0.12 0.16 0.04 Shortstop 0.14 0.14 0.12 -0.02 Left Field 0.11 0.11 0.10 -0.02 Center Field 0.09 0.09 0.05 -0.04 * Right Field 0.09 0.09 0.13 0.04 Designated Hitter 0.04 0.04 0.06 0.02 N 1269 1106 163 Notes: The sample is restricted to non-pitchers and new contracts only. All hypothesis tests are two tailed, with *** indicating p<0.01, ** indicating p< 0.05, and * indicating p< 0.1. Source: Information in the table comes from the author's calculations, based on data from Lahman (2018), Baseball-Reference.com (2018), and the Baseball Prospectus (2009-2018b) Table 2 Descriptive statistics Variable All No option With option Difference Length 2.74 2.41 3.72 1.31 *** Salary 6.57 M 6.08 M 8.03 M 1.95 M *** Age 31.3 31.3 31.3 0.01 Has Option 0.25 0 1 1 *** WINS 1.45 1.36 1.73 0.38 *** N 2143 1606 537 Source: Information in the table comes from the author's calculations, based on data from Lahman (2018), Baseball-Reference.com (2018), and the Baseball Prospectus (2009-2018b) Table 3 Contract has option results (WINS) Variable WINS WINS WINS WINS Contract -0.084 ** years left (0.038) Two years left -0.317 *** (0.100) Three years -0.239 left (0.153) Four plus -0.229 -0.071 years left (0.182) (0.168) One year left 0.284 *** (0.093) Has club -0.278 * -0.243 -0.270 * -0.248 option (0.154) (0.155) (0.158) (0.155) Has player 0.124 0.140 0.150 0.131 option (0.666) (0.671) (0.671) (0.669) Age 0.403 0.387 0.309 0.398 (0.288) (0.283) (0.281) (0.281) Age Sq. -0.013 *** -0.012 *** -0.011 ** -0.013 *** (0.005) (0.004) (0.004) (0.004) N 2143 2143 2143 2143 Adj. R-sq. 0.213 0.215 0.210 0.215 Notes: The sample is restricted to non-pitchers. Player- and year- fixed effects are included in all regressions. Regressions are run using heteroskedasticity robust standard errors, clustered on player. All hypothesis tests are two tailed, with *** indicating p<0.01, ** indicating p<0.05, and * indicating p<0.1. Source: Information in the table comes from the author's calculations, based on data from Lahman (2018), Base- ball-Refcrence.com (2018), and the Baseball Prospectus (2009-2018b) Table 4 Contract has option results with additional controls (WINS) Variable WINS WINS Contract years -0.088 ** left (0.038) Two years left -0.288 *** (0.104) Three years -0.198 left (0.152) Four plus -0.258 -0.122 years left (0.177) (0.159) One year left 0.264 *** (0.097) Has club -0.349 ** -0.317 ** -0.343 ** -0.320 ** option (0.151) (0.151) (0.154) (0.151) Has player -0.208 -0.216 -0.186 -0.224 option (0.826) (0.826) (0.838) (0.825) Age 0.570 0.546 0.482 0.554 (0.846) (0.840) (0.841) (0.845) Age Sq. -0.012 ** -0.011 ** -0.010 ** -0.011 ** (0.005) (0.005) (0.005) (0.005) N 2143 2143 2143 2143 Adj. R-sq. 0.356 0.356 0.353 0.356 The sample is restricted to non-pitchers. Player fixed effects, fielding position dummy, and team-year fixed effects are included in all regressions. Regressions are run using heteroskedasticity robust standard errors, clustered on player. All hypothesis tests are two tailed, with *** indicating * p<0.01, ** indicatingp<0.05, and *** indicating p < 0.1. Source: Information in the table comes from the author's calculations, based on data from Lahman (2018), Baseball-Reference.com (2018), and the Baseball Prospectus (2009-2018b)
|Printer friendly Cite/link Email Feedback|
|Author:||Paulsen, Richard J.|
|Publication:||Atlantic Economic Journal|
|Date:||Dec 1, 2018|
|Previous Article:||How Social Security's Earning Test, Age and Education Affect Female Labor Supply.|
|Next Article:||Income Elasticity Decomposition Models and Determinants of U.S. Pharmaceutical Expenditures.|