Printer Friendly

Negotiated settlement under mlb final-offer salary arbitration system.

This paper provides a detailed analysis of negotiated salaries under Major League Baseball's final-offer arbitration process using data from the 2007-2010 seasons. There is a wage premium of 25% for hitters and 14% for pitchers filing for arbitration. Interestingly, there is an additional premium for exchanging offers for hitters but not for pitchers. The additional premium in salary for hitters who exchange offers with their clubs amounts to 7%. (JEL J3 1, J52)


In negotiated settlements, the costs of reaching an agreement can be expensive in terms of both time and money. If both sides could realize the objective, unbiased, end result from the beginning, then an accord could be reached quickly. This is rarely the case. In industrial relations, the use of mediators and arbitrators is designed to assist union and management negotiators in reaching a contract settlement. The use of mandatory interest arbitration to resolve bargaining disputes is found predominantly in the public sector; this is designed to reduce the negotiation costs to taxpayers. Unable to reach a settlement, unionized police, firefighters, teachers, and the respective governing bodies in charge of the budgets of these employee groups are often forced to allow a neutral, third party to decide the major fiscal contents of their new collective bargaining agreements (CBAs). In rare cases such as in New Jersey, the wages of police and firefighters are determined by final-offer arbitration, a system in which the arbitrator is constrained to choose either the offer proposed by the union or the offer proposed by the management. Final-offer arbitration was designed to reduce the use of expensive arbitration as, in theory, the requirement that the arbitrator must accept the final offer of one side or the other would push both sides closer together and allow for a negotiated settlement.

Major League Baseball (MLB) has used a system of final-offer salary arbitration (FOSA) since 1973 to determine the salaries of players of a certain experience level. Baseball's FOSA process was thought to provide a "laboratory experiment" that would provide greater insight into the workings of such a system of interest arbitration because of the level of data available for analysis. The historical record of these cases has been studied extensively by economists to analyze topics such as the determinants of arbitrators' decisions, the determinants of the probability that the dispute will end in arbitration, and the effect of each party's willingness to bear risk on the salary outcome. Much of the previous research into baseball's final-offer system has centered on the final offers themselves and the arbitrators' decisions. However, under baseball's system, a negotiated salary agreement can come at any point before the arbitration panel renders a decision.

This paper uses data from the 2007 through 2010 seasons and analyzes negotiated salaries under three out of the four separate stages of the FOSA system: those who are eligible for arbitration and do not file for it; those who file for arbitration and do not exchange salaries; and those who exchange salaries and do not reach the arbitration stage of the process. Specifically, the analysis provides estimates of the average treatment effect for players moving to each subsequent, aforementioned stage in the bargaining process. No other studies of the FOSA process in MLB have examined the effects of filing for arbitration on the salaries for players. Other published studies have used the list of players who filed for arbitration as their dataset. This ignores the first step in the baseball arbitration process.' Burgess and Marburger (1993) concluded that arbitration awards won by players were higher and those won by owners were lower than negotiated settlements for comparable players. Miller (2000b) found that negotiated salaries that occurred after the exchange of final offers under arbitration were significantly different from those for free agents, while Marburger (2004) found that the average free agent salary was a significant determinant of both player and management final offers. Other researchers (Farmer, Pecorin, and Stango 2004; Miller 2000a) have examined the role of risk in shaping the outcome of negotiated and arbitrated settlements after the exchange of final offers. Therefore, the majority of the earlier literature has focused on the stages of the process after each side has made and exchanged final offers. Because the majority of players who are eligible for salary arbitration today settle before the exchange of offers, it is important to understand the earliest stages of the process, which is the focus of this paper.

The rest of this paper proceeds by examining the FOSA process in MLB in Section II. Theory and the previous literature are discussed in Section III. Section IV provides a discussion of the data, model, and empirical methodology, while Section V details the results of the analysis. Section VI offers conclusions.


The FOSA system began with the 1973 CBA. The FOSA system has four distinct stages: players being eligible for arbitration; filing for arbitration and not exchanging offers; exchanging offers and settling prior to arbitration; and arbitrating salaries. (2) Service time determines whether a player is eligible for FOSA. (3) Currently, eligible players are those who have between 3 and 6 years of service time. (4) Players with 2 years of service who had at least 86 days of service during the preceding season and ranked in the top 17% in total service among this group are eligible as well (these players are known as the super-twos). (5) Players know if they meet these eligibility criteria shortly after the conclusion of the season. Either eligible players or owners can file for arbitration between January 5 and 15 following a season. Players and clubs exchange offers by January 18. The league then schedules arbitration hearings between February 1 and 20. A three-member arbitration panel renders a decision usually within 24 hours after the conclusion of the hearing. (6) The panel must choose either the salary offer of the player or the club and cannot render a compromised settlement.

The parties can negotiate a salary any time before the arbitration panel renders its decision, and this occurs in the majority of cases. For example, between 2007 and 2010, 567 players were eligible *for FOSA. Of those players, 467 (approximately 81% of the eligible) filed for arbitration, I 82 players (about 32% of the eligible) filed for arbitration and exchanged offers, and 22 players (about 4% of the eligible) went through the entire arbitration process. In fact, 385 players (68%) settled before the exchange of offers. Using data from 1993 to 1996, Farmer, Pecorin, and Stang[degrees] (2004) had a sample of 527 players who filed for arbitration, with only 82 (16% of those who filed) players settling before an exchange of offers, 374 (71% of those who filed) exchanging offers, and 71(13% of those who filed) receiving arbitrated settlements. Because the authors only use a sample containing players who filed for arbitration, it is difficult to get a direct comparison of the numbers. Still, however, there is a dramatic contrast in some of these figures: 374 players exchanging offers in the 1993-1996 period versus 182 players in the 2007-2010 period; 71 settlements reached by .arbitration from 1993 to 1996 versus 22 settlements reached by arbitration in the 2007-2010 period.

The contrast in these figures illustrates the belief that a change in the approach to the use of salary arbitration has taken place despite no change in the rules and regulations governing the process. If there is an accepted relationship between salary and past performance, then this may explain why some agreements on salary are being reached during earlier stages of the negotiation process. Another potential explanation follows the reasoning of Brown and Link (2010) that labor relations are more harmonious now between the owners and the players following previous decades of discord over collusion by owners and strikes by the players.


A beginning point for almost all of the theoretical models on FOSA in baseball is the work of Farber (1980). Using his basic construct for final-offer arbitration, it is assumed that an arbitrator will choose the offer made by a club if:

( 1) |[Y.sub.A] - [Y.sub.c| < |[Y.sub.A] - [Y.sub.p|,

where [y.sub.A] represents the arbitrator's calculation of an objective salary for the player given the criteria set forth in the CBA, ye is the offer rendered by the club, and [y.sub.3c] is the offer from the player or his agent: it is assumed that ye and [y.sub.p] are a function of [y.sub.A]. The player's offer will be chosen if the inequality holds in the opposite direction.

Under the assumption that clubs' and players' salary offers are risk-neutral and arbitrators are unbiased and interchangeable, Faurot and McAllister (1992) find that four of the criteria listed in the CBA are all significant determinants of the expected value of the arbitrators' fair settlement. Fizel (1996) finds racial bias in the decisions rendered by arbitrators. Marburger and Burgess (2004b) use a probit model to predict the winning offers in arbitration cases. They conclude that the FOSA process not only favors reasonable offers but also creates an incentive to settle before arbitration to avoid unfavorable rulings.

Farber's (1980) model can also be used to analyze the effect of each side's level of risk-aversion on the outcome. If F([y.sub.A]) represents the distribution of the arbitrator's fair settlement and is assumed to be known by both sides, then the club chooses a bid y, that will maximize

(2) (1- F([y.sub.c] + [y.sub.p])/2][y.sub.c] + F([y.sub.c] + [y.sub.p])/2][y.sub.p]

assuming that utility from the bids are strictly concave and increase monotonically. The player selects his bid. [y.sub.p] in a symmetrical fashion. From these, each side derives a reaction function to the other's optimal offer and the simultaneous solution of each yields Nash equilibrium.

Farber (1980) discusses the so-called contract zone as an area in which parties may reach an agreement because of a convergence of their final offers. This contract zone can be defined as:

(3) [Y.sub.c] < [Y.sub.L] < [Y.sub.u] < [Y.sub.p]

where YL is the lower bound of the zone and Yu is the upper bound. Some research has focused on the contract zone. Marburger (2004a) suggests that final offers by both management and players are a weighted average of the player's past season compensation and the average free agent salary of the current season; this same result holds using only cases that ended in arbitration. Hadley and Ruggiero (2006) use nonparametric analysis and conclude that arbitrators and the FOSA process are approximately mimicking the free agent process.

Faber (1980) contends that if uncertainty over [y.sub.A], the arbitrator's fair settlement, disappears, then the contract zone shrinks to [y.sub.A] as both [Y.sub.c] and [Y.sub.P] are a function of [y.sub.A]. However, if uncertainty increases the gap between the final offers, then arbitrated settlements may increase. These are said to be of "low quality" by Farber (1980) as they fall outside the contract zone. The conclusion by Burgess and Marburger (1993) that arbitrated salaries won by players were higher and those won by management were lower in the baseball FOSA system than negotiated salaries for comparable players offers empirical verification of Farber's conclusion.

Miller (2000b) incorporates negotiating costs into his theoretical model to allow for negotiated settlements in baseball's FOSA system. Farber (1980) concludes that costs of arbitration widen the contract zone and increase the likelihood of a negotiated settlement. However, if costs are asymmetrical, then the party subject to the higher costs will become more risk-averse and the resulting contract zone will move unfavorably for this party and get larger overall. According to baseball's CBA, the hearing costs of arbitration are borne equally by the club and player, and each is responsible for his own expenses and those of his counsel or other representatives. Since a club may reap some economies of scale if hearings are set for more than one player, it is possible that costs are asymmetrical.

The basic Farber model can be used to frame the current empirical research, but some interesting questions arise. To what extent does filing for arbitration differ from simple eligibility for arbitration in the effect on a negotiated salary? Second, to what extent does exchanging offers differ from simply filing in the effect on a negotiated salary? Previous studies of baseball's FOSA system have not dealt with these issues. Marburger and Scoggins (1996) use a probit model to determine that higher quality players are more likely to file for arbitration and press for an arbitrated settlement. This means that selectivity bias is a likely problem in the estimation of salary equations for a self-selecting group of players. Farmer, Pecorin, and Stang[degrees] (2004) use a two-stage process to adjust for any selectivity bias to isolate the effect of aggressive bargaining behavior on negotiated versus arbitrated salaries. Likewise, Miller (2000a) uses probit models for arbitration-eligible and free agent players to correct for any selectivity bias in salary regressions.

This paper has reviewed the Farber (1980) model in this section. It discusses the Miller (2000a) model in Section 4B. Both serve as starting points for the research presented here. Recall that the purpose of this paper is to analyze negotiated salary outcomes at the three early stages of IVILB's FOSA process. Farber (1980) focuses on how final offers are created along with the consequences for negotiated settlements. Miller (2000a) examines bargaining after final offers have already been exchanged. None of these papers models the bargaining that takes place before the exchange of final offers, however. Therefore, there are no implications from these models that provide a hypothesis that is being tested in this paper. The estimated equation presented below would require a bargaining model that allows negotiations to occur before the exchange of final offers. Many bargaining models already exist that can be usefully extended to include bargaining behavior before the exchange of final offers, and, specifically, the decision to file for arbitration. While this type of extension is beyond the scope of this paper, it would be a fruitful area for future research.7


A. Data

The dataset for this study includes the salaries of all arbitration-eligible players from the 2007 through 2010 seasons. The data also include previous season and career performance statistics for these players. (8) The data is disaggregated into two subgroups. The first subgroup contains hitters and the second subgroup contains pitchers. Recall that the purpose of this study is to investigate the effect of advancement through each of the first three stages of the FOSA process on players' salaries. To this end, each subgroup is delineated by the four stages; players who are eligible; players who file for arbitration; those who exchange salary offers; and those who proceed through the entire arbitration process.

Table 1 provides summary statistics by subgroup and arbitration stage. The top panel of the table contains information for hitters, while the bottom panel is for pitchers. Each stage of the FOSA process listed in Table 1 is mutually exclusive. In other words, the column labeled "Eligible" provides statistics for those players who are eligible for arbitration and do not file for arbitration; the column "Filed" contains statistics for those who file for arbitration and do not exchange salary offers with the teams' owners. The column labeled "Exchanged" displays statistics for those players who exchange salary offers and do not proceed through the arbitration process; "Arbitrated" provides calculations for those players who proceed through the entire FOSA process. (9) Finally, the first column of the table lists the variables of interest.

Table 1 shows that for both hitters and pitchers, the previous season's salary is lower and the current season's salary is higher as players proceed through each stage of final-offer arbitration. This provides transient evidence that there is a premium associated with each FOSA stage. Table 1 also shows that those who receive more playing time are more likely to proceed through each stage of the arbitration process. This is true for hitters and pitchers as evidenced by the increases in at bats and career at bats for hitters and innings pitched and career innings pitched for pitchers. Hitters and pitchers with less service time appear to be more likely to proceed through each stage of the arbitration process. Finally, better players are more likely to enter into the later stages of the FOSA process as evidenced by the changes in the performance statistics displayed in Table 1. These descriptive statistics indicate that there is selection, potentially, into the various stages of the FOSA process.

Descriptive Statistics by Player Type and Eligibility Status


                  Eligible      Filed  Exchanged  Arbitrated

Previous salary    772.491    700.595    637,329     578.874

Current salary     925.868  1,249.929  1,556.606   1,969.986

Service lime          4.03       3.55       3.44        3.53

Team attendance  2,443.523  2,538.231  2,604.545   2,135.266

Team winning          0.49       0.51       0.51        0.51

At bats             316.37     365.33     490.00      471.73

Career at bats     1404.92    1401.84    1540.56     1616.91

Slugging              0.39       0.43       0.45        0.43

Career slugging       0.41       0.42       0.44        0.44

On base               0.32       0.33       0.35        0.34

Career on base        0.33       0.33       0.34        0.34

Sample sizes            51        127         77         1 1


                  Eligible      Filed  Exchanged  Arbitrated

Previous salary    651,451    606.065    614,656     912.857

Current salary     957,774  1.054.612  1,304.721   1.832.635

Service time          3.89       3.80       3.72        3.74

Team attendance  2.435,287  2.536.583  2.614.206   2,697.242

Team winning          0.50       0.50       0.50        0.47

Saves                 3.31       4.67       3.67       12.45

Innings pitched      75.61      87.25     121.18       95.37

Career innings      365.97     380.94     445.37      400.89

Earned run            4.73       4.10       3.80        3.57

Career earned         4.31       4.12       4.05        3.81
run average

Relief pitcher          36        111         48           6

Sample sizes            54        153         83          11

Notes: The categories listed in the column are mutually exclusive.
In other words, the category Eligible indicates that the players
are eligible for arbitration but did not file. Those who filed
for arbitration did not exchange offers. Those who exchanged
offers did not go through the arbitration process.

Source: Authors' calculations taken from the data. See text for

B. Model and Methodology: Treatment Effects of Filing and Exchanging Offers

As stated previously the underlying model for the previous work in this area is the theoretical model developed by Farber (1980) in which a contract zone is developed based on the objective salary for the player estimated by a neutral, unbiased arbitrator, and the final offers of the player and the club. Miller (2000a) offers an approach that better fits the focus of this research but it must be adapted to allow for negotiated settlements both before and after the exchange of final offers. Consider an intertemporal version of the Miller (2000a) cooperative bargaining model in which Nash equilibrium is found from

(4) [y*.sub.ns] = arg [max.sub.yns]([U.sub.ns](1-[y.sub.ns])-[d.sub.cs]) x ([] ([Y.sub.ns])-[]),

where [y.sub.ns] is the salary reached through negotiation in stage s of the FOSA process. The utility of the club in step s, [U.sub.cs], and the utility of the player in step s. [] , are a function of the bargained salary and the respective disagreement outcomes in step s, [d.sub.cs] and [] , the expected utility of proceeding to the next step, s +1, in the arbitration process. Let [y.sub.cs] and [] be the salary offers of the club and player respectively in stage s; these offers are unobservable in the early stages of the process but represent the final offers in the latter stages of the process. From the disagreement functions (10) and Equation (4) above, the first-order condition defines the function for the negotiated salary

(5) [y*.sub.ns], = [Y*.sub.ns] [[d.sub.cs] ([Y.sub.cs] , [])[] ([Y.sub.cs], [] )

Miller (2000a) notes how Equation (5) is an increasing function in the disagreement point for players and a decreasing function in the disagreement point for clubs. Recall that there are three stages under scrutiny in this research: negotiated settlements for eligible players who do not file for arbitration, negotiated settlements for players who file for arbitration but settle before an exchange of offers, and negotiated settlements for players who file and exchange offers but settle without arbitration. Therefore, if there is an increase in negotiated salary during each subsequent stage in the bargaining process, then it can be interpreted that the arbitration process is either increasing the disagreement point for players or decreasing the disagreement point for the clubs.

Obviously, the wage offers of the club and the player are a function of the player's past performance and years of experience. Asymmetric information may cause a divergence between the player and club offers. These differences may shrink as more information is exchanged between parties in subsequent stages of the process. The willingness to assume risk in the negotiation process will shape offers as well. It seems likely that both sides may be willing to assume more risk in the earlier stages of the process before the exchange of final offers. The act of making a final offer leaves either party vulnerable to an adverse decision should the opposing side opt for arbitration. The costs of negotiation, both real and psychic, can also play a role in the utility maximization of each side. Real costs of negotiation may be minimal during early stages where offers can be exchanged via phone or fax. Costs of an arbitrated decision are borne equally by parties and include airfare and/or hotel stays for the arbitrator, expert witness, and/or lawyers/agents. Psychic costs include the psychological stress for players not knowing where they and their families may be living next season. For club general managers and coaches there is stress from dealing with players disgruntled by the negotiation process,

The cost of disagreement favors management in the pre-filing stage of the process as players must accept a contract from management if they do not file for arbitration. This may cause better players to advance to the filing stage to increase their bargaining power, that is, their disagreement point. Marburger and Scoggins (1996) find that higher quality players are more likely to file for arbitration and press for an arbitrated settlement. After filing, but before the exchange of final offers, it is unclear which side has an advantage. Once final offers are exchanged, both sides face the prospect of an adverse decision by the arbitration panel. Perhaps better players feel that they have an advantage because clubs may want to keep them happy so their performance is not negatively affected by the rancor that can surround negotiations. This implies that there should be a relatively higher salary premium associated with filing for arbitration instead of for exchanging final offers.

Recall that the purpose of this paper is to analyze how the process of moving to each subsequent stage of MLB's FOSA process influences the salary of players. As suggested in Table 1, there appears to be a salary premium associated with moving to each stage. Therefore, it is logical to think of the process of filing for arbitration, or the act of officially exchanging offers with the team as a type of program in which players participate, and these various programs should influence salaries in one way or another. The empirical methodology will estimate the average treatment effects on players' salaries of moving to each subsequent stage in the FOSA process. Table I also shows that better players appear to be more likely to move on to each stage. Therefore, the estimated equations need to account for selection into the participation of each program, that is, filing and/or exchanging final offers. If selection on observable characteristics is assumed, then a standard model for estimating the average treatment effects is the following:

List of Variables for Regressions

Common Variables         Hitters Only           Pitchers Only

Log of previous       At bats             Saves

Service time          Career at bats      Innings pitched
                      divided by service

Team attendance       Slugging average    Career innings pitched
11.000.000)                               divided by service time
Team winning          Career slugging     Earned run average
percentage            average

Year dummy variables  On base percentage  Career earned run average
First time            Career tin base     Relief dummy
eligibility dummy     percentage
Multiyear contract    First base dummy    Relief dummy x innings
dummy                                     pitched
                      Second base dummy   Relief dummy x (career
                                          pitched/service time)

                      Third base dummy    Relief dummy x saves
                      Catcher dummy
                      Short stop dummy

(6) ln([]) = [[beta].sub.0] + [[beta].sub.1][ - 1] + [[beta].sub.2][ - 1] + [[beta].sub.3][ - 1]([ - 1] - [bar.X]) + [].

In Equation (6), In([]) ) is the natural log of player i's salary in season t. The [ - 1] contains a set of previous season performance characteristics, team-specific variables, and dummy variables for position played, year, a player being eligible for arbitration for the first time, and a negotiated salary resulting in a multi-year contract. The full set of variables contained in [ - 1] is in Table 2.

Equation (6) is estimated separately for hitters and pitchers. For each sample, Equation (6) is estimated twice. The first time, the sample used includes players who are eligible for arbitration and those who filed for it and did not officially exchange salary offers with their team. The second time uses the sample of players who filed for arbitration and exchanged salary offers but did not move to the final stage of the arbitration process. Given that only 22 players (11 hitters and 11 pitchers) actually moved to the arbitration stage, they are excluded from the analysis sample. The variable [ - 1] is a dummy variable equaling 1 if the player filed for arbitration (for the first estimation) or officially exchanged a salary offer with his team (during the second estimation) between seasons t - 1 and t. The estimate of [[beta].sub.1] is the average treatment effect for moving to each subsequent stage in the arbitration process. The interactions between [ - 1] and the de-meaned variables contained in Table 2 help to control for selection on observable characteristics. (11)


The results from Equation (4) are presented in Tables 3 and 4. Table 3 presents the results for hitters, and Table 4 is for pitchers. The first column in each table contains the independent variables used in the analysis. The second and third columns present the parameter estimates by different subsamples of the data. Model 1 uses the sample of players who did not exchange salary offers. Put another way, Model 1 focuses on the movement from being eligible to filing for arbitration but not exchanging salary offers. Model 2 uses the sample of players who file for arbitration and do not go all of the way through the arbitration process. In other words. Model 2 examines the movement from filing for arbitration to exchanging salary offers.
TABLE 3 Treatment Effects of Filing and/or Exchanging
on Salary--Hitters

Sample              No Exchange Model 1  File and Exchange Model 2

Log of previous             0.218                      0.245
salary                    (1.70)*                  (4.91)***

Service time                0.081                      0.174
                           (0.89)                  (4.15)***

Team attendance            -0.069                     -0.045
(1.000.000)                (1.08)                     (1.37)

Winning percentage          1.063                      0.575
                          (4.91)*                    (1.70)*

At bats                     0.002                      0.002
                        (3.27)***                  (8.07)***

Career at                   0.001                      0.002
bats/service time        (2.05)**                  (7.24)***

Slugging average           -2.049                      0.751
                           (1.56)                     (1.00)

Career slugging             7.092                      3.174
average                 (4.65)***                   (2.75)**

On base percentage          3.447                      1.261
                         (2.50)**                     (0.89)

Career on base             -3.162                     -1.985
percentage                 (1.50)                     (0.94)

First base                  0.423                     -0.075
                         (2.11)**                     (0.55)

Second base                 0.047                     -0.092
                           (0.48)                     (1.45)

Third base                 -0.214                      0.050
                         (2.18)**                     (0.66)

Catcher                     0.019                      0.137
                           (0.13)                   (2.51)**

Short stop                 -0.106                      0.017
                           (1.01)                     (0.25)

First time                 -0.012                      0.048
eligible                   (0.10)                     (0.76)

Multiyear contract          0.023                     -0.111
                           (0.29)                     (1.18)

Filed                       0.223                         --
                         (3.54)**                         --

Exchange                      --                       0.072
                              --                    (2.06)**

Observations                 178                         204

[R.sup.2]                   0.93                        0.94

Notes: Robust t statistics in parentheses. Standard errors
clustered at the team level used in the calculations. All
regressions include year dummy variables and interactions
between the treatment dummy variable and the demeaned
independent variables listed in Table 2. See text for

* Significant at 10%; ** significant at 5%; ***
significant at 1%.

Before analyzing the average treatment effects, the coefficients associated with the variables found in Table 2 are discussed. Results in Tables 3 and 4 show some similarities with regards to pay between hitters and pitchers. Service time and the previous season's salary appear to influence positively the current season's salary. Additionally, those players who receive more playing time receive higher salaries. This can be seen through the positive and significant coefficients associated with at bats for hitters and innings pitched for pitchers. As expected, better performance also results in higher negotiated salaries as evidenced by the significant coefficients associated with career slugging average for hitters and career earned run average for pitchers. Interestingly, being eligible for arbitration for the first time and salary negotiations resulting in multi-year contracts have no discernible impact on players' current salaries. One difference between hitters and pitchers in Tables 3 and 4 is the effect of team variables on salaries. The team's previous winning percentage positively affects current season's salary for hitters but not pitchers.
TABLE 4 Treatment Effects of Filing and/or Exchanging on

Sample              No Exchange Model 1  File and Exchange Model 2

Log of previous             0.323                      0.346
salary                  (3.01)***                  (5.89)***

Service time                0.163                      0.111
                         (2.66)**                  (4.66)***

Team attendance            -0.005                     -0.019
(1.000,000)                (0.07)                     (0.49)

Winning percentage          0.860                     -0.657
                           (1.20)                     (1.21)

Saves                      -0.019                     -0,019
                           (0.09)                     (0.10)

Innings pitched             0.007                      0.006
                        (3.66)***                  (4.48)***

Career innings           4.45E-04                      0.002
pitched/service            (0.19)                    (1.85)*

Earned run average         -0.002                     -0.069
                           (0.18)                  (3.91)***

Career earned run          -0.132                     -0.130
average                   (1.72)*                  (4.26)***

Relief dummy               -0.082                      0.414
                           (0.38)                    (1.91)*

Relief x saves              0.043                      0.046
                           (0.22)                     (0.24)

Relief x innings           -0.003                     -0.003
pitched                    (1.10)                     (1.43)

Relief x (career            0.003                     -0.002
innings                    (0.98)                     (1.17)

First time                 -0.025                      0.015
eligible                   (0.30)                     (0.19)

Multiyear contract          0.152                      0.030
                           (0.60)                     (0.35)

Filed                       0.134                         --
                         (2.16)**                         --

Exchanged                      --                     -0.028
                               --                     (0.82)

Observations                  207                        236

[R.sup.2]                    0.88                       0.88

Notes: Robust t statistics in parentheses. Standard errors
clustered at the team level used in the calculations. All
regressions include year dummy variables and interactions
between the treatment dummy variable and the demeaned
independent variables listed in Table 2. See text for

* Significant at 10%; ** significant at 5%; ***
significant at 1%.

As Table 1 suggests, there may be a salary premium for those players who proceed through each stage of the FOSA process. Table 3 presents estimates of the average treatment effect for hitters who move to each subsequent stage. Focusing on Model 1, Table 3 indicates that there is a positive and highly significant gain for hitters who file for arbitration but do not exchange salaries with their clubs. Those who file and do not exchange offers increase their salary by 25%. (12) Model 2 indicates that those who move from filing to exchanging offers receive a salary premium of 7%, which is statistically significant at the 5% level. Therefore, the conclusion presented by these results is that filing for arbitration substantially increases a player's salary when compared to simply being eligible for it. Furthermore, exchanging final offers does provide an additional increase in salary. It is not nearly as large, however, as the initial increase from filing for arbitration.

Table 4 presents the treatment effects for pitchers. Focusing on Model 1, the results show that those pitchers who move from eligibility to filing and not exchanging offers increase their salary by 14%. This treatment effect is statistically significant: however, it is much smaller in magnitude when compared with the results in Table 3 for hitters. Results in Model 2 show that there is a negative and statistically insignificant relationship for moving from filing to exchanging offers. Combined, these results imply that there is a premium for filing for arbitration as opposed to being eligible, and there is no salary premium for exchanging offers once a pitcher has already filed for arbitration.

The results in Table 4 are different from those found in Table 3 for hitters. Hitters receive significant salary gains when moving to each stage of the arbitration process, and the gain is larger when filing for arbitration. Pitchers, on the other hand, only receive a gain in salary for filing, and this gain is approximately 11 percentage points lower than that for hitters. This finding suggests that hitters and pitchers are treated differently during the negotiation process.

Overall, the empirical results suggest that both hitters and pitchers can improve their salaries by filing for arbitration rather than just being eligible to do so. It is possible that filing for arbitration spurs owners into more good faith negotiations through the threat of possibly moving forward to the final stage of the arbitration process, which is costly.[degrees] This result is somewhat surprising. Why would players not file for arbitration, a relatively costless process, if it meant an increase in salary? Previous studies have not analyzed separately the filing phase of the FOSA process. Perhaps the adjustment for selectivity bias used here is not accounting for all of the potential selection. Instead, the result could be capturing the effect of better players filing for arbitration while lesser players do not. However, different sets of explanatory variables and different methods for accounting for the potential selection into each stage of the arbitration process have been used. The quantitative and qualitative results are similar and available upon request. Therefore, the results presented in Tables 3 and 4 are robust to different empirical specifications.

A potential explanation for the large salary premium associated with filing and the small/no premium associated with exchanging offers for hitters/pitchers, respectively, is that there is only a short amount of time between these stages, typically 3 days. Perhaps the rush to finish negotiations before the official exchange of offers, threat of arbitration, and the cost incurred by such, are responsible for the larger bump in salary for filing as opposed to exchanging. Once offers are exchanged, valuable information is then available for all participants in the process to see and analyze. There is a longer timeframe in which to continue negotiations before arbitration hearings are held, typically 2 weeks to a month.

The size of the salary increases in the filing stage and smaller and/or lack of significant increases for the exchange phase suggest that the process is starting to push agreement to resolution much earlier. Ashenfelter and Dahl (2005) analyze settlements determined by final-offer arbitration for wages of police and firefighters in New Jersey. The authors conclude that individuals who use experts during the arbitration process fair better than those who do not, and knowledge of this benefit develops over time. The extensive use of agents by players during the negotiation process in MLB and the historic institutional knowledge created over the years may be finally creating the intended result of the process design.

The difference in the size of the salary increases for hitters versus pitchers is not surprising. Pitchers are more susceptible to career shortening/ending shoulder or arm injuries. This risk creates greater variability in their performances and lessens the expected value for the future from good performances in the past. Miller (2000a) also found differences in risk-assuming behavior by negotiators for position players versus starting pitchers.


The number of salaries actually determined by arbitration in MLB has declined in recent years, suggesting that there has been a shift in the use of the FOSA system even though there has been no change in the operation of the system. Because of this, the focus of this paper is on the salary determination process at the three initial stages of the FOSA process. Treatment effects are estimated for filing for arbitration and exchanging offers. Results indicate that, among hitters who become eligible for arbitration, those who file for arbitration receive a wage premium of 25%; the exchange of offers increases salaries by 7%. For pitchers, there is a salary premium of 14% for filing for arbitration and no premium associated with exchanging offers. The difference in findings for hitters and pitchers suggests that they are treated differently during the negotiation process.

This paper methodologically contributes to the literature in a number of ways. This is the first paper on MLB to use information from the most recent seasons, 2007 to 2010, a time period in which salary negotiation cases that were ultimately determined by arbitration were at historic lows. Additionally, the analysis makes use of statistics on service time as opposed to experience. This is important since the CBA states that it is service time, as opposed to experience, that dictates eligibility for arbitration. Finally, this is the first paper to focus explicitly on the salary determination process of players in each of the first three stages of the arbitration process: being eligible for arbitration; filing for arbitration and not exchanging offers; and exchanging offers and not continuing through to arbitration.

The results of the analysis offer interesting conclusions that the previous literature has not shown. There is a salary premium for both hitters and pitchers for filing for arbitration; however, there is only a premium for exchanging offers for hitters, and this premium is substantially smaller than that for filing for arbitration. The previous literature has not shown this. This particular result is similar to the literature on free agency, which has shown that the threat of free agency seems to increase salaries, while going to the free agent market does not.

Given the rise in the use of final-offer arbitration, particularly in the public sector, the results from this analysis may shed some light on the construct of the final-offer arbitration processes. The analysis does not necessarily give insight into the evolution of the FOSA system used in MLB. However, the limited number of cases that proceed to the arbitration stage in the time period of this study and the finding that players benefit significantly, on average, from filing for arbitration suggests that the design of this system may be finally achieving the overall goal of final-offer arbitration to encourage cooperative bargaining.


Ashenfelter, 0., and G. B. Dahl. "Strategic Behavior, Self-Serving Biases, and the Role of Expert Agents: An Empirical Study of Final-Offer Arbitration." National Bureau of Economic Research, Inc., NBER Working Papers: 11189,2005.

Brown, D. T., and C. R. Link. "Final Offer Arbitration in Major League Baseball: An Empirical Analysis of Bargaining Failure.- Unpublished Manuscript, University of Delaware, 2010.

Burgess, P. L., and D. R. Marburger. "Do Negotiated and Arbitrated Salaries Differ Under Final-Offer Arbitration?" Industrial and Labor Relations Review, 46, 1993, 548-58.

Farber, H. S. "An Analysis of Final Offer Arbitration." Journal of Conflict Resolution, 24(4). 1980, 683-705.

Farmer, A., P. Pecorin, and V. Stango. "The Causes of Bargaining Failure: Evidence from Major League Baseball." Journal of Law and Economics, 47, 2004, 543-68.

Faurot, D. J., and S. McAllister. "Salary Arbitration and Pre-Arbitration Negotiation in Major League Baseball." Industrial and Labor Relations Review, 45, 1992, 697-710.

Fizel, J. "Bias in Salary Arbitration: The Case of Major League Baseball." Applied Economics, 28(2), 1996, 255-65.

Hadley, L., and J. Ruggiero. "Final-Offer Arbitration in Major League Baseball: A Nonparametric Analysis." Annals of Operations Research. 145, 2006, 201-09.

Marburger, D. R. "Arbitrator Compromise in Final Offer Arbitration: Evidence from Major League Baseball." Economic Inquiry, 42, 2004, 60-68.

Marburger, D. R.. and P. L. Burgess. "Can Prior Offers and Arbitration Outcomes Be Used to Predict the Winners of Subsequent Final-Offer Arbitration Cases?" Southern Economic Journal, 71(1), 2004, 93-102.

Marburger. D. R., and J. F. Scoggins. "Risk and Final Offer Arbitration Usage Rates: Evidence from Major League Baseball." Journal of Labor Research, 17(4). 1996, 735-45.

Miller. P. "An Analysis of Final Offers Chosen in Baseball's Arbitration System: The Effect of Pre-Arbitration Negotiation on the Choice of Final Offers." Journal of Sports Economics. I. 2000a, 39-55.

----. "A Theoretical and Empirical Comparison of Free Agent and Arbitration-Eligible Salaries Negotiated in Major League Baseball." Southern Economic Journal, 67(1), 2000b, 87-104.

Wooldridge, J. M. Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press, 2002.

Hill: Department of Economics, Central Michigan University, Mount Pleasant, MI 48859. Phone 989-774-3706, Fax 989-774-2040, E-mail hill

Jolly: Department of Economics, Marquette University, PO Box 1881. Milwaukee, WI 53201-1881. Phone 414288-7576, Fax 414-288-5757, E-mail nicholas.jolly@


CBA: Collective Bargaining Agreement

FOSA: Final-Offer Salary Arbitration

MLB: Major League Baseball

(1.) Beginning with 2007, an Internet website (Cot's Contract website at made available a list of players who were eligible for arbitration based on service time.

(2.) Players who are eligible for free agency and offered salary arbitration by their team are not considered here.

(3.) The CBA defines one day of service as each day a player is on a team's active roster. It takes 172 days to get one service year. The days begin with the first regularly scheduled game in a season and conclude with the last regularly scheduled game in a season.

(4.) Before the 1985 CBA, players used to be eligible after two years of service time.

(5.) Super-twos were added to the 1990 CBA.

(6.) The use of a three-member panel rather than a single arbitrator began with the 1997 CBA. The arbitration panel can only consider the following six criteria when rendering a decision: (1) the quality of the player's contribution to his team in the past season, including performance, leadership, and public appeal; (2) the length and consistency of the player's career performance; (3) the record of the player's past compensation; (4) comparative baseball salaries; (5) mental or physical player defects; and (6) recent performance by the club, including league standing and attendance.

(7.) The authors thank an anonymous referee for the points made in this paragraph.

(8.) The salary data comes from the USA Today online database found at baseball/mlb/salaries/team. Most data on service time and arbitration eligibility come from the Cot's Contract website at Some biographical data, arbitration eligibility data, and service time data on players are from the Baseball Reference database at Performance data and some biographical data come from the Baseball Almanac at

(9.) Although this paper does not concentrate on arbitrated salaries, the descriptive statistics are provided here for completeness. Only 22 players proceed through the arbitration stage. Therefore, any econometrics performed on this subgroup would be very imprecise.

(10.) These are not shown but can be found in Miller (2000a), 42.

(11.) See Wooldridge (2002) for a complete discussion of this model,

(12.) The exact percentage changes come from the formula [e.sup.[beta]]--1.

(13.) The authors thank an anonymous referee for making this suggestion.
COPYRIGHT 2014 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hill, J. Richard; Jolly, Nicholas A.
Publication:Contemporary Economic Policy
Article Type:Report
Geographic Code:1USA
Date:Apr 1, 2014
Previous Article:Estimating productivity growth in the korean economy without restrictive assumptions.
Next Article:Labor market impacts of smoking regulations on the restaurant industry.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters