Loading...
Please wait, while we are loading the content...
Similar Documents
Selection of risk and effort levels among low-stakes players : a case study in online poker
| Content Provider | Semantic Scholar |
|---|---|
| Author | Weiss, Justin |
| Copyright Year | 2010 |
| Abstract | Firms pay workers using a variety of different pay structures. The structure that governs executive pay in many instances is a tournament pay structure. This paper examines the applicability of a tournament pay structure to lower wage workers by examining the effort and risk responses of players to tournament incentives and the role these responses play in determining the tournament's outcome. Players from 19 different tournaments are observed on a hand by hand basis. It is found that players adjust effort and risk taking levels but only in response to certain incentives. This study finds evidence that tournaments are a viable pay structure for low wage workers under certain conditions. Introduction The recent financial crisis has led to debate regarding executive compensation. Executives are paid based on a tournament pay structure. The top-level executives get far more money than those who work for them. Although this is a common pay structure for executives, it is only one of many possible structures. Many workers get paid a salary reflective of their perceived value to the firm. Other workers are paid a piece-rate wage equal to the marginal revenue product of their labor. The piece-rate wage is a common form of pay structure for lower earning workers, while tournament pay structures are used mostly for higher earning workers. This study examines the applicability of a tournament pay structure to lower wage workers by examining how low stakes poker players respond to effort and risk taking incentives and use effort and risk to affect the tournament outcome. While tournament pay has been shown to increase the productivity of skilled workers, no previous studies have examined the effect of similar structures on the behavior of semi-skilled workers. If lower stakes players are sensitive to tournament incentives, an extension of tournament pay structures to semi-skilled workers could be an efficient alternative. To best determine players' sensitivity to incentives, effort and risk taking will be considered as endogenous decisions made by each player in an attempt to maximize his net payout. Literature Review Tournament or contest pay structures arguably provide incentives for workers to increase their effort and productivity, consistent with efficiency wage theory. The study of the tournament payment structure reveals that a top-heavy, or accelerated payout structure, leads to the efficient resource allocation under many conditions. For example, tournaments are found to be superior to a piece rate payment structure when it is easier to measure relative output than it is the marginal product of labor. Tournament pay structures lead to increased participant effort by incentivizing players to try and earn the disproportionately high payout for first place relative to all other ranks. The impetus for this enhanced effort has been theorized as the incentive effect. A second effect, the selection effect asserts that the institutional structure of a tournament as well as its payout structure governs who will opt to play and who will select out. The incentive effect motivates individuals who have opted into the tournament to strive hard to earn the disproportionately large rewards at the top. It will be the concern of this paper (Lazear and Rosen 1981). Lazear and Rosen (1981) find that under ideal circumstances (homogeneous players, riskneutral firms, and purely effort driven production) tournament contests will produce the same resource allocations as piece rate pay schemes. Thus, either has the potential to increase efficiency. However, tournaments are preferable if it is difficult to measure a worker's absolute output but trivial to ordinally measure a worker's output relative to his peers. Lazear and Rosen find that with risk-averse workers tournaments and piece rate pay schemes lead to different resource allocations. If a worker's production is uncertain, there is inherent risk in accepting compensation based on the marginal product of his labor. Yet, accepting payment based on the relative production rank eliminates the danger of systematic randomness. Telemarketing is an example of this given the variable quality of inputs, potential customers' numbers. In this uncertain production scenario, workers prefer the tournament pay structure because it reduces risk. Alternatively, if productivity is certain, risk-averse workers prefer a wage based on marginal revenue product of labor over the less certain tournament payout structure. In the case of workers of heterogeneous abilities, there is no feasible efficient allocation of resources because there exists an inefficient pooling of low and high ability workers (Lazear and Rosen 1981). Predicated upon these findings, several works have examined the extent of the incentive effect of tournaments including Rosen (1986) and Ehrenberg and Bognanno (1990). Rosen (1986) finds that elimination tournaments, in which there are sequential rounds of competition and the loser is eliminated in each round, necessitate a large grand prize to sustain motivation. Effort is costly, and thus players exert the level of effort that maximizes the expected value of payoffs net effort costs. Regardless of risk-neutrality or aversion, homogeneity or heterogeneity, and number of rounds, there must be a disproportionately large first prize to sustain effort in later rounds. Rosen demonstrates this is a result of two motivating factors that incentivize a player's effort: the payoff of winning in the given round and the potential payoffs of advancing to the next round. As the tournament progresses, if there is no tournament style first prize, the payoff incentive diminishes and effort slackens. Ehrenberg and Bognanno (1990) use the PGA tour, controlling for the accelerated prize structure examined by Rosen, and investigate the incentive effect of larger prize pools. Controlling for factors such as course difficulty and weather conditions, they find that for players whose membership on the tour is guaranteed, performance improves significantly with the size of the prize pool. More recent study of this prize pool effect reported by Simmons and Frick (2007) bring into question the reported magnitude of performance improvement and suggest that there is no incentive effect to larger prize pools. The incentive effect research also set the framework for the examination of the role of risk in tournaments. Hvide (2002) tests Rosen's (1986) theory that when both risk and effort can be manipulated by the player, the equilibrium would be one of low effort and high risk. Hvide (2002) studies why the theorized relative performance evaluation hypothesis is not born out in real-world payout structures. He argues that risk taking will predominate and this can have negative consequences to the firm as effort is dropped and variance of production increases. To counter these outcomes, he suggests an alternate pay scheme which sets a reasonable benchmark and those that fall closest to it are rewarded in tournament fashion. Furthermore, Hvide (2002) contends that this benchmark explains the discrepancy between the theory that those who excel ought to get rewarded the most and the empirical reality that mediocrity is often disproportionately rewarded beyond excellence. Grund and Gurtler (2005) examine the effect risk taking has on outcomes in professional soccer in Germany. They find that when risk is taken by a team, the team earns fewer points and the variance of points earned is higher. In the case of soccer, when a team assumes more risk by adopting an attacking formation, it loses more frequently than otherwise expected and more total goals are scored. Poker is used to test the theoretical and empirical findings of tournament theory. In order to use the game of poker as a basis for empirical analysis, it must be shown to be a game of skill and not chance. Dreef et. al (2003) study the relative skill level in poker and find that a simplified form of poker does involve more skill relative to other casino games and enough to be considered a game of skill. Dreef et. al used a simple limit two-player game as opposed to the more common many player no limit game. It is reasonable to extend their findings and suggest that due to the greater number of competitors and increased cost of mistakes, many player no limit tournaments are likely to contain greater skill than Dreef et. al find in the simplified version. Davidson (2007) examines the decision-making of top professionals in prestigious tournaments on the World Poker Tour. Using World Poker Tour data allows Davidson to see the cards each player holds, hole cards. He finds that despite the aforementioned designation of poker as a game of skill, top professionals base their decision making on noise and that Monte Carlo simulations outperform professionals in maximizing chip stack. A distinction is made between the static question of maximizing chip stack and the dynamic question of maximizing payout. Davidson finds elite players to be more risk averse than the Monte Carlo simulations predict. The observed behavior of elite players may be due to the fact that tournament poker is not a game of static optimization but dynamic optimization. Each player must maintain his stack throughout the tournament to avoid elimination. Davidson concludes that top professionals include “noise” in their decision-making process and would likely be better served by using the Monte Carlo odds. In an attempt to isolate the strategic employment of risk taking, Lee (2004) examines high stakes poker tournaments on the World Poker Tour. He contends that after risk selection, effort is a trivial issue. The amount of risk a player assumes is hypothesized to be impacted by the spread between payouts, relative positioning, and stability of relative positioning. The change in chip stack over time is used to proxy risk taking. Lee finds that the predictions of tournament theory hold in tournament poker |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://scholarship.richmond.edu/cgi/viewcontent.cgi?article=1160&context=honors-theses |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
