The Economic Impact of Changing CBT Thresholds and Penalties

This past Saturday, as part of the ongoing collective bargaining agreement negotiations, Major League Baseball sent its second proposal on core economic issues to the Major League Baseball Players Association. We’ve already covered how the two sides differ on pre-arbitration compensation, and examined how changing the arbitration eligibility rules would alter player salaries based on recent arbitration awards. MLB and the MLBPA have also laid out proposals regarding the competitive balance tax, proposals that would have strikingly different effects on team spending.

To compare the two approaches, I started with the actual tax regime from the previous CBA, which was in effect from 2017 through ’21. I made one modification: the abbreviated 2020 season led the league and the union to bilaterally amend the CBA to drop the competitive balance tax for that season. Payrolls also ended up being quite different than their original projections due to the 60-game slate. For the purposes of this analysis, I’ve turned each payroll into a full-season number and calculated the tax as if 2020 were a regular year (hopefully, how the new CBT handles a pandemic will not be relevant for future seasons).

As a reminder, the previous CBA had three thresholds (over the cap, but by less than $20 million; $20 million-$40 million over the cap; and more than $40 million over the cap) as well as escalating penalties for repeated overages. The tax rate started at 20% and climbed from there. It also carried draft pick penalties for teams that exceeded the cap by $40 million or more. I took each team’s yearly CBT salary from Cot’s Contracts to determine which teams were over the cap. I valued the draft picks using past research by Craig Edwards and looked at the total tax cost levied by the old agreement:

Of note, the 2017 tax year used a hybrid approach that split the difference between the 2016 tax regime and the new one. Repeater penalties kicked in fully in 2018; no team has paid the tax three years in a row since then. That three-time offender tax rate is 50%, with surcharges kicking it up to 62% and 95% at the higher thresholds.

The league has proposed new tax rates, as Jay Jaffe detailed for us on Monday. There would no longer be repeat overage penalties. Instead, the new tax rate would be 50% for all teams exceeding the cap by any amount. At $20 million, the marginal rate would climb to 75%, then to 100% for teams over by $40 million or more. In addition, the draft pick penalties would change to a forfeited second round pick for being $20 million-$40 million over and a first round pick for being $40 million over.

The penalty would actually be two picks each – one in the current domestic amateur draft, and one in a heretofore-nonexistent (but reportedly forthcoming) international draft – but for the purposes of today’s article, I’ll ignore the international draft. I also assumed that the value of a first-round pick was the average of picks 6-30, rather than 1-30, because teams in the tax aren’t generally picking in the top five, though that assumption could certainly change. The old CBA included protections if a team’s forfeited pick would have been in the top six; I believe my estimation roughly replicates that effect, though the details of any protections in the league’s most recent proposal have not been announced.

Using the old tax thresholds and the new penalty rates, here’s how the competitive balance tax would have been assessed from 2017-21:

MLB’s proposal would see roughly $100 million more in assessed tax penalties, not including draft picks. The monetary value of drafts picks forfeited would skyrocket from roughly $10 million to roughly $80 million. Note: a previous version of this article under-stated the Padres’ potential luxury tax bill in this theoretical situation by roughly $2 million.

The effect of an increased CBT isn’t a one-for-one loss of player revenue. The assessed tax flows into two buckets: roughly half (slightly more, because the first $13 million of total taxes also count toward this bucket) goes to defraying player benefit costs, while the rest is remitted to teams that didn’t exceed the tax, though those teams aren’t required to reinvest that money into their roster. To approximate the effect on payrolls, I came up with a workaround. I worked out a new payroll level that would lead to the same amount of total league-wide tax dollars. In other words, teams would need to spend less in salary to end up with the same tax bills that they actually received from 2017-21. I calculated how much less to ballpark the economic impact of the tax. This doesn’t exactly handle team behavior, but I believe it’s a good first-order approximation.

More specifically, I reduced each team’s payroll by a consistent fraction while they were over the tax. When they reached a non-taxable level, I stopped reducing that team’s payroll, but continued to reduce the rest of them proportionally, until another team reached the lowest threshold, and so on. In the end, that would lower payrolls like so:

That comes out to an aggregate $207.5 million in reduced payrolls, or $41.5 million in salary per year. To reiterate, that reduction represents the payroll level that would result in constant total tax collected for the new tax rates proposed by the owners. The new tax rates proposed by the players are the same as the old tax rates, only minus the draft pick penalties, which means that team behavior would have been roughly unchanged under their proposal, albeit with higher initial thresholds to work with.

Next, let’s cover the thresholds that trigger the CBT. The league proposes increasing the first threshold for the CBT by 1.9%, 0%, 0.9%, 0.9%, and 1.8% in the five years of the new CBA. To calculate how those levels would impact total spending, I did the same thing: I started with the 2016 CBT level and increased it by those rates each year, then applied the proposed new penalties based on the same cap formula. That would make the 2017-2021 first thresholds lower: $192.6 million, $192.6 million, $194.3 million, $196.1 million, and $199.6 million respectively. This would bring more teams over the cap, which I’ve accounted for below:

If teams didn’t change their spending at all but were governed by both higher rates and lower CBT thresholds, they would have paid $383 million in tax and lost draft picks worth $136 million. To reach an amount of tax equivalent to what was actually paid (including the monetary value of draft picks), they would have needed to spend an aggregate $383 million less from 2017-21, or $76.6 million per year. That’s a hefty chunk of change relative to the status quo.

The MLBPA’s proposal would have the opposite effect. The tax rates are unchanged (except for the removal of draft pick compensation), but the tax levels would change significantly. I used the same methodology, starting with the 2016 tax level and adding the percentage increases the players are calling for:

To reach the same amount of tax paid, teams could have spent an additional $474 million, or $94.8 million per year. This is the estimate I’m least sure about; given that several teams spent up to just below the first threshold in past years, there would likely be some knock-on behavior from teams spending up to the cap. But it’s also unlikely that keeping the tax bill constant is a binding constraint in this direction to the same extent that it is to the downside; every team has behaved to minimize its tax bill, but a team that went $5 million over the cap under the existing regime (like the 2021 Padres) wouldn’t necessarily increase spending by $30 million dollars (as they did in this example) to get near the luxury tax again. I’d argue that both proposals I ran with changed thresholds don’t reflect actual team behavior; modeling how teams would approach different CBT thresholds is trickier than modeling how they would approach different penalties.

The most likely place for the players’ proposal to change is in the first CBT increase, and that’s also the part of the model most sensitive to change. The 16.7% increase they propose is essentially a catch-up that accounts for the fact that the CBT threshold increased by an average of 2.1% over the course of the last CBA while revenues increased at a faster rate. Lower that first increase to 5% and the extra spending number drops from $474 million to $137 million.

Where you think the CBT threshold should land is trickier than taking the midpoint of these numbers. Neither reflects how teams will actually behave; it’s likely that a more restrictive tax would lead to some high-spending teams accepting slightly higher tax bills, while a less restrictive tax would only raise some salaries, not all.

The tax regime that began in 2017 had a huge impact on salaries. Harsh repeater penalties and a 2% threshold growth rate meant that almost no one exceeded the tax rate for two years in a row, and that top-end team payrolls stayed roughly constant. It’s no accident that total MLB payrolls peaked in 2017 and have been declining ever since.

My main takeaway from this analysis: changing the competitive balance tax thresholds and rates will likely be one of the two biggest levers in the CBA negotiations, along with pre-free-agency compensation. Changing the structure of the tax has huge and hard-to-model changes for total league payrolls.

If the players’ proposal for CBT thresholds had been in effect in the last CBA, with no other changes to the CBA at all, and my modeling holds true, aggregate payrolls would have increased by just under 1% per year over the past five years (instead of declining by roughly 1% per year). If the owners’ proposal for CBT thresholds and rates had been in effect and my modeling holds true, aggregate payrolls would have decreased by nearly 2.5% per year. Want an example of this? This estimation of the owners’ proposed CBT regime has the Dodgers’ prospective payroll in 2021 dropping from $286 million to $257 million, a $29 million decrease. Despite that, they’d actually pay more in tax, $59 million counting draft penalties rather than $37.7 million in the current regime. Might they cut payroll even further?

What’s the right number to arrive at? That’s not for me to decide. The players’ proposal comes close to matching revenue gains, but you could perhaps argue that revenue gains have outstripped the long-term rate of league growth. One thing I feel comfortable saying: the league’s proposal isn’t one with a serious chance of being accepted. It’s a drastic escalation of the restrictive cap from the last bargaining agreement, one that has already eroded the players’ piece of the pie.

Could this merely be a bargaining position from the league’s negotiators? Certainly. But it’s absolutely not an extension of the status quo. It’s significantly harsher than that, in terms of both the rate at which the cap grows and the penalties for going over. Both sides have other levers to pull, but on this issue, the league’s suggestion looks to me like a massive overreach.