Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Download
Standard view
Full view
of .
Look up keyword
Like this
2Activity
0 of .
Results for:
No results containing your search query
P. 1
Cheating in Contests

Cheating in Contests

Ratings: (0)|Views: 31 |Likes:
Published by baschance

More info:

Published by: baschance on Nov 26, 2008
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

06/05/2010

pdf

text

original

 
612
 © 2003 OXFORD UNIVERSITY PRESS AND THE OXFORD REVIEW OF ECONOMIC POLICY LIMITED
CHEATING IN CONTESTS
OXFORD REVIEW OF ECONOMIC POLICY, VOL. 19, NO. 4
IAN PRESTON
University College London
STEFAN SZYMANSKI
 Imperial College London
1
 Much of the interest in the study of sports from the perspective of an economist lies in the empirical applicationof contests as efficient mechanisms for eliciting effort. Contestants respond to contest incentives, and theseincentives include the incentive to cheat. This paper discusses different forms of cheating: sabotage, doping,and match fixing. The paper discusses how these forms of cheating arise and how they can be treated. In particular, we look at specific forms of cheating in soccer, baseball, and cricket. In the appendix wedevelop a simple model of match fixing.
I. INTRODUCTION
Commercial sport is a distinctive form of an eco-nomic contest. Unlike most economic contests,demand arises more from interest in observation of the contest itself (e.g. a race or a match) than in theoutcome of the contest. In the economics literature,the use of a contest as a means of eliciting effortcontributions stems from the perception that in aconventional principal–agent framework any con-tract will be second best in the face of asymmetricinformation. If only the agent can observe effortaccurately, then a reward scheme based on inputsis subject to moral hazard (the hidden-action prob-lem), while an output contract is likely to imposeexcessive risk on the agent. If the ability of the agentis also private information, then even with a risk-neutral agent the first-best contract cannot be im-plemented owing to the combination of moral hazardand adverse selection (the hidden-information prob-lem; see, for example, Laffont and Tirole, 1993).The insight of the contest literature is that rewardingagents according to their relative, rather than abso-lute performance can help to overcome these incen-tive problems. However, in a sporting contest ob-served by thousands, possibly millions, of specta-tors, the possibilities for on-the-job shirking arerelatively limited. For the vast majority of athletescompeting at the higher levels of a sport, where mostof the commercial interest lies, shirking will rapidly
1
We are grateful to Pascal Courty, Hamish Low, Tommaso Valletti, and the participants at the editorial conference for helpfulcomments. Errors are our own.
 
lead to omission from future contests. However,asymmetric-information problems are not entirelyabsent, and in this paper we discuss three of them,all of which fall under the broad heading of ‘cheating’.(i)
Sabotage
. Contestants win if their perform-ance is better than that of their rivals. Reducingthe performance of rivals may be as effectivea means to achieving this end as improvingone’s own performance. In some sports, this isan accepted part of the game, but often itfrustrates the wish of spectators to observeopponents exercising skills to the full and isaccordingly illegal under the rules of the sport.Typically, organizers employ umpires or ref-erees to monitor and punish sabotage activities.However, sabotage may also take the form of deceiving officials.(ii)
 Doping
. Contestants can improve their prob-ability of winning by the right kind of prepara-tion, in relation to training and diet. Manysubstances have the potential to enhance ath-letic performance, but only some of these aredeemed to be acceptable within the spirit of thesporting contest. However, detecting the con-sumption of banned substances is problematic.As with sabotage, contest organizers attempt todiscourage such activities through monitoring,but in general this is a much more difficultproblem than with on-the-field sabotage.(iii)
 Match-fixing
. Individual contestants may bewilling to reduce their effort contribution forspecific matches if the rewards for so doing arelarge enough. Sometimes this occurs eitherbecause the opposition values the victory sig-nificantly more and is willing to pay to secure it,and sometimes it occurs because there is anopportunity to generate returns on the insiderinformation (for example, through gambling).Match fixing is felt to violate the spirit of thegame and is also perceived to undermine spec-tator interest, and is, therefore, prohibited byorganizers.Broadly, these kinds of cheating fall into two catego-ries. First, there is ‘cheating to win’, where it mightbe argued that the incentives inherent in the contestreward scheme have ‘gone too far’ (a genericproblem discussed in a wider social context byFrank and Cook, 1996). Second, there may be aproblem of ‘cheating to lose’, such as the casewhere a team is bribed to lose a match. Here, theproblem is that the direct contest incentives areoverwhelmed by some external incentive. In thispaper we explore each of these forms of cheating intheory and in practice, and discuss what optionsthere are to discourage such activities.
II.WHAT IS WRONG WITHCHEATING IN SPORT?
Arguments for not permitting cheating in sporttypically fall into two kinds.
 Legal and ethical
. Cheating in the formsdescribed above is typically harmful or fraudu-lent to varying degrees. Assaulting fellow com-petitors is frowned on for obvious reasons.Fixing matches for gambling gains is fraudulentand harmful to those who lose money as aresult. These sorts of behaviour violate mostreligious and other ethical codes, but the desireto win or to make money is so great that in manycases ethical codes have a limited ability torestrain cheating. In more extreme cases, theseactions violate civil or criminal legal codes.Bribing players and officials is typically illegal,as is assault. Thus, even without the interven-tion of the sporting authorities, there exists bothmonitoring (in the form of the police and otherpublic agencies) and punishment for cheating.
Commercial
. Cheating undermines interest inthe sport. This is itself an interesting and test-able hypothesis, although we know of no em-pirical research. While it is possible to think of many cases where cheating scandals havegenerated ‘bad’ publicity, it is also possible tothink of many cases where the prevalence of cheating does not seem to undermine interest.For example, the disciplinary record of playersin many sports has deteriorated over recentdecades, but there is little evidence that this hasdiminished spectator interest. Indeed, it may beargued that causation runs in the opposite direc-tion—increasing (decreasing) interest in a sportleads to increased (decreased) cheating.
613
I. Preston and S. Szymanski
 
614
OXFORD REVIEW OF ECONOMIC POLICY, VOL. 19, NO. 4
III. SABOTAGE
Sabotage as an activity has been considered in bothindustrial and labour economics. In the context of competition between firms, sabotage may be thoughtof as an act of raising rivals’ costs (Salop andScheffman, 1983). In the context of an internallabour market, sabotage may be one kind of re-sponse to the use of tournament incentives (Lazear,1989). In the first case, sabotage is clearly beneficialto the perpetrator since it weakens competition.However, in the latter case the effect is ambiguous,since, although the sabotage may have a directeffect on expected returns by raising the probabilityof winning the prize, if the sabotage indirectlyaffects the productivity of the firm then this may, infact, reduce the expected income of the firm. Forthis reason, Lazear suggested that employers willreduce the spread between winning and losing inorder to ensure that co-workers cooperate. A simi-lar argument can be applied to a sporting competi-tion. The attractiveness of a sporting contest de-pends on the balance of the competition (the uncer-tainty-of-outcome hypothesis) and the quality of theperformance. If sabotage reduces the quality of theperformance of the opposition, then, even if itincreases the contestant’s probability of winning, itmay reduce the contestant’s expected return. This,then, would seem to be an argument in favour of reduced incentives for winning.Sabotage in sports can take a number of forms:
illegally restraining or assaulting competitors;
attempting to provoke illegal responses fromcompetitors (e.g. by goading);
attempting to persuade the referee that oppo-nents have engaged in illegal acts.
2
Each of these tends to undermine the attractivenessof a sporting contest to spectators, because it limitsthe opportunity to observe the skills of the opposingteam or generally slows down the game with toomany interventions from the officials. Thus, sabo-tage reduces productivity of the sporting competi-tors and makes their joint product less attractive.Garicano and Palacios-Huerta (2000) suggest thata recent change in the rewards for winning in leaguefootball provides a natural experiment to test thehypothesis that increasing the prize spread increasessabotage activity. During the 1980s most nationalsoccer moved from a system of awarding two pointsfor every match won to three, with the intention of increasing the incentive to win matches.
3
Garicanoand Palacios-Huerta found that the change wasassociated with a significant increase in sabotageactivity measured by the number of yellow cardsawarded.
4
They also found no tendency for thenumbers of goals scored to increase, hence con-firming the insight of Lazear, that a larger prizespread can lead to increased sabotage and reducedproductivity. However, there is no evidence that thisled to any reduction in fan interest.Sabotage is a particular problem in contests withsmall numbers of players, a point stressed by Konrad(2000). In a two-person contest, any reduction in theopponent’s probability of success leads to a one-for-one increase in one’s own probability of success.However, with many contestants, sabotage aimedat a particular rival provides an externality for all theother rivals and only a small gain for the perpetrator,making sabotage unlikely to be privately profitable incontests with large numbers. This suggests thatcontest designers should seek to organize contestswith many competitors. Of course, in a soccer orbaseball match there can only be two teams, andthese institutional constraints are binding. There isalso a more subtle problem with this prescription.Increasing the number of competitors in a contestcan be shown, for a wide variety of plausiblefunctional forms, to reduce individual effort, andpossibly even aggregate effort (Nti , 1997). Thus,
2
To this list we might add the case of deceiving the referee as to what has occurred in the game. For example, it has been traditionalin cricket that, whenever a batsmen knows himself to be out, he should ‘walk’, i.e. quit the field, even if the umpire is unable todetermine whether the batman truly was ‘out’. However, in recent years this tradition has more or less died out, and most batsmenwait for the umpire to decide, and may even try to suggest by their body language that they were not, in fact, ‘out’. However, inmost professional sports it has long been the case that umpires and referees pay no attention at all to the opinions of the players.
3
In the world of soccer a draw (or tie) has always been considered a legitimate outcome. In the USA, sports leagues have tendedto adopt rules that guarantee each match produces a winner, thus avoiding the problem.
4
In soccer a yellow card is shown to player who commits one of a number of offences, principally related to illegal tackles onopponents or seeking to intimidate the referee. Two yellow cards shown to a player in the same game leads to dismissal from thepitch, and accumulating yellow cards over a number of matches can lead to suspension.

You're Reading a Free Preview

Download
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->