CARROTS AND STICKS: PRIZES AND PUNISHMENTS IN CONTESTS

We study optimal contest design in situations where the designer can reward high performance agents with positive prizes and punish low performance agents with negative prizes. We link the optimal prize structure to the curvature of distribution of abilities in the population. In particular, we identify conditions under which, even if punishment is costly, punishing the bottom is more effective than rewarding the top in eliciting effort input. If punishment is costless, we study the optimal number of punishments in the contest. (JEL D44, D82, J31, J41)     

PROPERTY RIGHTS AND LOSS AVERSION IN CONTESTS

We analyze the effects of property rights and the resulting loss aversion on contest outcomes. We study three situations: in “gain” two players start with no prize and make sunk bids to win a prize; in “loss” both the players start with prizes and whoever loses the contest loses their prize; and in “mixed” only one player starts with a prize that stays with him if he wins, but is transferred to the rival otherwise. Since the differences among the treatments arise only from framing, the expected utility and the standard loss aversion models predict no difference in bids across treatments. We introduce a loss aversion model in which the property rights are made salient, and as a result the reference point varies across treatments. This model predicts average bids in descending order in the loss, the mixed, and the gain treatment; and higher bids by the player with property rights in the mixed treatment. The results from a laboratory experiment broadly support these predictions. There is no significant difference in bids in the loss (gain) treatment and bids by property rights holder (nonholder) in the mixed treatment. A model incorporating both loss aversion and social preferences explains this result. (JEL C91, C72, D23, D74)     

CHEATING IN CONTESTS

Wherever competition is used to motivate a desirable activity or productive effort it may also motivate undesirable and therefore prohibited behavior—that is, cheating—that the organizer of the contest attempts to police. For example, when workers compete for promotion, bonuses, or other rewards, they may misrepresent their output (i.e., commit fraud) or increase their output by unacceptable means (e.g., violate regulations). We show how the extent of cheating is determined by the payoffs at stake in the contest, the random component of output, probability of cheating being detected, number of contestants, and the penalty associated with being found to have cheated. We find that while greater enforcement reduces cheating, it may also reduce productive effort. We also identify how two particular aspects of enforcement, the awarding of default victories and use of correlated rather than independent audits, affect cheating behavior. (JEL J33, K42)     

CAPS IN SEQUENTIAL CONTESTS

We study a sequential two‐stage all‐pay auction with two identical prizes. In each stage, the players compete for one prize and each player can win either one or two prizes. The designer may impose a cap on the players' bids in each of the stages. We analyze the equilibrium in this sequential all‐pay auction with bid caps and show that capping the players' bids is profitable for a designer who wishes to maximize the players' expected total bid. (JEL D44, D82, J31, J41)     

RISK TAKING IN CONTESTS AND THE ROLE OF CARROTS AND STICKS

We study contests in which contestants choose both work effort and the variance of output (risk). Winner-take-all contests generate incentives for contestants to engage in costly risk taking, which is inefficient if the contest organizer values the aggregate output of all contestants. The addition of a penalty for ranking last (retaining a prize for ranking first) enables the organizer to independently control contestants' incentives to exert productive effort and to increase output variance. In this way, the organizer can eliminate risk-seeking behavior in settings where it is wasteful, but also control risk seeking when it is desirable, such as in research tournaments. ( JEL J33, C72)     

CHAOTIC DYNAMICS IN CONTESTS

Under a myopic best‐reply dynamic, efforts in repeated contests may exhibit chaotic behavior. This may help explain, for example, why experimental data often show nonconvergence to one‐shot equilibrium efforts. (JEL C61, C72, D72, D74, D83)     

INCENTIVES AND BEHAVIOR IN ENGLISH, DUTCH AND SEALED-BID AUCTIONS

The Pareto optimality and price behavior of English and Dutch oral auctions, and First-Price and Second-Price sealedbid auctions are compared under various procedures for assigning valuations among cash motivated bidders. The Vickrey propositions with respect to the mean and variance of prices under the English, Dutch and Second-Price auctions are not falsified by the data. Individual behavior and prices in the First-Price auction deviates considerably from Vickrey's Nash postulate. Behaviorly, the English and Second-Price auctions appear to be isomorphic, but the Dutch and First-Price auctions may not be isomorphic.     

ENDOGENOUS TIMING IN CONTESTS WITH DELEGATION

We study two‐player contests in which each player hires a delegate, and the delegates decide endogenously when to expend their effort. First, we look closely at the delegates' decisions on when to expend their effort, given contracts between the players and the delegates, and look at the players' decisions on their contracts. Then, we compare the outcomes of the endogenous‐timing framework with those of the simultaneous‐move framework. We show that the higher‐valuation player offers her delegate greater contingent compensation than her opponent, the delegate of the higher‐valuation player chooses his effort level after observing his counterpart's, the equilibrium expected payoff of the delegate of the higher‐valuation player is greater than that of his counterpart, and economic rent for each delegate exists. We show that, in the endogenous‐timing framework, each player offers her delegate better contingent compensation, each delegate's expected payoff is greater, and each player's expected payoff is smaller, as compared with the simultaneous‐move framework. (JEL D72)     

CONTESTS WITH DRAWS: AXIOMATIZATION AND EQUILIBRIUM

We introduce and axiomatize a class of single‐winner contest success functions that embody the possibility of a draw. We then analyze the game of contest that our success functions induce, having different prizes delivered in the occurrence of a win and a draw. We identify conditions for the existence and uniqueness of a symmetric interior Nash equilibrium and show that equilibrium efforts and equilibrium rent dissipation can be larger than in a Tullock contest (with no possibility of a draw) due to increased competition even if the draw‐prize is null. These results suggest that a contest designer may profit from introducing the possibility of a draw. Finally, we show that this approach naturally extends to multiprize contests with multiple draws across different subsets of the set of players. (JEL C72, D72, D74)     

INCENTIVES AND OUTCOMES IN A STRATEGIC SETTING: THE 3-POINTS-FOR-A-WIN SYSTEM IN SOCCER

I exploit a major structural change that has occurred in world soccer to study the impact of incentives on outcomes in a strategic setting. A game-theoretic model is developed that captures some essential strategic elements of soccer vis-à-vis the number of points awarded to a win. The observable implications of the model are tested using a large data set that spans 30 years and 35 countries. The empirical results support the theoretical model and show that the 3-point system has led to a statistically significant increase in the expected number of goals and a decrease in the fractions of drawn matches. ( JEL C72, L83, C23)     

INCENTIVES, PREDESTINATION AND FREE WILL

Without wholeheartedly endorsing Max Weber's views on Protestantism and economic growth, we present a model formalizing his claim that belief in predestination was a component in economic development. Our model suggests that belief in predestination is more effective when heterogeneity of beliefs is high, when the desire for homogeneity is high, or when accurate signals about ethics are important. Using the General Social Survey, we find that Protestants generally and Presbyterians in particular exhibit a stronger connection between individual and group behavior and between worldly success and church attendance. ( JEL A13, Z10)     

TIME‐USE AND ACADEMIC PEER EFFECTS IN COLLEGE

This paper examines academic peer effects in college. Unique new data from the Berea Panel Study allow us to focus on a mechanism wherein a student's peers affect her achievement by changing her study effort. Although the potential relevance of this mechanism has been recognized, data limitations have made it difficult to provide direct evidence about its importance. We find that a student's freshman grade point average is affected by the amount her peers studied in high school, suggesting the importance of this mechanism. Using time diary information, we confirm that college study time is actually being affected. (JEL I20, F21, J01)     

ON EQUILIBRIUM PLAYER ORDERING IN DYNAMIC TEAM CONTESTS

We study equilibrium player ordering in a dynamic all-pay contest between two teams. The contest lasts two periods, and each team consists of two players who perform in different periods on behalf of their teams. The team with the higher aggregate output wins the prize, which is a public good to its players. Each team has one stronger player and one weaker player, and the two teams can differ in their values of the prize. The teams maximize their winning odds by strategically assigning their players to different periods. We find that when the intrateam heterogeneity in player ability is not excessive, the teams would allocate their stronger players to the late positions as the “anchormen.” When both the intrateam ability gap and interteam heterogeneity in teams' values become excessively large, the team with high value always places its stronger player in the early position, who will place a large bid to preempt late competition. (JEL C7, D7, D8)     

MEASURING EFFORT INCENTIVES IN A TOURNAMENT WITH MANY PARTICIPANTS: THEORY AND APPLICATION

This article provides an empirical measure to compare incentive effects of different tournament structures. Although there have been many theoretical attempts to analyze incentive effects of a tournament competition, empirical applications are limited as the provided conditions cannot easily be empirically observed. We show that the empirical density of the performance distribution can be used to measure effort incentives in a tournament with many participants. We apply this method to evaluate the grading policy change in the 2008 College Scholastic Ability Test in South Korea. (JEL C70, C51)     

OPTIMAL CONTESTS

Firms frequently use contests to compensate their employees: an employee's pay depends on the ranking of his output compared to that of others, rather than on the absolute level of his output. This paper analyzes the design of a contest which maximizes the contestants' expected aggregate output, and describes two settings which yield opposite results. In one, prizes should be equal except for that given to the contestant with the lowest output. In the other setting, only the contestant with the highest output should obtain a meaningful prize.     

WHY ZERO TOLERANCE OF MISCONDUCT IS UNDESIRABLE IN CONTESTS

Competition can foster misconduct in any circumstance where the organizer of a contest cannot perfectly monitor contestants' actions or when doing so is prohibitively costly. Although misconduct comprises all actions that are contrary to the interest of the organizer, it is not necessarily the case that it is optimal to prohibit all such behavior. In this paper, we determine the equilibrium level of misconduct chosen by players in a symmetric rank‐order tournament between two competitors in which the organizer tolerates some level of misconduct. In addition to showing that zero tolerance may not minimize the level of misconduct in equilibrium, we show that there exists a range of tolerated misconduct where a symmetric mixed strategy equilibrium exists with players cheating (i.e., misconduct above the tolerated level) with some probability. When the gain from misconduct is uncertain and unknown the contest organizer faces a tradeoff: tolerating more misconduct will reduce such behavior when the state of temptation is high, but increase it when temptation is low. (JEL J33, K42)     

PRODUCT BUNDLING AND INCENTIVES FOR MERGERS AND STRATEGIC ALLIANCES

This paper analyzes firms' choice of a merger or a strategic alliance in bundling their products with other complementary products. Tying two products of unequal value makes them equally valuable as they become inseparable for purchase. Consequently, firms can charge a higher price for the bundled products than before. If foreclosure is not the main purpose of bundling, firms would prefer strategic alliances to mergers because mergers only intensify competition by internalizing the complementarities of two products. In equilibrium, bundling occurs only through strategic alliances. (JEL L4, L11, L13, L23)     

CLASSROOM DIVERSITY AND ACADEMIC OUTCOMES

This paper tests how the race and ethnicity of one's college classmates affect academic performance. Incoming students at a Catholic college are assigned to their first semester, team‐taught, required course. Statistical tests support that this assignment is uncorrelated with a variety of student characteristics. Controlling for team fixed effects and student characteristics, I find evidence of racial peer effects that differ for white students and students of color. White students earn higher grades in classes with more students of color. Students of color with more nonwhite classmates earn lower grades; these effects occur exclusively among those with lower SAT scores. (JEL I21, I28)     

THE EFFECT OF MONETARY INCENTIVES AND FOLLOW-UP MAILINGS ON THE RESPONSE RATE AND RESPONSE QUALITY IN MAIL SURVEYS

The joint and comparative effects of the use of monetary incentivesand follow-up mailings were examined in a mail survey of suburbanWashington, DC cable television subscribers. Four experimentalgroups received monetary incentives enclosed with the firstmailing only ($0.25, $0.50, $1.00, or $2.00) and three follow-upmailings. These groups were compared with each other and againsta control group that did not receive an incentive. The resultsindicated that the response rate from the first mailing increasedsignificantly as the incentive amount increased from zero to$0.25, and from $0.25 to $1.00. Four mailings without an incentiveproduced a higher response rate than a single mailing with anincentive, but a combination of follow-up mailings and a $1.00or $2.00 incentive produced a significantly higher responserate than an equivalent number of mailings without an incentive. There was some evidence of intertreatment response bias. Largermonetary incentives tended to produce: (1) a greater degreeof effort expended in completing the questionnaires, as measuredby the number of short answers and comments provided, and thenumber of words written, and (2) comments that were more favorabletoward the survey sponsor.