Thursday, May 03, 2012
Does the NBA Draft Lottery Need Welfare Reform?
In one of the coolest recent efforts in crowd sourcing, ESPN's basketball blog "True Hoop" has been running a project called "Hoop Idea,", The blog discusses some structural problems with the NBA and asks readers to tweet solutions to #HoopIdea or email them to the editors (and even the questions have been crowd sourced). One recent effort has focused on an on-going problem in the NBA. Teams that are out of the playoff hunt "tank" games, or don’t try to win. They don’t throw them in a traditional sense – that would violate league rules – but they find other ways to do something other than their best to win (for instance, they bench their best players, claiming that they are suffering from fake-sounding injuries). The reason teams try to lose is to get a better position in the NBA draft, in which the teams are put in worst-to-first order to select new players out of college or international ball (the top three slots are determined by lottery among the worst 14 teams, but the number of lottery balls each team gets is based on how bad its record was the season before.) When teams tank, their lack of effort messes up the competitive balance in the league -- the teams that get to play the tanking teams at the end of the season get an advantage. And the lack of entertainment created by teams trying to lose is an externality, as the whole league shares a national TV contract.
True Hoop has gotten ideas from all sorts of people, but as far as I can tell, no lawyers, tax experts, or public economics scholars have gotten into the act. Which is too bad because the problem of tanking is effectively the same (or at least is similar) to the incentive problems created by any type of public assistance.
When designing a welfare program, the desire to redistribute money always has to be balanced against the incentive effects it creates for those receiving help. The big problem is usually with how aid is phased out. For instance, if there is a sharp cutoff -- say, everyone who earns under $15K gets benefits but if you earn over $15k, you get nothing -- then policy has effectively created a really high marginal tax rate at $15K, giving people who earn below that amount little incentive to engage in extra work. (That's a made-up example, but effective can exceed 100% due to cut-off benefits.) The desire to redistribute ends up messing up the incentives of those to whom you want to redistribute. Which is effectively the problem in the NBA with tanking. The league wants to redistribute talent to the worst teams, but the existence of the draft messes up their incentives, harming everyone.
While I'm not an expert here, a few tools that are used and discussed in debates about welfare seem like they could be translated into the NBA context.
Below the fold, see three proposals, modeled on 1996's welfare reform, a proposal to stretch redistribution out into the middle class, and the Earned Income Tax Credit.
Welfare reform: Two key elements of the 1996 welfare reform were requiring recipients to work within two years of receiving benefits and putting a time limit of five total years on federal benefits. Whatever one thinks of these in terms of federal policy, they do suggest NBA reforms.
The lifetime ban is easiest. Put a limit on the number of years in a row a team can appear in the draft lottery (which still wouldn't be as strict as federal welfare policy). If a team doesn't make the playoffs (the cutoff for the draft lottery) for three consecutive years, bar it from receiving a lottery pick until it does make the playoffs. Think this is too strict? Just create a temporary ban -- a season or two. This will give teams all sorts of incentives not just to try hard at the end of a season, but also to try to sign better players at the start of a season so they can avoid being banned from the draft lottery.
But I think the work requirement is even more interesting. What we are worried about in the context of tanking is that the game will get less interesting. We have a pretty good method for telling whether the games are interesting -- attendance. It is really hard to get people to show up to games when stars aren't playing and the outcome is certain. What if the league required teams to meet certain attendance figures in order to get lottery balls/draft spots? If attendance fell below 10K a game, you lose X number of lottery balls and fall in draft order (you'd have to control for market size and ensure that they weren't giving away tickets, but that doesn't seem too hard). This way bad teams have an incentive to put a good product on the court. Or even more interestingly, they can tank, but only if their fans are in on it. This would also goose attendance and revenue.
Stretch it out: One answer to the problem of phasing out benefits is to slowly remove benefits, to remove only a few cents of benefits for each dollar earned. We could do the same in the NBA. The league could provide lottery balls -- that is, chances to get one of the top three picks -- to all teams, reducing on a slowly sliding scale, reducing the marginal benefit to losing games.
Notably, this would face the same problem as stretching out solutions do in welfare programs. By definition, stretching out assistance means giving aid to more people. This means the program costs more or, alternatively, redistributes less. In this case, the amount of redistribution is fixed (there are only so many new good players to allocate), and the result of stretching out benefits would be that the draft redistributed less.
Earned Income Tax Credit: The Earned Income Tax Credit provides a refundable tax credit designed to make joining the work force worth it. (this post is getting long, so I’ll leave it to the NBER for a longer explanation and empirical analysis) . The basic idea is that, among the poor, we want to create incentives for joining the formal work force. Same logic applies for the NBA draft lottery. P.H.D. student Adam Gold has argued that we should reward wins, rather than losses. As summarized by True Hoop, Gold argues that the NBA should give:
“the first pick in the draft to the team that wins the most games after being officially eliminated from playoff contention. Then the team with the second highest number of wins gets the second pick. And so on. The theory is that the worst team in the league will be the one that is mathematically eliminated first. Thus, it will get the most chances to pile up wins. If it takes advantage of those opportunities, it will be rewarded with the No. 1 pick.”
The theory strikes me as similar to the logic of the EITC, an attempt to reward effort inside a redistributive system.
As I said, I'm not an expert here. But it strikes me that this is pretty fertile territory. Thoughts?
Posted by David Schleicher on May 3, 2012 at 09:09 AM | Permalink
TrackBack URL for this entry:
Listed below are links to weblogs that reference Does the NBA Draft Lottery Need Welfare Reform?:
These are all good. The NBA's goals are a little different than those of traditional transfer programs, though, in that they don't really care about distributive justice -- they're all about ratings, and hence competitive balance. That's a problem for Gold's proposal, and for the similar ideas from Simmons about having a loser's tournament for the top pick: the worst teams get worse picks.
You know I like price instruments, so no surprise my suggestion would be trying one here. What if the NBA charged teams an increasing percentage of their tv revenues for top lottery picks? That seems like the closest thing to internalizing the tv-revenue externality you identified.
Posted by: BDG | May 3, 2012 10:31:25 AM
Thanks for this post. We are hungry for all kinds of ideas, and especially those with a track record. Strategies that have been deployed elsewhere and may have applications in the NBA, I love that stuff!
Anyone reading this who has thoughts ... firstname.lastname@example.org is a great way to reach us.
Posted by: Henry Abbott | May 3, 2012 10:32:27 AM
Brian -- The key for me is not the goal but the question of how redistribution can be achieved without messing up incentives. But you're right that the reasons are different.
I like the price instrument idea a lot, but there's one hiccup -- money is a direct input in competitive success (can you pay the luxury tax? Do you spend on coaches, training facilities etc.) Also, would the charge count against the salary cap? What would happen to the money? Also, wouldn't a more direct method be to charge them based on ratings in games on their opponents local tv contract? Very interesting stuff..
Posted by: D.Schleicher | May 3, 2012 10:50:29 AM
There's a simpler, easier, and better solution, I think: A hard salary cap with no draft. I make that argument here:
Posted by: Frank Snyder | May 3, 2012 12:36:37 PM
Brian (& David, & Frank):
That assumes the "league" -- which doesn't exist as a coherent entity, see, e.g., the most recent labor dispute and the splits within ownership and with the players (and within the union, too!) -- wants competitive balance. To grow the pie, the league needs dynastic teams built on marketable superstars in major media markets. Superstars, too, want the opportunity to play, preferably with each other, in major metropolitan areas.
Also, tanking doesn't affect national tv contracts, as they too are built upon the revenue generated by top teams playing each other in games scheduled for national broadcast (playoffs) and cable (regular season and playoffs) networks. Charlotte, I imagine, has hardly been on TNT or ESPN all year, and certainly not after they revealed themselves to be one of the worst teams in NBA history. There might be some externalities for the teams they play (diminished game day ticket sales and lower tv ratings), but the subscriber fees and advertising rates aren't moved much by a late season game with a tanking team, and it's possible that a playoff bound team might be willing to trade some diminished ticket sales -- which are likely low in number, since many of the top teams have pre-sold seats via season tickets and multi-game packages -- for an easy late-season win.
For small market teams (hello, Cleveland), tanking is rational. I wish my beloved Milwaukee Bucks (a very poorly-run franchise, to be sure) had tanked this year rather than make a run for a low playoff seed and a certain first round exit. It might not have worked, but it would have given me more hope for the near- and middle-term. And as for getting rid of the draft, one of the draft's key attractions for ownership is being able to mitigate the risk of acquiring new talent with low salaries for a given number of years, all set and locked in by the collective bargaining agreement. That too helps veterans (which is to say, union members) to grab a larger piece of the salary pie. Everyone loves it, except the kids who get treated like chattel, but they have no voice in the league.
Posted by: Mark Fenster | May 3, 2012 8:56:06 PM