Frequently Asked Questions About the Baseball Labor Negotiations

by David Grabiner

Last modified:

11/23/11: New 2011 agreement (A1, A2, minor changes throughout), NBA union decertifies (C3)
11/14/11: No changes
10/14/11: No changes

This is an attempt to get quick answers to some common questions, to reduce repetitive discussion from newcomers who often don't understand the issues, and to provide single, authoritative answers so that doesn't get overwhelmed with several people making the same point, whether correct or incorrect.

Constructive comments, corrections, and suggested additions are welcome; please send them to me at In particular, I would appreciate suggestions from people who have some expertise. I have left square brackets in several sections; comments there would be particularly appreciated.

If you want to comment on an economic or legal point, make sure that you understand the underlying economics or law first.

You can get a current copy of this FAQ in the Usenet group, or from

Table of Contents

A. Salary cap and tax plans

A1. What are the terms of the new agreement?

The new agreement was announced on November 22, 2011. As in the 2006 negotiations but in contrast to all previous negotiations, neither side publicly announced its proposals. The contract will run through December 1, 2016; thus, if the contract expires in 2016 with no agreement, players will have some time to sign contracts in the 2016-2017 off-season before a lockout is possible.

The luxury tax will be imposed on the portion of payrolls over $178M in 2012 and 2013, $189M in 2014-2016. The tax rate depends on the number of times a team has gone over the cap, including times under previous agreements. In 2012, the tax rate will be 20% for the first time over, 30% for the second, and 42.5% for the third or subsequent time. In 2013-2016, it will be 17.5% for the first time (or for any team which did not go over the the limit in the previous year), 30% for the second time, 40% for the third, and 50% for the fourth or subsequent time.

The Yankees were the only team to pay the tax in 2003, with a payroll $66M over the limit and thus a tax of $11M. In 2004, the Yankees were $83M over the limit and paid a tax of $25M; the Red Sox were $14M over, and the Angels were $4M over, both paying small taxes. In 2005, the Yankees paid $34M at a 40% rate, and the Red Sox paid $4M at a 30% rate. In 2006, the Yankees paid $26M at a 40% rate, and the Red Sox paid $500K at a 40% rate. In 2007, the Yankees paid $24M at a 40% rate, and the Red Sox paid $6M at a 40% rate. In 2008, the Yankees paid $27M at a 40% rate, and the Tigers paid $1.3M at a 22.5% rate. In 2009, the Yankees paid $26M at a 40% rate. In 2010, the Yankees paid $18M at a 40% rate and the Red Sox paid $1.5M at a 40% rate.

The net effect of the tax is that there is only a small incentive for teams not to go over the limit once while making a short-term push to win, but a strong incentive for teams not to go well over the limit for a long period.

Revenue sharing in the 2006 CBA was 31% of all local revenue, and no change was reported in the amount. However, there is a phase-out of eligiblility to receive the shared revenue. Teams in the 15 largest markets (with a waiver for Oakland while its ballpark situation is unresolved) will lose 25% of their shared revenue in 2013, 50% in 2014, 75% in 2015, and 100% in 2016. The money that these teams would receive goes back to all paying teams in 2013, and to all paying teams which are under the luxury tax threshold in 2014. The net effect should be a slight decrease in shared revenues, but also a strong incentive for teams in the larger markets with below-average revenues to increase their revenues; unlike small-market teams, the large-market teams will not have as much revenue-sharing money to lose.

Blood testing for Human Growth Hormone will be done during the off-season; there is no provision for random testing during the season but tests for cause may be made at any time. The number of off-season urine tests will increase.

There will be very strong taxes on amateur bonuses. In the amateur draft, each team has a different limit "based on 2011 total spending" [according to AP; the formula has not been disclosed yet]. The limit applies to all bonuses paid in the first ten rounds, and the amount over $100K of any bonus paid in round 11 or later. The tax rate is 75% for going up to 5% above the limit, and 100% for going more than 5% over. In addition, the team loses its next first-round pick for going more than 5% over, first- and second-round for 10%, two first-rounds for 15%. This makes 105% of the limit an effective cap; a team will probably prefer to leave this year's third-round pick unsigned rather than pay a bonus which will deny the team a first-round pick the next year. It will also decrease the value of extra picks; a team with three first-round picks is less likely to be able to pay normal signing bonuses to all three.

There will be a similar international limit for international signings. The limit will be $2.9M for 2012-2013 (the year is July-June), varying by won-lost record for 2013-2014, and may be changed by a study committee for 2014-2015. All bonuses count towards the limit except six exceptions per team of $50K or less and any bonus of $7500 or less. The tax rate is 75% for going up to 5% over the limit, and 100% for going more than 5% over. In addition, a team which goes more than 5% over can sign only one player for more than $500K the next year, a team which goes more than 10% over cannot sign any player for more than $500K, and a team which goes more than 15% over cannot sign any player for more than $250K. As with the draft, this creates an effective cap; a team is unlikely to offer a bonus which will severely limit its ability to sign players in the next year.

If a worldwide draft is not agreed by July 2014, the penalties will increase. A draft is another method for limiting compensation, as a player is more likely to accept a smaller bonus if only the team which drafted him is allowed to sign him.

There will also be a competitive balance lottery. The ten teams with the lowest revenue and the ten in the smallest markets will be entered into a weighted lottery for six draft picks following the first round; the teams which do not receive picks in the lottery and other teams receiving revenue sharing will be entered into a weighted lottery for six draft picks following the second round. These picks, unlike normal draft picks, may be traded.

There was no report on compensation for unsigned picks. In the 2006 agreement, for a first-round or second-round pick, the compensation pick will be in the same spot as the unsigned pick; for a third-round pick, the compensation pick will be following the third round. If the compensation pick is unsigned, there is no further compensation.

The draft-pick compensation for free-agent signings will be reduced in 2011 and restructured after 2011. In the 2011-2012 off-season, there will only be eight Type A free agents; the signing club loses its first pick in the next year's amateur draft as long as the pick is not one of the top 10 overall. Starting in 2012, in order to be eligible for compensation, a player must have been with the team the entire season, and the team must make an offer which is at least the average of the highest 125 contracts by annual value ($12.4M in 2011; this amount cannot decrease). This created a slight incentive not to trade a player in mid-season if he is eligible for free agency, as the receiving team will no longer get a draft pick if the player leaves.

Arbitration eligibility was changed from 17% to 22% of players with three years of service, effective in 2002.

The most visible change to fans will be the addition of a wild-card playoff game; each league will have two wild cards who play a one-game playoff before the Divisional Series. Since this is a scheduling issue, requiring players to play an extra game, it had to be negotiated with the union.

Many of the minor details have not been disclosed but are probably the same as in the previous CBAs. Payrolls are figured based on the average value of multi-year contracts for all players on the 40-man roster, earned bonuses, and benefits; as a result, the tax payrolls are higher than the officially reported payrolls. Team payrolls are based on the money actually being paid to players; if a team trades a high-salaried player away, along with $5M to help pay the player's salary, the $5M counts against the tax limit for the team paying it. In 2006, the tax money will be used for MLB-wide expenses: player benefits, an industry growth fund, and the development of players in countries lacking organized high-school baseball. For example, some money from the 2010 luxury tax was used to make payments to retired players who were ineligible for a pension.

A2. Why were the last three agreements reached with no strike?

Any negotiated agreement requires that both sides be willing to negotiate. If one side takes a position which the other side considers unreasonable, and refuses to compromise, there will be no agreement. Every labor negotiation before 2002, except for 1985, involved the owners taking such a position. In 2002, 2006, and 2011, both sides were willing to engage in reasonable negotiations.

In the 2002 negotiations, the main issues were increased revenue sharing and a luxury tax. Both are issues on which the union has been willing to negotiate, and the owners were willing to negotiate on the values. The 2006 bargaining on the luxury tax is a good illustration of succesful bargaining. The union did not want a tax to carry over to 2007 and thus proposed plans with no tax in 2006; the owners wanted a tax for every year of the agreement. The final compromise had a tax in 2006, but the tax would not carry over if the 2007 season was played under an expired agreement.

The 2006 and 2011 negotiations were carried out with no public negotiations, but the final deals in 2002 and 2006 have similar structures, and it does not appear that either side made requests on which the other side was unwilling to negotiate. In 2006, the union also had additional leverage from the lack of a 2007 tax if the agreement expired; the owners could only avoid this by negotiating a new contract, either in cooperative negotiations or after a lockout.

In contrast, in the 1994 negotiations, the owners proposed a salary cap which would have resulted in a 15% reduction in total salaries. The union opposes a salary cap on principle, and had no reason to believe that one was necessary; the owners' official figures showed that MLB had been profitable for nine consecutive years, and the union had access to the actual numbers which presumably showed higher profits. The owners refused to make any proposal other than the salary cap until November 17, 1994, three months into the strike, and that proposal had a luxury tax which was so strong that it would have been a virtual cap. With the owners making demands that the union could not accept, the strike continued until stopped by the court ruling that the owners had not been negotiating in good faith and could not unilaterally impose the cap. (In contrast, if negotiations had reached an impasse after the productive bargaining of August 2002, the owners probably could have imposed their last proposal, as it was the result of good-faith negotiations.)

The previous long strikes and lockouts were the result of similar issues. In the first strike, in 1972, the union requested that pensions and health-care benefits be increased to keep up with the increased costs; the money was already available because of a surplus in the pension fund. The owners refused to make the increased payments until after the union struck. In 1976, the union had won the right to free agency with the Messersmith-McNally decision, and the owners locked the players out in spring training while proposing that free agency be severely limited. In 1980 spring training and 1981 during the season, the owners proposed that a team signing a quality free agent could protect 15 players on its 40-man roster, and the team losing the free agent could take any of the others as compensation. The union saw this as destroying the value of free agents, and struck. In 1990, the owners first proposed a salary cap, and a lockout was already under way before the proposals changed.

While a strike or lockout could result from demands which were made by either side, it appears from the previous paragraph that the owners were making all of the rejected demands. The reason is that the union never demanded, nor received, any significant concessions in these negotiations. The union's most important gains were grievance arbitration in 1970 and salary arbitration in 1973, both obtained in negotiations without strikes; and free agency in 1975, which was the result of an arbitrator's ruling. The 1976 negotiations defined the terms of free agency (which was completely unrestricted under the 1975 ruling, although neither side expected that to remain), and the union's goal in subsequent negotiations was primarily to preserve the status quo. Even in 1972, the union's goal was essentially to get benefits to keep up with their costs.

The only short strike which actually affected games was in 1985. There was no issue seen as non-negotiable, with the main issue the delay of arbitration eligibility from two years to three. An agreement was reached after only two days. Similarly, on May 23, 1980, the union had announced a strike on the issue of free-agent compensation. When the owners and union agreed to create a study committee, postponing implementation for one year, the non-negotiable issue was gone, and the strike was settled after five hours. (The settlement was temporary, as the study committee was not able to reach a recommendation acceptable to both sides, leading to the strike in 1981.)

A3. What are the effects of a salary cap, tax, or revenue sharing?

A salary cap is an agreement which places an upper limit (and sometimes a lower limit) on the money each team can spend on player salaries. In the 1994 negotiations, the owners proposed to limit each team's salaries to 50% of average team revenues for the previous year; every team would be required to have salaries between 84% and 110% of that level. (Before the strike, the players got 58% of average team revenues, according to the owners' methodology; the actual reduction in salaries would be greater because salaries of players on the 40-man roster and incidental expenses such as meal money would be counted against the cap.)

A decision by an individual team to set a budget is not a cap. Several teams did this, publicly announcing their budgets, with no complaints of collusion.

For example, if the Tigers refuse to spend more than $80M on salaries, they can do that without a cap. If that means that they have $75M already allocated and there is a player (either one of their own players or a free agent who is interested in playing for them) who wants $10M, they have to do without that player. And if the Yankees think that the player is worth $10M to them and are willing to pay that, he will become a Yankee. The Tigers cannot do anything to stop this except for going over their budget. They might decide to do this; for example, if the player at stake is a hometown hero, he could produce a lot of revenue for the Tigers.

A salary cap would force every team to have the same budget. Thus, in the above example, say that the cap is $80M. Now the Tigers cannot pay more than $5M for the player. If the Yankees are already at or near the cap, they cannot make a better offer; if he receives no better offer, the player will probably sign with the Tigers for $5M. If the Yankees have $7M or more free under the cap, they can offer $7M to the player, and the Tigers will not be allowed to match the offer; now the player will sign with the Yankees at a reduced salary.

This is an essential feature of the salary cap; the Yankees and Tigers have an agreement which affects what the Yankees can pay for a Tigers player.

A very high tax rate would have the same effect as a cap. Consider the example in the section on the cap, but now assume that the Tigers are over the limit and subject to a 100% tax rate on any increase in their payroll. Thus, if the Tigers want to sign the player for $10M, they would pay an additional $10M in tax, for a total cost of $20M to make an offer of $10M. Meanwhile, if the Yankees are under the limit, they could offer $7M to the player at a cost of only $7M to the team. This would make it almost impossible for teams which are in the high tax range to compete for free agents, but not completely impossible as a cap would. If the Tigers really want to keep the player, they can offer $10M and pay an extra $10M in tax, but they would expect to lose a lot of money by doing this.

A lower tax rate will reduce salaries but not prevent teams over the limit from competing. For example, if the tax rate were 11% rather than 100%, the Tigers could offer $9M (paying taxes of $990K) for a player worth $10M. This would reduce the player's value to the Tigers by 10%. He might accept that offer to stay with the Tigers, either for non-economic reasons or because no team offered the $10M.

If the tax money is redistributed to small-payroll owners, or the tax is imposed on small-payroll owners and redistributed evenly, it may have the same effect on them. For example, if the Yankees are under the tax limit and will get 11% of the difference between their payroll and the league average, they will lose $990K in tax receipts if they sign the player for $9M. In this case, the player's value to the Yankees is also reduced by 10%. If every player's value to every team is reduced by 10%, salaries should drop by 10% across the board, and owners' profits should increase by the amount that salaries drop.

The effect on the players of a tax on revenue, or of revenue sharing, would be similar. If the Tigers pay a 10% tax on revenue or must share 10% of their revenue, then the player who adds $10M in revenue is only worth $9M, because the Tigers don't get to keep the other $1M. (The effect on the owners would be different, because owners with high revenues would pay a higher fixed sum with a tax on revenues than with a tax on payrolls. The revenue-tax plan would do more to help low-revenue teams make a profit and reduce high-revenue teams' profits.) This is why the union offered lower tax rates in plans which accepted the owners' revenue sharing plan in the previous labor negotiations.

This effect may have contributed to the decline in free-agent salaries in recent years. _Sports Business Journal_ reports that the average free agent in 2004 took a 26.6% pay cut. All teams paid revenue-sharing money, and the Yankees, who paid the most for free agents, were facing a 30% tax in 2004 and a 40% tax in 2005 and 2006, and continue to face it in 2007.

Note that the effect of all of these proposals depends on the *marginal* tax rate. A 5% payroll tax across the board and a 50% tax on the top payrolls might raise the same amount of revenue, but the 50% tax on the top payrolls might do more to reduce salaries because it would force teams down to the tax level.

A4. What is the status of contraction?

While contraction was ultimately postponed in 2003 until at least 2007, the threat of contraction in 2002 and 2003 was an important issue both in the labor negotiations and in negotiations among owners. The owners had the right to contract two teams in 2007 without the union filing a grievance, but they would still have to deal with lawsuits from other parties such as the cities with which they have signed stadium leases. There was no talk of contraction in 2006, and thus the agreement not to contract teams in the 2007-2011 agreement is only a minor issue; it was not even mentioned as an issue in 2011.

The owners announced on November 7, 2001 that two teams would be contracted before the 2002 season. The two teams were not announced, and the details had not been worked out. The owners claimed that they did not need the union's approval to contract two teams because the labor contract had not yet expired (it expired November 8); the union challenged this and asked for an arbitrator to rule on it. The owners and union tried to negotiate a compromise but failed; the arbitrator's ruling was postponed several times, and may have been put on hold pending the outcome of the labor negotiations. The labor agreement made the ruling moot.

In addition, the commission which owns the Metrodome obtained an injunction forcing the Twins to honor their lease for the 2002 season, which they had already exercised the option to renew; this injunction was upheld by the Minnesota Court of Appeals, and the Minnesota Supreme Court refused to hear the case. As a result, contraction was postponed to 2003. The Twins renewed their lease for 2003, and the stadium commission settled a lawsuit against MLB with an agreement that the Twins could not be contracted in 2003.

If two teams are eliminated, profits for the remaining teams will increase, because the national revenue will be divided among fewer teams, and the revenue-sharing money which would have gone to these teams is more than the shared revenue which they provided for other teams. However, the owners of the teams involved will need to be paid enough by the remaining teams to make the deal acceptable.

The two teams were believed to be the Expos and Twins. The Expos had no local buyer available; the owners probably wanted to either be bought out or move the team. (Instead, MLB took over the team.) The Twins had been discouraging potential buyers at the time contraction was under discussion, but Donald Watkins had been negotiating for purchase. It is probable that the Twins could receive more if they are contracted; if MLB contracts the Expos, it must contract a second team as well, and thus the Twins will have bargaining power and may receive more than their market value. Watkins himself said that he would pay market value, not contraction value, for the Twins. If Watkins had bought the Twins and refused to be contracted, some other team would have been contracted and the value of the Twins would have gone up.

Contraction was unlikely to happen in 2002 from the beginning; it was more likely to be an opening move in the new labor negotiations. The owners reportedly told the union in September that it was not feasible to contract for 2002; contracting after this promise could have been seen as failing to bargain in good faith, a violation of labor law.

The threat of contraction also risks the owners' goodwill. The senators from Minnesota threatened to introduce a bill removing MLB's anti-trust exemption; in the past, such bills have been threatened by senators and congressmen to ensure expansion to their home states. The Twins offered to pay employees several months' salary after their jobs were lost to contraction, in order to discourage them from leaving for more secure jobs. Businesses are less likely to buy season tickets, and players may be less likely to sign with teams that may not exist.

The Twins' 2002 attendance is a good illustration of the cost of goodwill; even though the Twins had a huge lead for most of the season, and their ticket prices had gone up at about the rate of inflation, they drew fewer fans than the 1987 and 1991 Twins which also won titles unexpectedly.

B. Steroids and other performance-enhancing substances

B1. What changes were made to the drug policy in reaction to the Mitchell report?

On April 11, 2008, the union and commissioner's office agreed to a revised drug policy, implementing many of the recommendations of the Mitchell report which were subject to collective bargaining. The commissioner's office ratifified the policy on May 15 and the union ratified it on May 23. MLB had already implemented several other recommendations on its own because they did not affect the union.

As part of the agreement, amnesty was given to all players named in the Mitchell report; the amnesty included Jay Gibbons and Jose Guillen, who were suspended just before the report came out but whose suspensions were appealed and put on hold pending negotiations.

The new drug policy will be enforced by an administrator who serves a three-year term and cannot be removed except by an arbitator; this serves the goal of keeping the testing independent. Negative test results will be maintained for two years. The administrator will also release anonymized information such as the number of positive tests for each substance and of theraputic-use exemptions. The administrator will also audit the testing procedures annually.

Penalties will remain the same as in the 2006 steroid policy. The names of players who are under investigation will not be released unless they are disciplined. The list of prohibited substances has been expanded, adding some drugs which reduce the side effects of steroid use.

The number of tests will be increased to 3600 per year, about three per player. There were 125 tests per year during the off-season through 2011; this will be increased to 200 in 2012-2013, 225 in 2013-2014, and 250 after that. In addition, the top 200 amateur prospects must agree to be tested in order to be eligible for the draft. Teams will be notified of positive tests but the players will not be directly penalized, as they were not under MLB jurisdiction.

Starting in 2009, MLB will require certification standards for strength and conditioning coaches, and starting in 2010, there will be guidelines to ensure that certified coaches are available to all players.

B2. What is the history of MLB's policies on steroids?

Baseball has long banned the use of illegal drugs and unauthorized use of prescription drugs. Steroids were officially listed as prohibited in 1991, but the drug policy was not part of the Basic Agreement at the time, and thus the commissioner's office had to negotiate procedures for individual cases with the union, both for steroids and for illegal drugs such as cocaine. An informal agreement allowed players to be tested under "reasonable cause", but the negotiation of reasonable cause allowed players considerable advance notice. No players tested positive for steroids under this testing.

In 2001, Commissioner Selig unilaterally implemented a drug testing program on the minor leagues; it only applied to players who were not on the 40-man roster of MLB clubs because such players were not subject to the CBA. Players were subject to up to three tests per year (later increased to four, including tests outside the season), and a 15-game suspension for a first positive test. Androstenedione (a steroid precursor which was still legal at the time) and human growth hormone were added to the list in 2002. When the MLB penalties were increased in 2005, the minor-league penalties became the same.

The 2003-2006 CBA was the first agreement to include a steroid policy. For 2003, testing would be anonymous "survey testing"; the names of players testing positive would not be released, and if at least 5% of the steroid tests were positive, mandatory random tests would begin in 2004. First-time offenders in 2004 would receive counseling but would not be suspended and their names would not be made public. Even the weak penalties may have had some deterrent effect, as only 12 out of 1133 tests in 2004 were positive; for comparison, in 2003, 96 tests out of 1438 were positive for steroids or elevated testosterone levels, and 8 others tested positive for substances not banned at the time. Under the terms of the agreement, androstenedione was automatically added to the list of banned drugs when Congress declared it a controlled substance in 2004.

The Government seized the 2003 tests in 2004 as part of the BALCO investigation. As a result, MLB and the union suspended testing in 2004 for players who had tested positive in the past until they were notified, because there was concern that tests which should have been confidential would be released. The 2003 tests actually remained confidential until 2009. At that time, the Government wanted to release BALCO information about Bonds' tests in 2003 as part of his trial; a court has not yet ruled on the release. Six other names have also been released by the media from unnamed sources. An appeals court panel ruled that the seizure of the entire list of positive tests (as opposed to just the tests for the ten players named on the warrant) was improper, but by the time of the ruling, the list had already been available and had led to these leaks. The full appeals court confirmed the ruling.

MLB and the MLBPA could not comment on individual cases because of the confidentiality, but both have cautioned about drawing conclusions from the list, because some of the tests were contested (and many more would have been contested at the time if there had been any consequences), and some positive tests were not violations of the policy in effect at the time.

On January 13, 2005, the commissioner and union agreed to a stronger steroid-testing policy. Players were tested randomly, both during the season and in the off-season, with at least one test during the season. The substances banned included anabolic steroids (including "designer steroids" such as THG), steroid precursors such as androstenedione, and masking agents and diuretics. Amphetamines were not banned. A player testing positive was suspended without pay for 10 days for a first offense, 30 days for a second offense, 60 days for a third offense, and one year for a fourth offense. In addition, the positive tests were made public; first-time tests under the previous agreement were not made public. It was initially announced that players could not defer the penalty during an appeal, but that was apparently not enforced; Rafael Palmeiro was allowed to appeal his suspension and served a suspension in August for a positive test in May.

On November 15, 2005, the commissioner and union agreed to a further revision of the steroid-testing policy; it was maintained in the 2007-2011 CBA, and in the 2012-2016 CBA with the addition of blood testing for HGH. This policy was reached under threat of Congressional legislation which would have imposed even harsher penalties. A player testing positive for steroids will be suspended without pay for 50 days for a first offense, 100 days for a second offense, and for life for a third offense. Positive tests under the previous steroid policy will not count as prior offenses. A player banned for life may petition for reinstatement after two years, and this review will be subject to arbitration. The union yielded to the commissioner's proposal on the length of suspensions, in return for the right of arbitration. Players will also be tested for amphetamines, with a first offense leading only to more frequent additional testing, and suspensions for 25 days for a second offense and 80 days for a third offense. A first-offense positive for amphetamines is supposed to be kept confidential, although it apparently leaked that Barry Bonds tested positive. Players will be tested during spring training and randomly during the year, with at least one test during the season.

If there was no new labor contract by August 1, 2006, the union had the right to consider the 2006 steroid policy to end with that labor agreement, on December 19, 2006. If that had happened and the 2007 season were played under the expired agreement, the previous steroid-testing policy, which applied in the 2005 season, would have applied until a new labor contract is reached. (While the union had the right to take that step, it was unlikely; there would be a significant cost in good will, and if the union wanted to change the steroid policy, its main interest would be in getting a long-term change in the new labor contract.)

It is rare for labor and management to agree to a change to an existing labor agreement, but the January 2005 steroid agreement was seen by both sides as in their interest. At that time, leaked grand jury testimony in the BALCO case reported that several MLB stars were using steroids; the union and owners both decided that it would be in baseball's interest to develop a new policy. Similarly, the threat of Congressional action convinced both labor and management to establish the November 2005 steroid policy, as otherwise Congress could have imposed its own regulations. There was no official threat of action in April 2008, but the Mitchell report had significant credibility, and both sides had an interest in implementing its recommendation.

B3. What are the conclusions and recommendations of the Mitchell report?

This section is based on the Mitchell report at Information from the report is presented here with little comment. The union did not file an official response, as it had little advance notice of the release of the report.

The most notable conclusion is that steroid use is a serious problem, and that it became serious because everyone involved allowed it to continue as a serious problem rather than taking action. The purpose of the report is to obtain closure on the past, and to make recommendations for the future.

The magnitude of the problem is indicated by the list of current and former MLB players who are linked to steroid use, along with the evidence of use by each player. Different sources count the number of players as 85 to 90, depending on which players are considered "linked".

Many of the players involved have direct links, including their own testimony, written prescriptions, checks written to to Kirk Radomski, Radomski's phone records, and documents provided in law enforcement investigations. However, the list of players includes some players for which the only evidence appears to be uncorroborated accusations. For example, Larry Bigbie, who admitted to using steroids and cooperated with a federal investigation, reports that two players told him they used steroids but does not say that he observed either one. (One of those two players, Brian Roberts, later admitted to having tried steroids once.) Other players are listed based on news reports.

The report also discusses the effectiveness of the current policies and makes further recommendations; the following discussion is based on Section X of the report. The report quotes the established principles of an effective drug program, from a report at the Duke Conference on Doping at Duke's Center for Sports Law and Policy: "independence of the program administrator; transparency and accountability; effective, year-round, unannounced testing; adherence to best practices as they develop; due process for athletes; adequate funding; and a robust education program."

Independence was improved in the 2006 program because it had an independent adminstrator rather than being administered jointly through the commissioner's office and the union. The administrator was still not independent until the 2008 program, as he could be removed by either side for any reason.

Transparency was a failure of the 2006 program. There were no aggregate reports, and they have not been audited. The data which Mitchell needed to compile aggregate reports had been destroyed, so it is not possible to determine what happened to most negative tests, or how well integrity and chain-of-custody procedures work. In addition, current best practices for testing depend on variations between the results of tests taken over time; a player with a naturally high testosterone level could prove the natural level with previous tests which were negative but just below the limit, or which were above the current limit but below the limit in effect at the time of those tests. The 2008 program provides for release of aggregate reports and retention of negative tests.

There are unannounced tests during the season, but given the difficulty of locating players during the off-season, there are fewer off-season tests (68 tests in 2006) and players may have up to three days notice before a test. The 2008 program allowed for 125 tests a year, further increased in the 2012-2016 CBA.

There are several failures in best practices. Diuretics and other masking agents, which are forbidden by the World Anti-Doping Agency, are not explicitly forbidden in MLB; the burden of proof is on the independent program administrator to show that the player took the masking substance with the intent to avoid accurate testing. Human growth hormone is not detectable in current urine tests, and blood tests are not allowed. Players were reported to have used human growth hormone rather than steroids specifically because there is no current test being used. This was not changed in 2008 but tests were added in 2012.

Mitchell was unable to investigate due process but has no reason to believe that the rights of players were not protected. Funding was also adequate. Education was provided but was apparently not adequately emphasized. The 2008 program does increase education.

The primary recommendations are given in Section XI of the report; many of them were put into effect quickly, although some are subject to collective bargaining.

The commissioner's office should have an independent department of investigations. The department should be headed by an official with law enforcement experience who can cooperate with law enforcement agencies. The deparemnt should interview players when it receives allegations (from law enforcement, for example) that they are involved with performance-enhancing substances. The 2008 program made the administrator independent throughout his term, although not a fully independent third party.

There is already a policy requiring clubs to report evidence of drug use, but it is not well publicized and has not been enforced. The policy should be enforced; all club officials, except for physicians and trainers for whom the information would be protected under patient confidentiality, must certify that they have reported all evidence, and the commissioner should have power to enforce the rule under the "best interests of the game" clause. MLB has announced that it will implement this rule, requiring team officials to sign a statement that they have no undisclosed knowledge.

Packages received for players at ballparks should be logged. The contents would not be recorded, but information about the sender and recipient would be available if the sender were subsequently linked to steroid distribution. MLB has adopted this rule.

Several deterrent provisions do not require collective bargaining approval because they do not directly affect the union. Clubhouse personnel should be subject to background checks and drug testing. An anoymous tip line should be available. MLB has done both of these. In addition, the top 100 prospects should be tested before the major league draft; the 2008 program includes provisions for testing the top 200 prospects.

Education about gambling and player safety have received much more emphasis than education about steroids; it is recommended that the emphasis on all three be comparable. An educational video on the use of performance-enhancing substances is shown to all players in spring training, and there are educational meetings. However, some players do not remember these programs, and others criticize the shallowness. There should be testimonials and presentations from physicians and law enforcement, and information about the risks of illegal purchases. The report gives examples of several effective presenters. The rules against gambling are posted in every major league clubhouse; the steroid policy and information about the risks should also be displayed in clubhouses and training rooms.

B4. What penalties have been imposed on players?

The report does not recommend that any players be disciplined for past violations of the policies, including players named in the report, except when the commissioner determines that "the conduct is so serious that discipline is necessary to maintain the integrity of the game." The primary reason is that the report's purpose is closure, and continued proceedings against the players named will emphasize the past problems. In addition, there is little cause for penalty, because the incidents are several years old, and would have to be penalized under the rules in effect at the time; the 2006 penalties cannot be imposed for violations in 2004.

The 2008 agreement included an amnesty for all players named in the report. Commissioner Selig originally announced that discipline for players in the report would be on a case-by-case basis, but no player was investigated based on the Mitchell report, making any discipline unlikely. Any disciplinary action based on the report would be subject to the usual procedure for union grievances. The only player interviewed after the report came out was Paul Byrd, who admitted before the report came out to using human growth hormone; he was also not disciplined.

The amnesty does not apply to club personnel such as managers and trainers; Commissioner Selig announced that they would not be fined or suspended but could be ordered to do community service. No club personnel have been penalized yet.

Jay Gibbons and Jose Guillen were each suspended for 15 days on December 6, 2007, which is before the Mitchell report came out. Both players were suspended for receiving shipments of human growth hormone in 2005 (and in previous years, but HGH was not banned in baseball in the previous years); they were also named in the Mitchell report. The union filed grievances against both suspensions, and the 2008 amnesty rescinded the suspensions before they took effect.

Three players were suspended for 50 games following positive tests after the Mitchell report, two players were suspended for 50 games for positive tests in 2009, and two in 2010. Manny Ramirez is the first player to be suspended 100 games for a second offense; he was suspended for 50 games in 2009 for having a prescription for HCG, which is not itself a steroid but which is banned by MLB because it is often used to counteract the effects of steroids and then tested positive in 2011. (He chose to retire rather than come back after the suspension.) Eliezer Alfonzo was the second player suspended for a second offense, also in 2011. Danny Gutierrez was suspended for 50 games in 2010 for testing positive for an amphetamine; he took it under a prescription and could have applied for a therapeutic use exemption but could not be granted one retroactively after a positive test.

Prior to the Mitchell report, thirteen major-league players tested positive in 2005, receiving 10-day suspensions. Three players tested positive in 2006 and four in 2007, receiving 50-day suspensions. In addition, Jason Grimsley was suspended for 50 days after receiving a package of human growth hormone and admitting in a Federal investigation that he had used it; the Diamondbacks also released him. The drug policy did not specifically provide for this suspension, so it was imposed under the "best interests of baseball" clause. The union filed a grievance, and a negotiated settlement established that the 50-game suspension was acceptable but that, given the established penalty, the team could not further penalize the player by terminating his guaranteed contract. (As part of the settlement, Grimsley received the remainder of his salary but donated it to charity.)

Miguel Tejada pleaded guilty to a crime, Barry Bonds was convicted in court, and Roger Clemens was charged with a mistrial and a retrial scheduled for April 17, 2012. None of the three were criminally investigated for using or buying steroids; rather, Tejada pleaded guilty to lying to Congress in an investigation (about another player's use), Bonds was convicted of obstruction of justice for giving evasive grand jury testimony in the BALCO case and the jury deadlocked on perjury charges for lying about his steroid use, and Clemens was charged with lying about his own steroid use under oath. In general, as the Mitchell report points out, criminal charges are brought against the manufacturers and distributors of steroids, not the purchasers.

It was never likely that any active players would be punished by MLB as a result of the BALCO case. There was some discussion of the Yankees threatening to void Jason Giambi's contract. However, the union would be able to challenge such a move, forcing the Yankees to demonstrate to an arbitrator that Giambi violated the contract. Victor Conte, the director of BALCO, pleaded guilty rather than allowing a trial, so there will be no open court testimony of MLB players' steroid use from that trial. Thus the Yankees' evidence could not be that Giambi admitted using steroids, only that the San Francisco Chronicle said that Giambi admitted using steroids.

US immigration authorities have discussed denying work visas to foreign players named in the Mitchell report, but no action was taken and these players were allowed to report to spring training.

B5. Who provided information to Mitchell? Why did few players provide information?

George Mitchell led the steroid investigation on behalf of baseball. He has no subpoena power and thus cannot force anyone to testify. He did warn that a lack of cooperation would "significantly increase" the chance that Congress will insist on its own investigation. In fact, after the Mitchell report, Congress launched an investigation, and compelled several players and former players named in the report to testify.

The only player required to testify before Mitchell was Jason Giambi, because the commissioner's office and union negotiated a plan for him to testify under threat of suspension. However, since Giambi only admitted to using steroids before the current policy came into effect, he might not be subject to suspension for using them. He agreed to testify only about his own use, not naming other players, in order to avoid a legal battle with the commissioner's office; the commissioner announced that Giambi would not be disciplined. Several other people, such as clubhouse attendant Kirk Radomski, agreed to testify before Mitchell as part of agreements with law enforcement.

Most of the people interviewed were employees of the teams and the commissioner's office, former team employees, and former players. The only active player who testified voluntarily was Frank Thomas. The teams and commissioner's office also provided a large number of documents; other information came from public records such as legal investigations.

Players who could be subject to legal cases will be unlikely to cooperate, as they do not want to say anything at MLB hearings that could then be used against them in court. This applies directly to Barry Bonds, who was under investigation. But the fear of discipline by MLB is just as significant; anything that a player said in the Mitchell report could be used against him, or against other players, in a disciplinary hearing. Mitchell did recommend against punishing players because of his report, but Selig was not bound by that recommendation and did not give any advance indication that he would follow it. (Even after the report came out, Selig said that punishments would be handled on a case-by-case basis.)

The union agreed to advise players who had questions, and recommended recommending that players get legal advice. It did not make a blanket recommendation that players should or should not cooperate, but the memo gave details of the union's reasons to discourage active players from cooperating. The union's memo said, "Commissioner Selig has not ruled out disciplining...players as a result of information gathered by the Mitchell investigation. Therefore, you should be aware that any information provided could lead to discipline of you and/or others." The memo also warned that Mitchell could not promise complete confidentiality; any statements which Mitchell agreed to keep confidential could be subpoenaed in a legal case, which would result in the statements becoming public and might cause them to be used against the player. The union's final warning was, "Any comments made to Senator Mitchell -- Commissioner Selig's lawyer -- by an individual player regarding the operation of the [negotiated steroid] Program might well be used by the owners in future bargaining with the union."

The union had legitimate reasons to be concerned about confidentiality. In the BALCO investigation, federal agents seized data from two firms which had conducted the survey tests in 2003; the seized information identified players who had tested positive in 2003. The commissioner's office and union agreed to a moratorium on testing, and the union notified the players who had tested positive. A legal case seeking the return of the seized material was still ongoing as of the time of the investigation, and some of the positive tests did get released in 2009.

Independent of legal issues, players have privacy concerns. They have an interest in protecting their own names from public allegations of steroid use (whether true or not), and the union has made several attempts to protect these rights. For example, the union denied Mitchell's request for anonymized medical records, because of concerns that the records would not remain anonymous because information such as age, height, weight, and blood type would identify the player.

The Associated Press and Hearst Corporation requested the official affidavits used in searches of former pitcher Jason Grimsley and former Mets clubhouse attendant Kirk Radomski, both of whom reportedly named multiple MLB players who used steroids. The news organizations claimed that the government could not selectively disclose records to Mitchell and not to the news organizations; the US attorney's office opposed disclosure of the names, saying that the names were blacked out in the disclosure to Mitchell. The union joined the cases on the government's side, opposing disclosure, and the documents were not disclosed at the time they were requested. However, following the Mitchell report and the winding down of the criminal cases, the affidavits no longer needed to be sealed to protect an ongoing investigation, and the government released them on December 19 and 20, 2007. The Grimsley affidavit named four players who were not mentioned in the Mitchell report, and the Radomski affidavit named two more.

B6. Which players admitted using steroids after the Mitchell report came out?

Andy Pettite admitted using human growth hormone in open Congressional testimony. Chuck Knoblauch admitted using performance-enhancing drugs in a recorded interview with the Congressional committee. These are the only players to have reported their own use under oath. Brian McNamee, testified before Congress that he injected Clemens with steroids, but Clemens denied it. Pettite also accused Clemens, but only based on an admission which he heard from Clemens; Clemens claims Pettite misunderstood the statement.

Immediately after the report came out, several players named in it voluntarily admitted their use. Gary Bennett, Brian Roberts, and F.P. Santangelo all admitted their use. Fernando Vina, who was named in the report as having used both HGH and steroids, admitted using HGH but denied using steroids.

Matt Herges and Glenallen Hill later admitted their use in statements which were made public on the same day as the Congressional testimony.

John Rocker admitted to failing a steroid test in 2000, and reports that Comissioner Selig knew about the failed test. However, under the policy at the time, he was not subject to discipline, and the test was required to be kept confidential under the terms of the Employee Assistance Program. (Rocker was under suspension at the time for making insensitive racial and ethnic remarks.) Similarly, Alex Rodriguez admitted using steroids in 2001-2003 after reports of his 2003 positive test were leaked; he would not have been subject to penalty under the policy at the time.

C. Anti-trust exemption

C1. Why does baseball have an anti-trust exemption?

It is not directly written into the law; it is the result of a Supreme Court decision.

The Federal League, which played as a rival major league in 1914-1915, filed an anti-trust suit against MLB. In 1922, the Supreme Court ruled for MLB, on the basis that MLB was not interstate commerce and thus was not subject to federal anti-trust laws.

In later rulings, the Supreme Court has called the 1922 decision "an anomaly", but has let it stand as a precedent, saying that it is Congress's responsibility to overturn the exemption. Bills to overturn the exemption have frequently been introduced in Congress, but they did not make it out of committee. The Court's interpretation of this action was that Congress intends to keep the exemption.

In the 1996 labor agreement, players and owners agreed to ask Congress to overturn the anti-trust exemption with respect to labor relations in major-league baseball. This bill was signed on October 27, 1998. This also effectively enshrines in law the ruling that baseball is exempt from anti-trust laws in other areas.

C2. Are other sports also exempt?

They are not legally exempt; the Supreme Court has ruled that the 1922 decision applies only to baseball.

However, it is legal to agree to terms in a labor contract which would normally be in violation of anti-trust law, provided that the contract was obtained in fair collective bargaining. For example, if the anti-trust exemption were repealed, MLB and the players' union could still agree to maintain the current system of free agency and arbitration. However, it might not be binding on minor-league players, who are not union members. [Zimbalist claims that it wouldn't be; do any labor experts have an opinion?]

As an illustration of the labor exemption, the NFL and NBA drafts are legal even though they prevent drafted players from offering their services to multiple teams. However, Maurice Clarett temporarily won a court challenge in 2004 against the NFL's attempt to exclude him from the draft. The court ruled that he was not a party to the union agreement which included the draft (since he was not being allowed to join the union), and thus anti-trust law applied. (The ruling was overturned on appeal.)

By act of Congress, all sports leagues are exempt from anti-trust in their negotiation of national broadcasting contracts. This allows the leagues to negotiate internal restrictions, such as baseball's rule that The Baseball Network had exclusive rights to all games on its date. Such an arrangement would also be legal in the NFL or NBA; an NBA restriction on superstation broadcasts was upheld by the Seventh Circuit in September 1996.

C3. What are the effects of the anti-trust exemption?

The owners in MLB have full control over franchise movement and probably more control over ownership. When the NFL tried to block the Oakland Raiders from moving to Los Angeles in 1982, team owner Al Davis won a suit against the NFL for restraint of trade. Without the anti-trust exemption, a team in MLB could probably have freely moved to Washington, which is further from Baltimore than Los Angeles was from the existing team in Anaheim; instead, it was MLB's decision when to allow the Expos to move to Washington, and to negotiate compensation to the Orioles for their territorial rights. (I have no idea of the anti-trust status of MLB taking over the Expos; a court might have ruled under anti-trust law that the team had to be sold as soon as bidders were available.) Likewise, when Nintendo wanted to buy a majority share in the Mariners, MLB forced the deal to be restructured because of a policy against foreign ownership; this might have been challenged under anti-trust law.

The exemption also allows the owners to create exclusive deals, even when they are anti-competitive measures. The right to such deals could also hurt a rival league (although it might not matter; the USFL won $1 in damages in an anti-trust case). MLB could sign a contract with ESPN which allowed ESPN to broadcast a certain number of its own baseball games, and no games from any other league during the MLB season. The rival league would then be unable to negotiate with ESPN.

The owners are also claiming that their reserve rules are protected by anti-trust law. That is, players who do not have signed contracts but were not eligible for free agency under the expired CBA may not sign with teams in a rival league; MLB can claim that the players are still under contract. Blacklisting players who move to the rival league would also be possible. For example, in the 1950's, several players signed with the new Mexican League. The Commissioner barred any Mexican League players from playing for MLB for five years, and this was upheld by the courts.

The minor-league agreements might also be forbidden by anti-trust law, because they bind a player who is not a member of the union to a single team's minor-league system.

An agreement which would normally be in violation of anti-trust law is allowed if it is reached in collective bargaining with a union. The NBA's salary cap was recently upheld in court because it was reached in such an agreement. (If the cap is imposed after a failure to negotiate in good faith, it is forbidden by labor law rather than anti-trust law.) A June 1996 Supreme Court decision affirmed this principle; a union cannot file an anti-trust suit on behalf of its members.

However, anti-trust law may still have an effect on labor relations. The NFL players decertified the union in both 1987 and 2011, which removed the labor exception and allowed the players to sue the NFL for under anti-trust. The NBA players decertified their union in 2011. The baseball players could have done the same to overturn the imposed cap if anti-trust law had applied in 1994. It now applies to MLB's labor negotiations.

Collusion would be forbidden by the anti-trust laws if they applied to baseball; instead, it is officially forbidden by the collective bargaining agreement. When the owners colluded in the free-agent market in 1985-1987, they paid their penalty under the terms of the CBA, which limited the penalty to the actual damages. If the players' union had been able to sue under anti-trust law, the damages would have been tripled. (All subsequent current CBAs specify triple damages for collusion.)

D. Other economic issues

D1. What effect would X have on ticket prices?

The answer to this question is usually "none", regardless of X. Most commonly, X is something like "higher salaries" or "a smaller TV contract"; in these cases, "none" is correct.

Baseball owners, like most business owners, are interested in maximizing profits or minimizing losses. Thus they set ticket prices with that goal in mind. Since having an additional fan attend the game does not have much effect on the cost of holding a game, this means that prices are set to maximize revenues.

For example, if there are two million fans willing to pay $10 to see the game, but only 1.6 million willing to pay $12, then it would not be a good economic decision to raise ticket prices from $10 to $12, because it will decrease revenues and profits. But suppose instead that there are 1.7 million willing to pay $12. In that case, ticket prices will be raised to $12, because the increase will generate an extra $400,000 of revenue.

Now, suppose something happens which affects the team's profit, without affecting the number of fans who want to attend games at any given ticket price. For example, the team could get less money from a new TV deal, or could have its payroll increase as several good young players became eligible for arbitration and others were kept as free agents, or the owner could lose money when one of his other businesses went bankrupt. The ticket price which maximizes revenue would not change, so the owner would continue to charge the same price, but make a lower profit.

When would a change affect ticket prices? Only if it affected the demand for tickets. This might happen if the team was improved by signing free agents; however, the teams which lost those free agents would have the opposite effect, so this would not cause ticket prices to change on a league-wide basis. The opening of a new stadium, or improvements in an existing stadium, might result in higher ticket prices for a better product; this is one of the main causes of recent ticket price increases. The end of a recession in the city would increase disposable income, and thus might increase the optimal ticket price.

Even in these cases, it isn't clear that ticket prices will go up. If the change means that 20% more people are willing to buy tickets at any price, then revenue for a fixed price will go up by 20%, so the optimal price won't change; however, revenues and thus profits will increase.

However, there is a current trend which may cause prices to go up. In some cities, the team is so popular that the park sells out regularly. If the team becomes more popular, demand will go up at a constant ticket price, but revenue will not go up with demand because some people who want to buy tickets cannot get them. (Some of the extra revenue would be collected by ticket agencies and scalpers instead of the team, since there are people who do not have tickets but who are willing to pay more than the ticket prices.) Thus these teams will have to increase ticket prices to maximize revenue, even if demand goes up by the same constant factor at all prices. This is why Fenway, the smallest park in baseball, has by far the highest ticket prices.

D2. Have ticket prices gone up?

Yes, they have gone up faster than inflation in recent years, primarily because of new parks, and the increasing popularity of the game in older parks which sell out. This is a new trend; a study in Baseball and Billions, by Andrew Zimbalist, compares average ticket prices to the cost of living from 1950 to 1990. There are minor fluctuations (adjusted prices were highest in 1970 by a small amount), but they had been essentially constant over time.

There are 10 teams which played in the same park in 1991 and 2006 (not counting Oakland, which raised its average ticket price by removing the cheap seats in 2006). Only the Royals and Blue Jays increased prices at about the rate of inflation over those 15 years. Most of the others are big-city parks which sell out. The Red Sox have the largest increase, but this is because they sold out almost every game even back in 1991, so the only way they could capture the increased demand was to raise prices. The Twins are the one unusual case; they capitalized on their 2002-2004 success by raising ticket prices rather than drawing more fans at the same price.

However, the new parks usually have much higher ticket prices than the old parks they replace; most new parks had about a 50% ticket price increase. Fans are paying more for a better product when they pay the higher prices for games in modern ballparks. Once the new parks have opened, there is no tendency for ticket prices to outpace inflation. Of the 14 teams that opened a new park between 1992 and 2001, two have lowered prices since the park opened (and neither was just a lowering from an overpriced first year), one barely raised them, eight were close to inflation, two were significantly above, and only the Orioles, who actually lowered their prices when Camden Yards opened, raised prices far more than inflation. Subsequent data confirms this trend, with some variation due to the economy; ticket prices followed inflation in 2007, increased more than inflation in 2008, but had no average increase in either 2009 or 2010 except for the Yankees (2009), Mets (2009), and Twins (2010) moving into new parks.

You will often see comparisons which say, "It costs a family of four $200 to attend a game today," along with some ticket price from the past. This comparison is meaningless unless it is adjusted both for inflation and for the difference between what is being purchased. The Fan Cost Index used in these figures is the cost of a once-a-year family trip, not a trip for regular fans; it includes tickets, food for everyone, and several souvenirs. If the tickets alone cost $25 thirty years ago and $90 now, the price has not increased in real value; the family which now makes $90,000 probably would have made only $25,000 back then in similar jobs, and now pays $1800 monthly to rent a house which rented for $500 then.

D3. Why do players make so much? What is the market value of a player?

The value of a player to his team is his marginal revenue; this is the amount of revenue which the team makes with him but would not make without him. If he is a good player, his team will win more games if he plays for them, and will thus sell more tickets, collect more from concessions, get more TV viewers, and have a better chance at World Series money. If he has extra drawing power as an individual, he will also help sell more tickets. All of these may be worth a lot of money to the team. If the team expects the player to be worth $4M in additional revenue, it should be willing to pay the player up to $4M, since it will make a profit on the deal; if he asks for more than that, it should let him go.

For baseball players, such a high value is reasonable. A study in Baseball and Billions estimates that the value of an extra win to a team in 1984-1989 was $400,000, independent of the team's market size. Projecting this to current revenues, that would be $2M, which means that a player worth five extra wins (typical for a superstar) would generate $10M in extra revenue. A later study by Nate Silver in Baseball Between the Numbers estimates that the value of an extra win averages $1.8M, but the value varies beteen $.7M and $4.5M, with the highest figure for teams expecting to win 90 games because wins around 90 have the greatest effect on the probability that a team will make the playoffs. A team expecting to win 84-96 games could pay over $15M for a player worth five wins.

The market value of a player is what he would earn if there were open competitive bidding for his services. In theory, this should be the expected value of the player to the team for which he has the second-highest expected value, since the team for which he is most valuable can offer that amount and nobody will beat it. This would be the player's actual salary in a free market.

Players who are not subject to a free market may not make their expected marginal value. Players who aren't eligible for arbitration usually make much less, because they have very little option. Players who are eligible for arbitration still tend to make less than their marginal value (see below). Players who are under long-term contracts may make more or less than their marginal value in one particular year; however, when the contract was offered, it was probably offered for the expected value or less, and both sides are now taking the risk. If a team misjudges expected values and signs players for more than their value, it should pay for its bad business decisions.

The same principles apply to any worker who is free to market his or her services. If employers X, Y, and Z all believe that you will generate $30,000 in additional revenue if they hire you, then all three will be willing to match each other's salary offers if they are under that level. If Z is stupid enough to offer you $35,000, you'll take the offer, and if Z makes too many of these mistakes, its profits will drop, and its executives will lose their jobs or the company will go bankrupt. If Z offers you $35,000 because it believes you are worth $40,000, and it turns out to be right, its profits will go up, and if Z makes more of these good deals, X and Y will be in trouble.

D4. How are salaries determined for players who have X years of service?

Players with less than two years of service, and the 78% of players with the lowest service time among third-year players, have no negotiating rights with their teams. The teams can offer whatever they want, subject only to the minimum salary. The players' only leverage is to refuse to sign any contract at all; they may not attempt to negotiate with another team. Such players often don't get just the minimum salary, partly because of the team's interest in maintaining good will; paying a player $1M instead of $500,000 for $4M worth of production is still a good deal, and may make the player more likely to stay for below market value when he becomes a free agent.

The top 22% of third-year players, and all players with at least three years of service, are eligible for arbitration. They may still negotiate with their teams for salaries. However, their teams cannot force them to accept an offer or go without a job. If a player and his team cannot agree on a salary, the team may choose to release the player or offer arbitration. The player cannot force the team to offer arbitration. If the team releases the player, he becomes a free agent, and the team which released him is not entitled to any compensation.

A player with less than six years of service must accept arbitration; a player with six years or more may refuse arbitration and become a free agent. The signing team may be required to give a draft pick as compensation to the team if it makes a qualifying offer of at least the average of the top 125 salaries. Except for the value of the draft pick, this is essentially a free-market negotiation. The player can demand X years for Y million dollars, but he won't get it unless at least one owner thinks he is worth that much.

A player in the middle of a long-term contract can ask to renegotiate the contract. However, the team is under no obligation to renegotiate; it can require the player to fulfill the previous contract. The team might agree to renegotiation in the interest of good will, hoping to retain the player when the contract expires. This rarely happens, but it has happened occasionally. An important example is Rickey Henderson, who signed an artificially low contract because of collusion, and wanted to renegotiate for his fair market value after the collusion had been exposed.

D5. How does arbitration work? Does it force teams to overpay for players?

The arbitration procedure was added to a previous CBA at the request of the owners; it was modified in the 1997-2001 CBA.

Arbitrators are members of the American Arbitration Association. Any member of the AAA may volunteer to be in the "pool" of arbitrators eligible to hear baseball cases. Lists of volunteers are examined by representatives of the players and the owners; an arbitrator must be approved by both groups to be in the pool. (This discourages arbitrators from appearing biased for either side; such arbitrators will be dropped from the pool.) The arbitrator for an individual case is chosen at random from the pool. Cases are heard by a panel of three people, with one arbitrator chosen from the pool, and one chosen by each side.

Arbitration can be offered only by the team; a player cannot force his team to offer arbitration, although he can refuse an offer of arbitration if he is eligible for free agency. The player and the team submit their proposed salaries to an arbitrator, and present their cases. The CBA specifies that team finances, including the luxury tax, may not be considered in arbitration; thus players do not have their arbitration salaries reduced because the team's payroll causes a tax to be imposed. The arbitrator must choose one figure or the other; this discourages unreasonable demands from either side. Contracts awarded in arbitration are always for one year, with no incentive clauses. However, the team and player may settle on some figure in between the two, or on a long-term or incentive-based deal, before the decision has been announced.

The arbitrator's decision is based on the salaries of comparable players. Thus, if other players and owners negotiate or arbitrate contracts which are unreasonably high or low, these will be considered as comparisons. This is why owners and fans talk about arbitration as "enshrining previous mistakes." However, a single anomalous contract is not likely to have a major effect on arbitration, since an arbitrator can recognize it as an anomaly.

Arbitration is unlikely to force an owner to pay a player more than his value. If the owner expects that the player will request and earn more than his value in arbitration, he can release the player instead of offering arbitration. This is a good economic decision if the player would have earned more than his value; that is, if the amount of salary saved is more than the amount of revenue lost. This rarely happened before the strike, but it has become more common recently. A fair number of players were released in 1995 rather than being offered arbitration. In 1997 and again in 2003, some of the top free agents were not offered arbitration; their teams thus gave up the draft pick they would receive as compensation, but avoided the risk of paying these players arbitration salaries which would be more than the teams were willing to pay.

It is often pointed out that players get huge raises in arbitration, but this is the result of the artificially lowered salaries for non-arbitration-eligible players. A player with two years of service will have to accept whatever the team offers, which may be near the minimum if he is not a star; when he has three years of service, arbitration increases his salary closer to his fair value.

D6. Are teams losing money?

While most businesses have an incentive to report high profits, MLB teams have an incentive to report low profits, and can manipulate reported profits more easily and legally than publicly traded corporations. Public corporations must make their audited records available; privately held businesses such as MLB teams need only report to their own owners. Public corporations improve their stock value, and thus their market value, by reporting profits; privately traded businesses are not valued in a stock market, but by potential buyers who know the true finances. The value of an MLB team does not depend directly on reported profits, but teams reporting losses are in a stronger position for negotiating with the public for new stadiums, lease concessions, and other benefits. In addition, the revenue-sharing rules give teams an incentive to minimize reported sharable revenue.

MLB claims to have operated at a loss every year from 1975 to 1985, then at a profit every year from then until the strike, although the profit was only $22M in 1992 and $36M in 1993. The strike led to large losses; the reported figures were $375M in 1994, $326M in 1995, and $185M in 1996; attendance was still down by 15% in 1996 from its pre-strike level. Reported losses have continued since then, $176M in 1997, $138M in 1998, $212M in 1999, about $500M in 2000, and $519M in 2001. The small reported profit in 1993 would be consistent with 12 of 28 teams losing money, which is what Bud Selig claimed. MLB's official figures claim that 25 of 30 teams lost money in 2001.

The 2007-2008 statements for the Pirates, Rays, Mariners, the 2008-2009 statements for the Angels and Mariners, and a draft workbook for the 2007-2008 finances of the Rangers, were leaked. Over those twelve seasons, the teams showed an average profit of $6M, with the Mariners reporting a loss of $7.8M in 2008 and the Rangers reporting a loss of $38M in 2007 and $10M in 2008. As a group, these six teams would not have reported profits without revenue sharing, as they received an average payment of $15M, with the Angels and Mariners paying both years, the Rangers reciving $5M in 2007 and paying $5.5M in 2008, and the other teams receiving over $30M a year.

Other sources claim much smaller losses. Forbes magazine, estimating finances from publicly available data, reported a profit of $130M in 2000, with ten teams losing money; a profit of $75M in 2001, with eight teams losing money; a loss of $39M in 2002, with sixteen teams losing money; a loss of $57M in 2003, with fifteen teams losing money; a profit of $132M in 2004, with ten teams losing money; a profit of $360M in 2005, with five teams losing money; a profit of $496M in 2006, with only the Yankees losing money; a profit of $492M in 2007, with three teams losing money; a profit of $501M in 2008, with the Yankees and Tigers losing money; and a profit of $522M in 2009, with the Tigers and Diamondbacks losing money. Older figures show similar differences. Financial World magazine reported profit of $168M in 1993, with eight teams losing money, a loss of $123M in 1994, and a profit of $59M in 1995. The Rockies' ownership claimed during the strike that seven teams lost money in 1993; The Sporting News reported in August 1995 that only three teams were losing money.

Which numbers are correct? It's difficult to tell without open books; however, it is almost certain that some of the teams which MLB claims are losing money are not actually losing any. When the owners opened their books to economist Roger Noll in the 1985 negotiations, he found enough hidden revenue and accounting techniques to turn the claimed $50M loss for 1984 into a $9M profit. Since the current books have not been opened for public analysis (although Noll looked at the 1993 books, and 2001 figures were released to Congress), it is impossible to tell how much such techniques are still being used. The 2001 figures were audited, but this means only that all transactions were counted, not that they were fairly valued.

The details of Noll's analysis are given in Baseball and Billions.

One of the biggest problems in determining a team's real profit or loss is the use of related-party transactions; that is, transactions between two entities which provide money to the same people. For example, AOL Time Warner owns both the Braves and TBS, so the Braves' TV contract can be set arbitrarily to cause either the Braves or TBS to show an artificially low profit. In MLB's official 2001 figures, the Braves' contract was at about the MLB average, despite the superstation; the Cubs (WGN is also owned by the Tribune Company) and the Dodgers (owned by Fox) showed similarly small deals. Similarly, Wayne Huizenga used to own both the Marlins and their stadium, so he could set the stadium rent at whatever price he wants, and assign luxury box revenue to either the team or the stadium. He could thus cause the Marlins to show an artifically high or low book profit. This is the basis for his claim that the Marlins would lose money in 1997 even if they sold out every game. Andrew Zimbalist anayzed the reported numbers, and concluded that the Marlins' actual profit in 1997 was $13.8M with a reported loss of $29.3M. Another possibility is for the team to pay a large salary or loan interest to the owner or to relatives of the owner, instead of distributing the same money as profits; the Brewers were reported to be doing this.

Undervaluation of related-party transactions may also reduce a team's revenue-sharing payments, and thus add to the team's actual profit. If a local TV contract is undervalued by $10M and revenue is shared at 30%, the team saves $3M which would otherwise go to the revenue-sharing pool. MLB was reportedly auditing the Yankees' deal with YES in an attempt to collect the shared revenue on the undervaluation.

Another problem, which could be resolved by open books, is the use (or abuse) of accounting practices. For US (but not Canadian) tax purposes, half the purchase price of a team may be attributed to player contracts. This is considered to be a purchase of short-term assets, which may be amortized. Amortization is normally used to reflect the declining value of assets which will expire, so that a new patent is worth more than a similar patent which will expire in one year. This does not make sense in MLB, because the true asset is not the specific players, but the right to acquire players for below their market value. Similarly, teams which own their stadium can depreciate the stadium, reducing its book value by a fixed percentage every year; while depreciation of buildings does make sense because buildings do eventually need to be replaced, the rate allowed by accounting principles is much faster than the actual deterioration rate of the building. The depreciation and amortization losses are paper losses, useful for their tax benefits, which will be regained by the owner when the team is sold at its real franchise value. MLB's own figures released to Congress in 2001 show $116M in interest and $174M in depreciation and amortization.

Also, part of the profits of owning a baseball team (as with other investments) comes from the appreciation of the franchise value. If you buy a team for $100M, break even in cash flow for five years, and then sell it for $150M, you have made a substantial profit, just as if you had bought $100M worth of real estate and it became worth $150M. And if you buy the team for $100M by taking out $40M in loans, show an operating loss equal to the after-tax interest you pay on the loans, and then sell it for $150M and pay off the loan balance yourself, you have made exactly the same real (economic) profit; the $40M loans allowed you to invest an extra $40M somewhere else.

A team can certainly lose money if it gets shut out of the market, or if the owner uses the team's money for another purpose. In 2011, the Dodgers were in an ownership dispute, with the owner's ex-wife claiming a half share of the team. MLB did not allow the Dodgers to sign a new television contract out of concern that the owner would divert the payment to his own use; without the contract, the Dodgers filed for bankruptcy.

It is possible for a team to lose money without overpaying for players, because the true values of the individual players, plus non-salary expenses such as stadium rent and player development, are not guaranteed to cover the total revenue in a particular team. If there are teams in this position, they could benefit from revenue sharing, or from a tax or salary cap which lowered salaries, and such moves might be necessary to keep teams in these cities.

But it is also possible that teams which are losing money are losing it because they are paying players, particularly free agents, more than their value. Some of these may simply be unlucky decisions, paying large contracts to players who got hurt or declined unexpectedly. This is a risk which should be calculated in the value of the contract, along with the potential gain if the player performs unexpectedly well; once this risk is considered, the problem won't consistently affect any team, although it may lead to single-year losses.

It is also probable that many players are overpaid because of poor talent judgment; for example, paying a 32-year-old player the value of his performance at ages 26-29, expecting him to repeat it, is usually a bad economic move. Such moves are made almost every time a free agent over 30 is signed. Owners who lose money because of such bad moves have the same interest in making a profit as other owners, and may be more likely to make a profit by a salary cap. From a free-market point of view, this is not a good idea; the free market will drive out incompetence, as such owners can make a profit by selling their teams to competent ownership or replacing their general managers.

D7. What is collusion?

Collusion occurs when a group acts in concert in business. Thus, if several owners agree not to make offers to each other's free agents, or to limit their offers, they are guilty of collusion; this is what happened in 1985-1987. (The cases were based on actual evidence of such communications, not simply the fact that some players received no offers.) It would also be collusion if the owners agreed to follow a salary cap without getting it in the CBA.

It is not collusion if owners act independently to reduce expenses. Many teams have publicly announced their budgets; this is fine as long as no team can control another team's budget. Likewise, if a free agent asks for $3M and he receives no offers because no owner thinks he is worth that much, this is not collusion; it happened to Jody Reed and Chris Sabo.

There is a clause in the CBA forbidding players or teams from acting in collusion. It was put there at the request of the owners, to prevent players from staging joint holdouts. In 1985-1987, the players charged the owners with violations of this clause. As specified in the CBA, the hearings were held before an arbitrator, not in a lawsuit.

There were rumors of collusion in the April 1995 free-agent signings, because so many players took large salary cuts. Charges were filed by the union, and there were some complaints by agents. However, this is probably not collusion; most of the owners discovered that they had less money after the strike and would make less in 1995, and were thus more careful where they spent it. There were similar rumors of collusion charges in 2002 and 2003. The union never filed any formal charges, but continued to negotiate privately with the owners. The charges were settled along with a list of other unresolved grievances, with an estimated $9M of the total $12M settlement going to settle the collusion cases.

The union announced that it would investigate the owners' exchanging information about team plans and needs in a November 2007 meeting. If the owners discussed plans to sign or keep free agents, the meeting could be a prelude to collusion; in 1985-1986, free agents received no offers if their own teams wanted to keep them. No charges have yet been filed.

D8. What are the owners' rights under labor law?

If labor negotiations reach an impasse, management may implement its last offer. This was the basis for the owners' imposition of the cap on December 23, 1994. (MLB made it clear that the salary cap was still the "last offer," although tax plans had also been offered.) One part of Judge Sotomayor's decision was that the owners would have had to bring a future declaration of impasse before her, rather than simply declaring it. In early August 1996, the owners announced that they would make a "last offer" and ask Judge Sotomayor to declare an impasse if the players refused, but this never happened because what followed was two weeks of productive bargaining, eventually leading to the final agreement.

In order to impose its last offer, management must negotiate in good faith. The National Labor Relations Act has the following definition: "To bargain collectively meet at reasonable times and confer in good faith with respect to wages, hours and other terms of employment...but such obligation does not compel either party to agree to a proposal or require the making of a concession."

Another important point is that management may not use poverty as a tactics in labor negotiations unless it opens its books. The owners did open the books to the union, but under an agreement that the union would not disclose the results.

Both management and labor are forbidden from engaging in unfair labor practices. One common example of an unfair labor practice is retaliation based on union activity, by either management or labor. Another example is a breach of contract in order to gain more negotiating leverage. The owners refused to make a pension payment on August 1, 1994, before the union went on strike in 1994. The union filed a complaint, and the owners settled by making the payment with interest.

D9. What is the situation in the umpires' labor action?

The MLB umpires' contract expired on December 31, 1999. After the contract expired, MLB could lock the umpires out. Fearing a lockout but forbidden to strike under their labor contract, the umpires' union asked umpires to resign effective September 2, 1999, and most of the union umpires honored this by submitting resignations.

When MLB ignored the action, many umpires rescinded their resignations; meanwhile, MLB started hiring replacements for the umpires who were resigning. The last 22 umpires to rescind their resignations were too late, as MLB had already hired replacements for them. On September 2, 1999, the replacements began work.

The umpires' union chose to threaten a labor action at a time at which they would have more leverage; the players' union did the same in 1994. The players' August 12 strike threatened the 1994 playoffs, the most profitable part of the season, but allowed enough time for the playoffs to be played on schedule if there was a quick settlement. Similarly, the umpires threatened to resign on September 2, which could force MLB to use inferior umpires for the pennant races and post-season if there was no settlement.

An important difference between the two labor actions is that the umpires' union had a no-strike clause in its contract. Such clauses normally forbid any concerted labor action, in order to prevent sick-outs and other labor disruptions which are not technically strikes but have the same effect. The players played 1993 under an agreement not to strike (which they made in return for an agreement by the owners not to lock them out or unilaterally impose conditions). The players' union made the same offer in 1994, but the owners refused, and the players then struck in 1994. In contrast, the umpires' action was effectively a breach of its own labor contract.

Since the union was not legally on strike, it has probably forfeited the rights it would normally have as a union on strike. MLB has replaced resigning umpires with new, permanent employees; these umpires cannot be considered strike-breakers because there was no strike. In fact, several have joined the union; strike-breakers in the players' strike were not allowed to join it. Likewise, it is illegal for an employer to dismiss a worker for union activity, but this is unlikely to apply to the umpires who resigned. (There has been some discussion about MLB's need to be careful in selecting which umpires to replace; it cannot selectively retaliate against union activists.)

The old umpires' union filed a charge with an arbitrator alleging that the umpires were unfairly dismissed. MLB offered to re-hire 13 of the 22 resigning umpires and allow the others to retire, but the old union refused this offer, insisting on re-hiring of all 22. The arbitrator ruled that the resignations were valid, but that seven umpires must be reinstated and two must be given back pay as if they had retired rather than resigned; two AL umpires had not officially resigned, and seven of the NL umpires were refused re-hiring arbitrarily.

Both sides appealed this ruling in court, and it was mostly upheld on December 14, 2001; the nine were ordered re-hired with back pay, and three others would have their cases re-heard by another arbitrator because the first arbitrator applied the wrong rules for umpires with less than five years of service. The nine who were ordered re-hired received their back pay; five were re-hired, while four retired with the back pay. The three whose cases were re-heard settled their cases; they were re-hired without back pay. The remaining ten umpires were not entitled to reinstatment or back pay.

The ten umpires who received nothing appealed their cases, while MLB appealed the order to give back pay to the nine umpires who were ordered re-hired. The arbitrator's ruling was upheld by the District Court on December 9, 2002, and by the Third Circuit Court of Appeals. MLB appealed to the Supreme Court, which declined to hear the case in January 2005.

Meanwhile, umpires dissatisfied with Richie Phillips, the union leader who proposed the mass resignations, called for a new union. The new union won a majority of votes in an NLRB election. The old union challenged the election on the grounds that the owners improperly supported the new union, but the election was upheld by the NLRB and on appeal. The new umpires' union signed an agreement on August 30, 2000.

The union reached a new labor agreement with MLB on December 22, 2004, which also included a nearly final settlement of the outstanding labor action. Of the ten umpires who received nothing in 2002, three were guaranteed re-hiring within MLB's next five openings for umpires, six were allowed to retire in 2004 with severance pay, and one, Rich Garcia, who had already been re-hired as an umpire supervisor, will also receive severance pay. The only remaning issue was the amount of back pay for the five umpires who were ordered re-hired in 2001; that was determined in court in the umpires' favor on August 24, 2006, with the umpires receiving $3.1M in back pay rather than the $1.9M that MLB claimed was owed.

Another new agreement was reached on December 23, 2009, covering seasons through 2014.

E. Credits

The reference for some of the data cited here (ticket prices and contract terms in particular), and a source of much more data about baseball finances, is Doug Pappas's Business of Baseball Pages, The details of proposals in labor negotiations are from Associated Press and news articles; the details of the 2007-2011 CBA are from the official MLB press release, with some clarifications added from Neil DeMause and Maury Brown, "You've got the Brawn, I've Got the Brains, Let's Make Lots of Money", in Christina Kahrl and Steven Goldman (eds.), Baseball Prospectus 2007, London, Penguin Books, 2007. The details of the leaked 2007-2009 statements are from "Detailed Financial Info From Pirates, Rays, Angels, Marlins, and Mariners Sees the Light of Day", by Maury Brown, in at, and from "Leaked Documents Show Rare Look at the Finances of the Texas Rangers", by Maury Brown, at F. Copyright and disclaimers

This document is Copyright 2011, David Grabiner. The document may be copied and distributed freely in unmodified form, provided that this notice remains intact. It may not be sold or included in a collection which is sold without the permission of the author.

The opinions in this document are those of the author, not necessarily those of Princeton University. Legal opinions included here should not be considered legal advice.

This document may be cited as,

David Grabiner, "Frequently Asked Questions about the Baseball Labor Negotiations", [date under "Last Modified"], available electronically from

Please note that this document was written as a reference point for an ongoing discussion rather than an authoritative article. If the facts may have changed, please check the "Last Modified" date to make sure that the current version is being cited. You may also want to check with the author or confirm facts from another source.