Contracts and bargaining with inequity aversion
(with Marcus Dittrich - DIT)
This paper analyzes bargaining over a labor contract in a principal-agent model when the agent is inequity averse. Although we assume full information, we find that the take-it-or-leave-it contract does not entail first-best effort. Effort may be inefficiently high (low) because the principal has to compensate the agent for costs of inequity aversion that decrease (increase) with rising effort. Bargaining is either always detrimental to the efficiency of the contract or can completely mitigate inefficiencies arising from inequity aversion.
Comparing the Behavior of Teams and Individuals in a Public Goods Game with Ostracism - A Null Result?
(with Stephan Huber - HS Fresenius, and Jochen Model)
We provide evidence from a public goods game with ostracism, i.e. the possibility to vote and consequently ostracize others from the game. We focus on how the decisions of individuals and teams differ in this setup. Participants either form groups of individuals or they form groups of two-member-teams to play the public goods game. Concerning contributions, we find a null-result. Concerning earnings, however, we find differences. The ostracism mechanism does not increase average earnings for individuals, but for teams. This is the consequence of a different use of the ostracism mechanism. Teams play a trigger strategy. The mere threat of being punished triggers cooperative behavior, i.e. higher contributions. The punishment as such, however, is seldom really executed. Individuals exclude more and earlier, yielding less earnings since ostracized members' contributions are missing.
Social interaction as a driver of information avoidance
(with Michael Eichenseer, Johannes Moser, Felix Peterhammer, Andreas Roider and Lars-Christopher Schlereth - all UR)
We develop a model in which it is rational for an individual to refrain from getting costless and precise information. Other than the literature so far, we do not need any behavioral assumptions regarding the utility function of the individual, i.e. the individual is a standard utility maximizer. We show that there exist cases, where information avoidance is rational for the individual depending on her ex-ante belief about her survival probabilities. Our model is able to explain empirical findings of (Oster et al., 2013, AER), especially why testing rates for diseases are low and why testing rates increase with higher ex ante risk of having the disease.
"Field in the Lab" – An Animal Approach to Human Lab Experiments
(with Tomer Czaczkes - UR, Department of Zoology, Animal Comparative Eonomics Laboratory)
Experimental economics is criticized by the now widely-held realisation that results from educated, western subjects are not representative of humans as a whole. To address this critical issue, we propose taking a novel ‘duel comparative’ approach to studying economic behaviour. Firstly, experiments on human subjects will follow a self-teaching paradigm, where subjects must learn the rules of each experimental manipulation. This mimics the way in which the economic behaviour of non-communicative subjects (animals or preverbal humans) is studied. This defuses issues of ‘second-guessing’ and may reduce issues of act rationality in standard laboratory experiments. In addition, it increases the realism and real-world relevance of experiments without sacrificing the tight control of the laboratory and may this enhance the external validity of lab experimental results.
In parallel, identical experiments will be run on non-human subjects. Parallel findings in humans and non-human animals acts as very strong evidence of broadly relevant economic behaviour, rather than cultural idiosyncrasies.
Smarter People are More Often Overconfident - Evidence From a Large-Scale Online Experiment
(with Marcus Dittrich - DIT)
We provide evidence on the relationship between overconfidence and cognitive ability from a large-scale online experiment in Germany. We use data from the 11–20 money request game to measure overconfidence and data from the cognitive reflection test to measure cognitive ability. Since our measure of overconfidence does not rely on people’s self-perception, our study does not face the problem of regressive measurement of overconfidence. We find that people with higher cognitive ability are more likely to be overconfident. Moreover, we find that men are more likely to be overconfident than women and that people that were born earlier tend to be less overconfident.
Are women really more patient than men? - New evidence from a large scale online experiment
(with Marcus Dittrich - DIT)
We provide evidence on gender differences in time preferences and financial literacy from a large scale incentivized online experiment. In contrast to the literature so far, we find women to be more impatient than men at least when stakes are not too high. Moreover, we confirm that women are significantly less financial literate and that financial literacy is related to time preferences in a way that less patient people are less financial literate.
Challenging IIA - Theory and Experimental Investigation with a Non-Cooperative Approach
(with Eric Cardella - Texas Tech University)
tba.
Team Behavior in Games: A Behavioral Theory
tba.
Contracts and bargaining with inequity aversion
(with Marcus Dittrich - DIT)
This paper analyzes bargaining over a labor contract in a principal-agent model when the agent is inequity averse. Although we assume full information, we find that the take-it-or-leave-it contract does not entail first-best effort. Effort may be inefficiently high (low) because the principal has to compensate the agent for costs of inequity aversion that decrease (increase) with rising effort. Bargaining is either always detrimental to the efficiency of the contract or can completely mitigate inefficiencies arising from inequity aversion.
Comparing the Behavior of Teams and Individuals in a Public Goods Game with Ostracism - A Null Result?
(with Stephan Huber - HS Fresenius, and Jochen Model)
We provide evidence from a public goods game with ostracism, i.e. the possibility to vote and consequently ostracize others from the game. We focus on how the decisions of individuals and teams differ in this setup. Participants either form groups of individuals or they form groups of two-member-teams to play the public goods game. Concerning contributions, we find a null-result. Concerning earnings, however, we find differences. The ostracism mechanism does not increase average earnings for individuals, but for teams. This is the consequence of a different use of the ostracism mechanism. Teams play a trigger strategy. The mere threat of being punished triggers cooperative behavior, i.e. higher contributions. The punishment as such, however, is seldom really executed. Individuals exclude more and earlier, yielding less earnings since ostracized members' contributions are missing.
Social interaction as a driver of information avoidance
(with Michael Eichenseer, Johannes Moser, Felix Peterhammer, Andreas Roider and Lars-Christopher Schlereth - all UR)
We develop a model in which it is rational for an individual to refrain from getting costless and precise information. Other than the literature so far, we do not need any behavioral assumptions regarding the utility function of the individual, i.e. the individual is a standard utility maximizer. We show that there exist cases, where information avoidance is rational for the individual depending on her ex-ante belief about her survival probabilities. Our model is able to explain empirical findings of (Oster et al., 2013, AER), especially why testing rates for diseases are low and why testing rates increase with higher ex ante risk of having the disease.
"Field in the Lab" – An Animal Approach to Human Lab Experiments
(with Tomer Czaczkes - UR, Department of Zoology, Animal Comparative Eonomics Laboratory)
Experimental economics is criticized by the now widely-held realisation that results from educated, western subjects are not representative of humans as a whole. To address this critical issue, we propose taking a novel ‘duel comparative’ approach to studying economic behaviour. Firstly, experiments on human subjects will follow a self-teaching paradigm, where subjects must learn the rules of each experimental manipulation. This mimics the way in which the economic behaviour of non-communicative subjects (animals or preverbal humans) is studied. This defuses issues of ‘second-guessing’ and may reduce issues of act rationality in standard laboratory experiments. In addition, it increases the realism and real-world relevance of experiments without sacrificing the tight control of the laboratory and may this enhance the external validity of lab experimental results.
In parallel, identical experiments will be run on non-human subjects. Parallel findings in humans and non-human animals acts as very strong evidence of broadly relevant economic behaviour, rather than cultural idiosyncrasies.
Smarter People are More Often Overconfident - Evidence From a Large-Scale Online Experiment
(with Marcus Dittrich - DIT)
We provide evidence on the relationship between overconfidence and cognitive ability from a large-scale online experiment in Germany. We use data from the 11–20 money request game to measure overconfidence and data from the cognitive reflection test to measure cognitive ability. Since our measure of overconfidence does not rely on people’s self-perception, our study does not face the problem of regressive measurement of overconfidence. We find that people with higher cognitive ability are more likely to be overconfident. Moreover, we find that men are more likely to be overconfident than women and that people that were born earlier tend to be less overconfident.
Are women really more patient than men? - New evidence from a large scale online experiment
(with Marcus Dittrich - DIT)
We provide evidence on gender differences in time preferences and financial literacy from a large scale incentivized online experiment. In contrast to the literature so far, we find women to be more impatient than men at least when stakes are not too high. Moreover, we confirm that women are significantly less financial literate and that financial literacy is related to time preferences in a way that less patient people are less financial literate.
Challenging IIA - Theory and Experimental Investigation with a Non-Cooperative Approach
(with Eric Cardella - Texas Tech University)
tba.
Team Behavior in Games: A Behavioral Theory
tba.