I was halfway through a draft post on some recent books when I read Brian Goff’s recent entry ‘LeBron “Stumbling on Losses”‘. Brian has highlighted some of the interesting arguments in the new work by David Berri & Martin Schmidt, “Stumbling on Wins” (2010, FT Press).  For the lay reader, ‘Stumbling on Wins’ is a very good introduction to the myriad of ways that people make sub-optimal, or irrational decisions on the sporting field and when recruiting, drafting or paying players and coaches.
Brian mentioned the tensions often found within the discipline of economics between behavioural and neoclassical approaches to economics, and this leads nicely into another superb book I recommend, by economics & finance journalist Justin Fox: “The Myth of the Rational Market” (2009, Harper Business). Fox has produced a highly readable account of the tensions within Wall Street offices, academic finance departments, and the economics profession at large between the simplifying assumption of the efficient market hypothesis and the reality that stock market noise cannot be simply described as a series of random walks, because people do, in fact, act irrationally in ways that produce non-random patterns in supply, demand and prices.Â
When you put these works together, I can’t help but wonder when the most prominent efficient markets hypothesis equivalent in sports economics – the zero transaction cost version of Rottenberg’s invariance proposition / Coase’s theorem will be torn down by scholars, journal/book editors and reviewers. Many have nibbled around the edges over the past 20 or more years, yet still the zero transaction cost assumption lives on all too often.
Sports economists are increasingly challenging the simplifying assumptions in the models of this discipline’s originating scholars; yet many prominent books and articles published this century highlight how far we have to go. Berri & Schmidt (2010) provide some good accounts for the layperson of how economists are identifying the effects of irrational behaviour in sport. But the edifice of the zero transaction costs assumption still retains prominence in both models of sporting leagues and labour markets, and the accepted leading commentaries on those models; for one, the excellent “Handbook on the Economics of Sport” (2007, Edward Elgar), can’t even find enough to justify an index entry for ‘transaction costs’. Surely that is irrational.
So what’s the point?? I’m all for destroying the Efficient Market nonsense! All one had to do was watch CNBC a week ago when the Dow fell nearly 1000 points and there was pure panic shooting from every hyperextended vein on that financial network. Markets only appear to be rational in hindsight…something like first round picks in the NFL that bust out…a year after the wasted choice any number of “insightful” commentators will tell fans what they saw as fatal flaws in the player well before he was chosen. Thank you so much!
Sure, many are wise after the event Greg. ‘So what is the point’, you ask? Two points:
(a) here are a couple of books on the questions of efficiency and irrationality that people might enjoy reading to stimulate their own thinking (or perhaps add to an economics syllabus to stimulate the thinking of the next generation of economists), and
(b) noting that there is a ways to go in developing robust economic models of a sports league to accommodate positive transaction costs. That highlights a challenge for those with the ability to model a sports league and the interest in doing so.
That Berri, Schmidt & Fox have been able to write thoughful and entertaining commentaries on the working of markets and the occasional shortcomings of market participants in a manner easily absorbed by the lay reader is something to be applauded, not dismissed.
What do you think of the current “Who Will Sign Lebron” markets at WSEX and 5dimes?
WSEX:
CLEVELAND +170
NEW YORK +284
CHICAGO +354
NEW JERSEY +733
MIAMI +900
LOS ANGELES C +3233
DALLAS +3233
5dimes:
Lebron James signs with Cavaliers +140
Lebron James signs with any other team -180
Interesting, thanks Harmy; I’m no NBA expert.
Hi Robert,
In reading your post it seemed there are really two separate issues. The first is whether the market makes mistakes in evaluating player talent or in putting together winning teams. The clear and obvious answer is yes. Player abilities and how performance translates from one level, say college, to the next, professional, is an inexact science. Ex post mistakes are sure to happen. Ex ante, are clubs and franchises making systematic and correctable errors? There is evidence that they do from a variety of sources. The interesting question is whether once they learn of these mistakes if they correct them. Skip and Jahn Hakes provide evidence that the mistakes identified in Moneyball were corrected in fairly short order. It is not clear if NBA executives have been able to readjust their thinking based on the evidence, and it may take an NBA insider to “discover” and correct the mistakes, and have success with the strategy before those mistakes are widely corrected.
The second issue is the continued use of the zero transactions cost assumption in models of sports leagues and sport labor markets. I don’t think anyone would argue transactions costs in sports are zero or anywhere near zero. I wonder though if any fundamental results would be substantially different if the assumption were jettisoned. For example, even with substantial transactions costs, it seems likely that by and large playing talent will end up on the clubs where it is most valued. Even with large transactions costs, it seems likely that salary caps, luxury taxes, rules about free agency and arbitration will have generally the same impact as in models that assume no transactions costs. (Such rules are, of course, also definitionally sources of transactions costs.) So on this second point, I am not sure what one gets from pitching out the zero transactions cost assumption.
Hi Dennis, thanks for your thoughtful contributions. It’s useful to play out these ideas.
I’ll jump straight to your second paragraph, for your first para stands alone as a nice summary of some relevant findings.
I question whether the existence of substantial transaction costs would, as you contend, leave us withe the same outcomes. At the minimum, I’d argue that while this outcome could be so in a model, I am not sure if that stands up to empirical analysis.
For mine, there are enough empirical examples in the world to suggest that positive TC’s are a factor worthy of empirical analysis in both the governance and labour market contexts, and subsequent incorporation into formal models of both (i) governance structures and (ii)labour market regulations. But no doubt many will agree with your competing view Dennis.
I don’t recall too much literature attempting to introduce positive TC’s to see what the effect would be upon (i) allocation of decision-making rights and residual claimancy rights, or (ii)wages, allocation of labour, winning percentages and competitive balance. Even a body of literature suggesting that model outcomes will be the same with positive and zeros TC’s seems to be be missing for the most part, so I regard that as a gap in our formal literature.
Modelling of leagues seems to jump forward every so often by incorporation of new ideas from outside of the orthodoxy. Even if the ‘wilder’ transaction cost economics of Williamson is a step to far and we feel safer within the Property Rights paradigm, surely it is time to bring all of Coase’s theory properly into the sports economics tent; lest we be smited by Coase himself, who suggested that detailed analysis of a zero transaction cost world is an exercise akin to divining the future by minute inspection of the entrails of a goose.
Hi Robert,
In the interest of possibly moving toward the modeling you suggest, I wonder how you would define transactions costs and place them into a model of sports leagues or sport labour markets. For example, I would include among the transactions costs in a sport labour market the resources devoted to negotiations, say the compensation of the lawyers/agents for the club and the player. I would not include things like amateur drafts, salary caps, or roster size restrictions which have been analyzed in any case.
I recognize there are also principal-agent issues, such as differing objectives of agents and the players they represent. I would not include them as transactions costs, though perhaps I should.
I hope that by asking such questions I have not begun butchering the goose.
Just want to say what a great blog you got here!
I’ve been around for quite a lot of time, but finally decided to show my appreciation of your work!
Thumbs up, and keep it going!
Cheers
Christian, iwspo.net