Forecasting risk attitudes: An experimental study using actual and forecast gamble choices

We develop and evaluate a simple gamble-choice task to measure attitudes toward risk, and apply this measure to examine differences in risk attitudes of male and female university students. In addition, we examine stereotyping by asking whether a person's sex is read as a signal of risk preference. Subjects choose which of five 50/50 gambles they wish to play. The gambles include one sure thing; the remaining four increase (linearly) in expected payoff and risk. Each subject also is asked to guess which of the five gambles each of the other subjects chose, and is paid for correct guesses. The experiment is conducted under three different frames: an abstract frame where the two highest-payoff gambles carry the possibility of losses, an abstract frame with no losses, and an investment frame that mirrors the payoff structure of the former. We find that women are significantly more risk averse than men in all three settings, and predictions of both women and men tend to confirm this difference. While average guesses reflect the average difference in choices, only 27 percent of guesses are accurate, which is slightly higher than chance.     

An experimental study of precautionary bidding

Auctions often involve goods exhibiting a common-knowledge ex-post risk. In such auctions, precautionary bidding predicts that under expected utility, DARA bidders reduce their bids by more than the appropriate risk premium. Because the degree of riskiness of an auctioned good and bidders׳ levels of risk aversion are difficult to observe in field settings, we conduct experimental auctions that allow us to identify the precautionary premium directly. We find strong evidence for precautionary bidding. The effect is robust to changes in experimental design features. Our experiment provides the first empirical demonstration of precautionary motives in a strategic setting.     

How luck and performance affect stealing

This paper investigates how the way of earning payoff affects the probability of stealing. The participants who earned their payoff according to performance were three times more likely to take the (undeserved) maximum payoff than participants with randomly allocated payoff. Conditional on stealing something, most subjects steal the full amount available.     

The effect of endogenous timing on coordination under asymmetric information: An experimental study

This paper investigates the role of endogenous timing of decisions on coordination under asymmetric information. In the equilibrium of a global coordination game, where players choose the timing of their decision, a player who has sufficiently high beliefs about the state of the economy undertakes an investment without delay. This decision (potentially) triggers an investment by the other player whose beliefs would have led to inaction otherwise. Endogenous timing has two distinct effects on coordination: a learning effect (early decisions reveal information) and a complementarity effect (early decisions eliminate strategic uncertainty for late movers). The experiments that we conduct to test these theoretical results show that the learning effect of timing has more impact on the subjects' behavior than the complementarity effect. We also observe that subjects' welfare improves significantly under endogenous timing.     

Corruption in public service delivery: An experimental analysis

To improve our understanding of corruption in service delivery, we use a newly designed game that allows us to investigate the effects of the institutional environment on the behavior of service providers and their monitors. We focus on the effect of four different factors: whether monitors are accountable to the service recipients, the degree of observability of service providers’ effort, the providers’ wages and the providers’ professional norms. In accordance with theory, we find that service providers perform better when monitors are elected by service recipients and when their effort is more easily observed. However, there is only weak evidence that service providers perform better when paid more. Monitors are more vigilant when elected and when service providers are paid more. Playing the game with Ethiopian nursing students, we also find that those with greater exposure to the Ethiopian public health sector perform less well, either as provider or as monitor, when the experiment is framed as a public health provision scenario, suggesting that experience and norms affect behavior.     

Premium auctions and risk preferences: An experimental study

In premium auctions, the highest losing bidder receives a reward from the seller. This paper studies the private value English premium auction (EPA) for different risk attitudes of bidders. We explicitly derive the symmetric equilibrium for bidders with CARA utilities and conduct an experimental study to test the theoretical predictions. In our experiment, subjects are sorted into risk-averse and risk-loving groups. We find that revenues in the EPA are significantly higher when bidders are risk loving rather than risk averse. These results are partly consistent with theory and confirm the general view that bidders' risk preferences constitute an important factor that affects bidding behavior and consequently also the seller's expected revenue. However, individual subjects rarely follow the equilibrium strategy and revenue in our experiment is lower than in the symmetric equilibrium.     

Valuing low probability risk: survey and experimental evidence

This study uses a survey question and an economics experiment to elicit the value of a risk mitigation investment. For 93 subjects (48 males and 45 females) the responses to hypothetical questions are compared with a risk mitigation decision that has a salient outcome. We find no gender-differentiated responses to either the maximum accepted price in the economics experiment or hypothetical willingness to pay questions from the survey. In our design, the two procedures generate aggregate measures that were in close agreement, however individual decisions were seldom consistent.     

An experimental study on individual choice, social welfare, and social preferences

We experimentally study subjects’ compliance with dominance relationships of income distributions in a ranking task. The experiment consisted of four different treatments: Lottery, individual choice, social preferences, and social planner. Our results suggest that people's risk attitudes do not adequately reflect their inequality attitudes. Uninvolved social planners exhibit randomization preferences, while self-interested social planners are generally more inequality averse and try to avoid extreme outcomes.     

The power of ESS: An experimental study

Abstract. Our experimental design mimics a traditional evolutionary game framework where players are matched pairwise to play a symmetric 33 bimatrix game that has two Nash equilibria. One equilibrium is an evolutionary stable state, or ESS; the other is an equilibrium in dominated strategies. Our primary experimental result is the observation that the ESS becomes extremely attractive when subjects have minimal information about the payoff functions, although the dominated equilibrium assures the highest equilibrium payoff. The attractiveness of the ESS is only moderate when players are completely informed about the 33 payoff matrix. Correspondence to: S.K. Berninghaus     

技术社会实验的理论起源与实践应用

现代技术的复杂性和不确定性、知识的不完备性等,使得实验室研究不能够完全验证技术知识的正确性,导致技术的社会引入呈现出明显的实验特征。技术社会实验是新近产生的、用以处理新技术不确定性、潜在危害和收益问题的理论。通过述评技术社会实验理论提出背景、追踪理论起源、阐述技术社会实验的基本特征、分析代表性研究案例,研究揭示技术社会实验的理论价值和现实意义,及对我国新兴科技创新治理的重要启示。     

An experimental study of decisions in dynamic optimization problems

Summary. We study decisions of subjects who are given an incentive to solve dynamic optimization problems with the structure of a single-agent, one-sector, closed economy macroeconomic model. The decision task involves a sequence of choices of consumption and investment levels. Treatment variables consist of the initial endowment of capital stock, the production technology available to the economy, and the method of creating the structure of an infinite-horizon model. The study includes and contrasts data from both American and Japanese participants. We find that whether over- or underinvestment relative to the optimum occurs depends on the production technology, but not on the initial endowment of capital stock, nor the subject pool used, nor the method of implementing the infinite horizon. Sudden episodes of maximal consumption called binges, which are always suboptimal, are widely observed. Received: December 7, 1998; revised version: March 12, 1999     

Incentive effects and overcrowding in tournaments: An experimental analysis

This study reports experiments that examine outcomes when agents choose between a payment scheme that rewards based on absolute performance (i.e., piece rate) and a scheme that rewards based on relative performance (i.e., a tournament). Holding total payments in the tournament constant, performance is higher when the tournament option is winner-take-all compared to a graduated tournament (i.e., second and third-place performers also receive a payment). Performance is higher in the winner-take all tournaments even among participants that choose the piece-rate option. While there is a modest amount of overcrowding, there are no significant differences in overcrowding across conditions. Entry rates into the tournament and the relative ability of tournament entrants (compared to non-entrants in the same condition) are higher in the graduated tournament condition than the winner-take-all conditions. Consequently, the winner-take-all tournament is more efficient than the graduated tournament (incentive effects are stronger and the overcrowding is about the same), but the graduated tournament provides a more effective mechanism to identify the most capable performer in a talent pool. Electronic Supplementary Material Supplementary material is available in the online version of this article at . JEL Classification D8 · J3 · L2     

Indicative bidding: An experimental analysis

Indicative bidding is a practice commonly used in sales of complex and very expensive assets. Theoretical analysis shows that efficient entry is not guaranteed under indicative bidding, since there is no equilibrium in which more qualified bidders are more likely to be selected for the final sale. Furthermore, there exist alternative bid procedures that, in theory at least, guarantee 100% efficiency and higher revenue for the seller. We employ experiments to compare actual performance between indicative bidding and one of these alternative procedures. The data shows that indicative bidding performs as well as the alternative procedure in terms of entry efficiency, while having other characteristics that favor it over the alternative procedure. Our results provide an explanation for the widespread use of indicative bidding despite the potential problem identified in the equilibrium analysis.     

Cheap talk with multiple audiences: An experimental analysis

We examine strategic information transmission in a controlled laboratory experiment of a cheap talk game with one sender and multiple receivers. We study the change in equilibrium behavior from the addition of another audience as well as from varying the degree of conflict between the senderʼs and receiversʼ preferences. We find that, as in cheap talk games with just one receiver, information transmission is higher in games with a separating equilibrium, than in games with only a babbling equilibrium. More interestingly, we find clear evidence that the addition of another audience alters the communication between the sender and the receiver in a way consistent with the theoretical predictions. There is evidence of the presence of agents that are systematically truthful as senders and trusting as receivers: deviations from the theoretical predictions, however, tend to disappear with experience, and learning is faster precisely in the games where deviations are more pronounced.     

Men, women, and competition: An experimental test of behavior

This study reports a series of experiments that examine outcomes when agents are able to choose between a payment scheme that rewards based on absolute performance (i.e., piece rate) and a scheme that rewards based on relative performance (i.e., a tournament). We test for the presence and persistence of gender differences in performance and the rate of entry into the tournament option and whether these differences are sensitive to the structure of the tournament rewards. In the winner-take-all (WTA) condition, only the best performer in the tournament for each round received a payment ($4.50). In the graduated tournament condition, the same payment ($4.50) was divided among the first, second, and third finishers in the tournament. In the WTA condition, men showed significantly lower forecast errors than women. In addition, a clear sorting effect occurs in the WTA condition. In early rounds of the WTA condition, male entrants into the tournament show significantly lower forecast errors than female tournament entrants. However, the difference disappears over time. After controlling for forecasting skill, gender did not predict entry into the tournament for the WTA condition. However, lower forecasting skill reduced the probability of entry. In the graduated tournament, the situation was reversed. Men entered the tournament at significantly higher rates, even after controlling for skill. Forecasting skill had no impact on the decision to enter the tournament. While the average male entrant to the tournament had lower forecast errors than the average female entrant, the men entered at much higher rates. As a consequence, men were much more likely than women to enter the tournament too frequently.     

Congestion allocation for distributed networks: an experimental study

This paper reports an experimental study of two prominent congestion and cost allocation mechanisms for distributed networks. We study the fair queuing (or serial) and the FIFO (or average cost pricing) mechanisms under two different treatments: a complete information treatment and a limited information treatment designed to simulate distributed networks. Experimental results show that the fair queuing mechanism performs significantly better than FIFO in all treatments in terms of efficiency, predictability measured as frequency of equilibrium play, and the speed of convergence to equilibrium. We thank Simon Anderson, Narine Badasyan, Catherine Eckel, Peter Katuscak, Edna Lohman, seminar participants at Caltech, Michigan, Vanderbilt, Virginia, Virginia Tech, ESA 2001 (Tucson, AZ) and Public Choice Society meetings (2002) for helpful comments. We are grateful to David Cooper and an anonymous referee for their constructive and insightful suggestions which significantly improved the paper. We thank Jing Deng, Belal Sabki, Pragya Sen, Brian Chan and Lisa Robinson for excellent research assistance, and Jim Leady for programming for the experiment. The research support provided by NSF grant SES-0079001 to Chen, SES-9973731 to Razzolini, and a grant from the Office of Naval Research to Razzolini are gratefully acknowledged.     

An experimental evaluation of tax-reporting schedules: a case of evidence-based tax administration

In an experiment in collaboration with the Australian Taxation Office (ATO), rental property schedules (RPS) were sent to taxpayers for them to itemize their deductions and return the RPS to the ATO. The intervention significantly reduced deductions compared to taxpayers who did not have to return the RPS, taxpayers who received only an information letter, and a no-contact control group—regardless of a cooperative versus deterring letter tone and first-time versus second-time participation of taxpayers. The study sheds light on the mechanisms underlying the effects of tax-reporting schedules and highlights the importance of systematic experimental fieldwork in evaluating regulatory strategies.     

Inefficiency and social exclusion in a coalition formation game: experimental evidence

This paper experimentally investigates the impact of reciprocal behavior in multilateral bargaining and coalition formation. Our results show that reciprocal fairness strongly affects the efficiency and equity of coalition formation. In a large majority of cases, inefficient and unfair coalitions are chosen when their coalition values are relatively high. Up to one third of the experimental population is excluded from bargaining and earns nothing. In monetary terms economically significant efficiency losses occur. We find that the interplay of selfish and reciprocal behavior unavoidably leads to this undesirable consequences. We also compare the predictions of recently developed models of social preferences with our experimental results. We find that some of these models capture the empirical regularities surprisingly well.