Interested in one of the top master’s in Economics programs in the country or a new, innovative microeconomics-focused  Ph.D. program in Economics?

 

Will the NBA Luxury Tax Take Effect in 2003-04, 2004-05, and 2005-06?

By Dan T. Rosenbaum

April 26, 2004

 

These projections are outdated as additonal information that I received in May and June suggested that David Stern's revenue warnings and my projections were too pessimistic about revenues. However, because of consulting negotiations with several teams, at this time I do not plan to make updated projections publicly available. For members of the press or team executives (or good friends) interested in updates of this page, please contact me at (336) 334-4872 or e-mail me at rosenbaum@uncg.edu. 

 

In 2003-04 salaries and benefits must be greater than 61.1% of basketball-related income (BRI) in order for the luxury tax to be triggered.  In 2004-05, the cut-off percentage of BRI increases to 63.3%.  Since July 2003 I had been predicting that salaries and benefits were likely to be high enough to trigger the luxury tax in 2003-04, but after 2003-04 the luxury tax likely was very unlikely to be triggered.  These predictions were based upon an assumption that revenue growth would not fall outside of historical bounds, an assumption that appears to have proven incorrect.  Following the April 2004 NBA Board of Governors meeting David Stern has stated that “overall, the league remains in a negative position, in a negative cash position.”  The League later clarified this claim, arguing that “while some teams lost money compared with last season, NBA revenue has increased overall.”  Subsequent conversations with multiple sources have confirmed that revenue growth in 2003-04 has been considerably lower than recent history would have suggested.  (I am still making sense of the conflicting claims of Frank Hughes and Phil Miller.  The former tells a more pessimistic story than that below, while the latter tells a far more optimistic story.)

 

Based upon this new evidence, I have updated my luxury tax predictions.

 

§         In 2003-04 there is a greater than 95 percent chance of the luxury tax being triggered.

§         In 2004-05 there is a 30 percent chance of the luxury tax being triggered.

§         In 2005-06 there is a 15 percent chance of the luxury tax being triggered if the next collective bargaining agreement (CBA) is like the current CBA using the 2004-05 63.3% BRI trigger.

§         In 2005-06 there is a 40 percent chance of the luxury tax being triggered if the next collective bargaining agreement (CBA) is like the current CBA using the 2003-04 61.1% BRI trigger.

 

In addition, I have updated the expected ranges for the luxury tax threshold and salary cap.

 

§         In 2003-04 I expect the luxury tax threshold to fall between $53.1 and $55.0 million.

§         In 2004-05 I expect the luxury tax threshold to fall between $56.0 and $60.1 million.  I expect the salary cap to fall between $41.3 and $44.4 million.  The mid-level exception is likely to be around $4.9 million once again.

 

Altogether, these predictions are significantly different than the conventional wisdom of a few months ago.  For example, many were predicting that the 2003-04 luxury tax threshold would hit between $55 and $56 million with the National Basketball Players Association predicting a threshold of $57 million.  With a much lower threshold that what was expected, many teams may be surprised by becoming taxpayers or by how much they will have to pay.  In parentheses, I list two numbers for each team.  The first is their likely luxury taxes paid. The second is their likely luxury/escrow tax distributions received minus luxury tax paid with negative numbers being net payments and positive numbers being net receipts.  See Larry Coon’s CBA FAQ or my paper on the luxury tax for more details on how these distributions are computed. All numbers are in millions of dollars and are for 2003-04 using my preferred estimates below.

 

§         Teams like New York ($41, -$34), Dallas ($26, -$20), Portland ($29, -$23), Minnesota ($17, -$10), Sacramento ($14, -$8) Los Angeles Lakers ($9, -$3), New Jersey ($9, -$3) , Philadelphia ($6, $0), and Toronto ($6, $0) all expected to be taxpayers.  They simply will be paying a couple million more than expected in 2003-04.

§         For Indiana ($4, +$2) and Memphis ($3, +$3) the combination of luxury tax/escrow tax distributions received minus luxury taxes paid is likely to range between break-even and about $5 million, much lower than they probably were expecting 

§         Houston ($2, +$9) may have hoped to avoid the luxury tax, but is likely to find itself paying taxes and losing distributions in excess of $6 million relative to non-taxpaying teams.  It looks like Boston will avoid the luxury tax because of the settlement with Vin Baker.

§         Chicago, Detroit, Golden State, and especially Phoenix may find themselves paying luxury taxes in some worst case scenarios.  Otherwise, they will join Atlanta, Cleveland, Denver, Los Angeles Clippers, Miami, Milwaukee, New Orleans, Orlando, Seattle, San Antonio, Utah, and Washington in being non-taxpayers and receiving $14 to $17 million in luxury/escrow tax distributions (it will be a little less after currency assistance and disabled player assistance is taken out).

 

The most significant outcome of this slow revenue growth will be felt in the summer 2004 free agent market.  Perhaps the most important aspect is that the salary cap is likely to fall.  The salary cap is based upon projected BRI, which assumes that non-national TV BRI will grow by 8 percent.  When that projection is not met, the shortfall is deducted from the next year’s projected BRI when determining the salary cap.  Thus, this summer the salary cap is likely to fall despite BRI being slightly higher.  Along with the luxury tax being a real possibility in 2004-05, this lower-than-expected salary cap is likely to lead to teams being quite frugal.

 

§         It may become increasingly difficult for the Los Angeles Clippers and especially Phoenix to clear enough salary cap space to make offers to maximum-salary players, such as Kobe Bryant.

§         Detroit and San Antonio are likely to find it more difficult to retain their own free agents.

§         Charlotte is likely to be able to charge an even higher price for taking on salary, further complicating the efforts of teams like the Clippers, Suns, Pistons, and Spurs.  Similarly, Utah, Atlanta, and Denver are likely to be offered very lopsided trades where they take on some salary from other teams.

§         Teams in the ballpark of the 2004-05 luxury tax threshold like Boston, Chicago, Houston, Memphis, Seattle, and Toronto may reconsider intentions to be major players in the free agent market given that the luxury tax might be rearing its head again.  And even teams like the Kings, Mavericks, Nets, Blazers, and Knicks may be more restrained than they would have been.

 

This all is likely to lead to a much, much tougher market for players this summer.  Unlike the past couple of summers, there is a deep pool of available free agents this summer and they may be in for a rude awakening as teams respond to the news about revenues.  Teams like the Jazz, Hawks, Nuggets, and Bobcats may be able to do some bargain-shopping this summer as reality sets in for the players.  Finally, there has been talk about the Bobcats not being overly aggressive in this summer’s free agent market.  In light of the available bargains, the Bobcats may want to revisit that decision.

 

Below I provide my reasoning for these conclusions.  For more on me, please see my main web-page (http://www.uncg.edu/eco/rosenbaum/) or my NBA-related web-page (http://www.uncg.edu/eco/rosenbaum/nba.html).  My NBA-related web page also highlights several of my NBA-related academic papers, including one dealing with the luxury tax.

 

COMMENTARY AND BACKGROUND:

 

Forecasting the luxury tax is the latest challenge laid at the feet of general managers and owners.  So now on top of (1) a salary cap with numerous exceptions and complex rules for trades, (2) rookie scale contracts, (3) minimum and maximum salaries that vary by years of experience, (4) taxes on players, (5) taxes on high-spending teams, and (6) complicated rules for distributing this tax money back to the teams, general managers and owners need to be able to forecast League-wide revenues and salaries, since that is what determines whether the luxury tax takes effect in a given season.  Without such forecasts teams don’t know the true costs of free agent signings or trades, since the luxury tax can double or sometimes more than quadruple the long-term cost of a potential free agent.

 

Through e-mail exchanges and phone conversations with numerous sportswriters and executives at many levels of various teams, I have gotten the distinct impression that what I present below possibly may be the most extensive analysis available of the question of whether or not there will a luxury tax.  My general strategy below has always been to present best case and worst case scenarios that are better and worse than anything the NBA has experienced over the past five or six years.  And I have argued that it is very unlikely that what actually transpires will fall outside of these bounds.  However, following the April 2004 NBA Board of Governors meeting I felt obliged these assumptions. Through multiple sources, I have come across information that makes me believe the revenue growth in 2003-04 will be low relative to historical standards.  If NBA revenues are actually increasing at an average or above average rate, then what I present below will greatly overestimate the probability of the luxury tax being triggered. 

 

In what I present below my “preferred” estimates are what I think is most likely to happen.  Simultaneously projecting both BRI and total salaries may seem farfetched, but by putting these projections in this kind of framework, I hope that I am able to bring some sense to what may seem like a very complicated forecasting problem.            

 

There is one final point I would like to make, before beginning the analysis.  For an economics professor like me who has spent much of his career researching the myriad of tax and welfare programs affecting low-income households, the NBA is a treasure trove of interesting economic applications.  For normal human beings (or even general managers or owners) whose lives are not centered around equations, crunching numbers, and thinking about incentive structures, I suspect this is all an annoying distraction from a beautiful game.  And it is a sad commentary on the NBA that its labor market has become so complicated (moreso than any other professional sports league) that having number crunchers like me around has become every bit as important as having people who can evaluate basketball talent.  I find it hard to believe that such a development will not have a long-term negative effect on the NBA, but that is a topic for another day.     

 

BRI HISTORY:

 

I start with an analysis of past BRI numbers.  The League generally reports total BRI, but the national-TV versus non-national TV breakdown is more difficult to get solid data on.  However, Rick Horrow of CBS Sportsline reported the 2001-02 and 2002-03 national TV BRI numbers and reports from various sources help make it possible to fill in the rest of the numbers in this table.

 

Table 1:

Basketball-Related Income Breakdown

(Dollar Amounts in Millions)

 

Year

BRI

National TV BRI

Percentage Change in TV BRI

Non-National TV BRI

 

Percentage Change in Non-TV BRI

2000-01

$2,450

$672

+20.2%

$1,778

+5.2%

2001-02

$2,664

$808

+20.2%

$1,856

+4.4%

2002-03

$2,662

$680

-16.7%

$1,982

+7.1%

2003-04**

$2,731

$689

+1.4%

$2,042

+3.0%

2004-05**

$2,833

$710

+3.0%

$2,123

+4.0%

2005-06**

$2,986

$731

+3.0%

$2,225

+4.0%

** Preferred estimates.  The 2005-06 estimate assumes that Charlotte generates about $47 million in new revenue for the League in 2005-06, which is not counted as part of the percentage change in non-TV BRI.  Charlotte’s BRI is not counted in BRI in 2004-05.   

 

Table 1 shows that the reason BRI decreased in 2002-03 was due to a large drop-off in the national TV contract.  Non-national TV BRI increased fairly rapidly.  Note, however, that about $13 million of this increase was due to the first round of the playoffs increasing from best-of-five to best-of-seven.  Excluding this increase, the non-national TV BRI growth rate in 2002-03 was about 6.3 percent – the growth rate that I use in my best case estimates below.  The rate of increase that I use for the national TV contract is the estimated 1.4 percent growth rate that is implied by the 2003-04 projected BRI (used to calculate the 2003-04 salary cap) less an adjustment for start-up costs for NBA-TV that I have heard is about $35 million.  Given the recent agreements with three major cable providers that will now carry NBA-TV, it is possible that this $35 million is an overestimate of the loss.

 

BRI PROJECTIONS:

 

So how can we use this past data to project BRI into the future?  I am going to use these data to come up with three estimates for BRI.

 

Worst Case BRI Estimate

-        Assumes national TV BRI increases by 1.0 percent in 2003-04 and 1.0 percent in 2004-05 and 2005-06.

-        Assumes non-national TV BRI increases by 1.0 percent in 2003-04 and 2.0 percent in 2004-05 and 2005-06.

Preferred BRI Estimate       

-        Assumes national TV BRI increases by 1.4 percent in 2003-04 and 3.0 percent in 2004-05 and 2005-06.

-        Assumes non-national TV BRI increases by 3.0 percent in 2003-04 and 4.0 percent in 2004-05 and 2005-06.

Best Case BRI Estimate  

-        Assumes national TV BRI increases by 2.5 percent in 2003-04 and 3.0 percent in 2004-05 and 2005-06.

-        Assumes non-national TV BRI increases by 5.0 percent in 2003-04 and 6.0 percent in 2004-05 and 2005-06.

 

Table 2 shows that BRI varies a lot under the scenarios that I consider.  Projecting BRI gets us only halfway.  The other half is projecting team salaries, which is more difficult, because it depends on a whole host of factors, including the perceptions by teams of whether or not there will be a luxury tax.  Thus, if this document changed conventional wisdom about the probability of the luxury tax, it could also change total salaries.  The best way to project team salaries is to go through team-by-team and estimate how much their total team salaries will be for 2003-04, 2004-05, and 2005-06.  I have done this for all 29 teams in 2003-04 and all 30 teams in 2004-05 and 2005-06.  (Again, I am assuming that the next CBA is fairly similar to the current CBA.)

 

Table 2:

Basketball-Related Income and Luxury Tax Projections

(Dollar Amounts in Millions)

 

Year/Estimate

Assumed

Change in

Non-TV BRI

(TV BRI)

Estimated BRI

Estimated Salary Cap

Estimated Luxury Tax Threshold

Estimated

Salaries

Estimated Salaries plus Benefits as

a Percentage of BRI

Probability of Luxury Tax

Overall Probability of Luxury Tax

2003-04

 

 

 

 

 

 

 

 

Worst Case

+1.0% (+1.0%)

$2,689

$43.84

$53.1

$1,643

65.0%

100%

> 95%

Preferred

+3.0% (+1.4%)

$2,731

$43.84

$54.0

$1,643

64.0%

100%

Best Case

+5.0% (+2.5%)

$2,778

$43.84

$55.0

$1,643

62.9%

100%

2004-05

 

 

 

 

 

 

 

 

Worst Case

+2.0% (+1.0%)

$2,736

$41.3

$56.0

$1,667

63.9%

70%

30%

Preferred

+4.0% (+3.0%)

$2,833

$42.9

$58.2

$1,703

62.9%

35%

Best Case

+6.0% (+3.0%)

$2,924

$44.4

$60.1

$1,735

62.0%

5%

2005-06

 

 

 

 

 

 

 

 

Worst Case

+2.0% (+1.0%)

$2,829

$43.8

$55.8

$1,655

62.7%

30% (90%)

15%

(40%)

Preferred

+4.0% (+3.0%)

$2,986

$45.8

$59.1

$1,709

61.2%

< 5% (45%)

Best Case

+6.0% (+3.0%)

$3,125

$47.4

$62.0

$1,754

59.9%

< 1% (< 5%)

** The 2005-06 estimate assumes that Charlotte generates about $47 million in new revenue for the League in 2005-06.  Charlotte’s BRI is not counted in BRI in 2004-05.  I assume that benefits are $103 million in 2003-04, $108 million in 2004-05, and $113 million in 2005-06.  In 2005-06 I assume the next collective bargaining agreement (CBA) is like the current CBA using the 2004-05 63.3% BRI trigger.  The numbers in parentheses in the last two columns assume a 61.1% BRI trigger.

 

2003-04 Season

 

The luxury tax is almost a certainty for 2003-04, but there are likely to be several teams (listed in the introduction) who surprisingly find themselves paying luxury tax.

 

In past versions of this document, I have discussed how high-spending teams could possibly conspire to inflate BRI.  For example, the six teams likely to pay the most luxury tax (Dallas, Minnesota, New York, Phoenix, Portland, and Sacramento) could each agree to increase BRI by a certain amount through buying advertising or tickets from the other five teams.  Such behavior could potentially eliminate the luxury tax in 2003-04, likely saving these teams about $90 million in net luxury taxes (luxury taxes paid minus the small luxury tax surplus that is distributed to all teams).  Since these owners would be receiving an amount in advertising (or ticket) revenue equal to the amount they paid out, the only costs involved would be due to (1) taxes paid on this revenue, (2) any revenue sharing with local governments or the League, (3) any related loss in advertising or ticket revenue, and (4) possibly a small reduction in escrow tax distributions.

 

It appears that such BRI manipulation is unlikely to occur in 2003-04, yet in 2004-05 when whether or not there is a luxury tax could be an extremely close call, efforts at manipulating BRI (up or down) could become more prevalent.  And before I leave this topic, I should mention again that as teams expend extensive resources manipulating BRI, efforts to improve their basketball product or the experience of their fans will suffer.  Such activity hurts the League as a whole and is one of the biggest drawbacks of the luxury/escrow tax system in its current form.   

 

2004-05 Season

 

The 2004-05 season will be very interesting.  Starting about midseason up until recently the conventional wisdom around the League was that the luxury tax was unlikely after 2003-04.  Even Russ Granik stated that “we have told the teams there is a substantial question about whether there will be a luxury tax after that season.”  Given the conservative nature of the League office, I think that many may have taken this to imply that the luxury tax was gone after 2003-04.  In anticipation of there being no luxury tax in 2004-05, teams like Portland and New York greatly increased their 2004-05 payrolls, exposing themselves to huge risks if revenues fell through the floor, which appears to be what has happened.  Given how close a call the luxury tax is in 2004-05, Isiah Thomas’s spending could possibly be what triggers the luxury tax, costing the Knicks in excess of $60 million in 2004-05 alone.  Similarly, Portland’s decision to take on Shareef Abdur-Rahim and Theo Ratliff could cost them tens of millions of dollars.  All in all, this is going to be a very interesting free agent season and there will be a premium on GMs for understanding (and understanding quickly) how salary cap and luxury tax concerns are likely to percolate through the market.     

 

2005-06 Season

 

Projecting the 2005-06 season is difficult because in addition to BRI and salary uncertainty, there is uncertainty in what the next CBA will look like.  Ironically, with the luxury tax possibly being in play for another season, this may actually improve the bargaining power of the Union.  The luxury tax redistributes hundreds of millions of dollars between teams and this is likely to result in some resentment among the owners, especially as many high-spending teams post losses totaling in the tens of millions of dollars.  It will be interesting to see how this all plays out.

 

CAVEATS:

 

This analysis represents the best that I can do with the data available to me, but as I have made analysis more widely available, my information sources have improved considerably.  In some cases, data completely unavailable to me would substantially change my estimates, but in most cases better information probably would just result in me refining my estimates more.  For the most part, I have tried to incorporate this source of uncertainty into my estimates and ranges.  But let me discuss a few caveats.

 

Is the increase in non-national TV BRI in 2002-03 largely due to positive revenue shocks that are unlikely to occur in later years?  Is there no reason to expect similar shocks in later years?

 

Clearly, BRI projections are crucial in this entire analysis.  If the 2001-02 and 2002-03 increases in non-national TV BRI are entirely due to positive revenue shocks that are unlikely to occur in 2003-04 and 2004-05 and nothing similar is likely to happen in 2003-04 and 2004-05, then the chances of seeing the low BRI estimates above (or something worse) increase.  This would also be true if the League is expecting large negative revenue shocks in 2003-04 and 2004-05.  Recent legal issues involving NBA players are one such possible negative revenue shock.

 

There also has been some mention that local TV deals may be decreasing for various reasons, including the fact that the national TV deal grants exclusive rights to ESPN and TNT to certain popular nights for advertisers.  However, remember that there are likely to be positive revenue shocks that counterbalance some of these negative shocks.  (I listed some of these above.)

 

One other related point.  I have heard that local media revenue increased by about $50 million in 2002-03 after not increasing in previous years.  One source informed me that part of this may be due to a one-time increase in Blazers’ media revenue.  If this is a one-time increase, then there is a possibility that I may want to lower some of my estimates of non-national TV BRI growth.  On the other hand, this same source also pointed out that the revenue from RealNetworks might be increasing fairly rapidly and that NBA-TV might not lose as much money as feared, so it is also possible that I am underestimating BRI growth.

 

I would greatly appreciate any speculation that anyone has about this or any other BRI shocks.

 

Do players with career-ending injuries, such as Terrell Brandon, count in the teams’ salaries that determine whether or not the luxury tax takes effect?

I have assumed that once a player no longer counts towards team salary, he no longer counts in the determination of whether the luxury tax takes effect.  Thus, in 2002-03 Terrell Brandon and Bryant Reeves counted in the determination of whether the luxury tax took effect, but Luc Longley and Larry Johnson did not.  This assumption seems most consistent with the data that the League is reporting.  It also seems more consistent with the collective bargaining agreement.  I also have received confirmation from several sources that these salaries do not count.  However, counting these salaries would cause me to increase my projection of the probability of the luxury tax in 2003-04 up to about 90 percent.   It would not have a big effect on my 2004-05 estimates and none on my 2005-06 estimates.  And since there are likely to be players who will sustain career-ending injuries that I am presently counting, my probabilities of the luxury tax in 2004-05 and 2005-06 likely are overstated.

 

Does salary lost due to suspensions count in team salary calculations for determining whether the luxury tax is triggered?

 

I have assumed that it does, but if this salary does not count (and there are good reasons for thinking that it might not), then I would lower my projected probability of the luxury tax being triggered in all seasons, but in particular in 2003-04.

 

Part of the salary of veterans signing at the minimum salary is subsidized by the League.  I assume that these subsidies do not count in team salaries, but do they count in total salaries and benefits when determining whether or not the luxury tax takes effect?  Are they part of benefits?

Overall, this probably only amounts to $5 to $10 million, but I was just wanting to verify how these subsidies are accounted for (or not accounted for) when determining whether the luxury tax takes effect.


Does Charlotte's revenue count in 2004-05 when determining whether the luxury tax takes effect?  Does it count as a team, pushing the total number of teams up to 30?

I have assumed no for both of these questions.  Above I also account for the fact that Charlotte does not count for the first two years in projected BRI calculations, which are used to calculate the salary cap.  Charlotte is counted in the 2005-06 BRI calculations, since I have been told that they will fully participate in the luxury/escrow tax system in 2005-06.  In 2004-05 they will not receive any luxury/escrow tax distributions nor will they pay any luxury tax.  (I presume their players will still pay escrow tax.)

 

Also, I assume that Charlotte contributes $47 million in new revenue in 2005-06.  I have tried to be conservative with this estimate, and I would appreciate any better estimates that others may be able to provide.

 

Will the salaries of players that Charlotte selects in the expansion draft but waives prior to the beginning of the 2004-05 season count in the determination of whether or not the luxury tax is triggered?

 

According to Larry Coon, they will not count.  This is what I believe as well, but for the most part this is not incorporated into my analysis above.  If I was able to incorporate it, most likely this would lead to even lower estimates of the luxury tax being triggered in 2004-05 and 2005-06.

 

And here is a question that isn’t directly related to this analysis, but I would greatly appreciate an answer if anyone can provide it.  Rick Horrow reports that sales of NBA-licensed merchandise increased from $1.8 billion to $2.1 billion last year.  My impression is that vast majority of this revenue is not counted in BRI.  I was wondering how much the League, the teams, the NBPA, and the players get of this licensing revenue and how much is counted in BRI?

So far, the best information from an old source on this is that licensing fees average nearly 5 percent of merchandise sales, which would mean that an additional $105 million NBA revenue, most of which is not counted in BRI.  Some of this estimated $105 million may be shared with the players.  Other sources peg this percentage at higher and lower percentages.


If you have any questions or comments about this analysis, feel free to call me or e-mail me.  When possible, I will incorporate comments on this analysis.

 

** I thank Larry Coon, David Gersovitz, Jeff Kramer, Andy Stein, Laurel Tofflemire, and other anonymous individuals for comments on this analysis.  All errors or mistakes, however, remain my own.  Also, all conclusions are my own and do not reflect on anyone who might have aided me in this analysis.  This analysis can be used in whole or in part, as long as (1) this piece is credited to Dan Rosenbaum, (2) a link to the latest version of this analysis (http://www.uncg.edu/eco/rosenbaum/luxtax.html) is included, and (3) I have given express permission for this use (e-mail me).

Webmaster: Dan T. Rosenbaum

Return to Home Page
This page last edited 4/26/04