This thesis explores the use of generous behaviour as a costly signal to convey information about an unobservable social characteristic to other individuals in one's social environment. Building on recent contributions in this spirit, I develop a theoretical framework that contrasts signaling activities without social benefits with activities that benefit the observers in situations in which individuals compete for access to a scarce social good. The objective of the first part of the thesis is to characterize the possible separating equilibria in each case. While one obtains a multiplicity of equilibria when the agents employ neutral signals to convey information, one can make a unique prediction with respect to the individuals' equilibrium behaviour if they use beneficial signaling activities, the agents are of two discrete types, behave symmetrically within their respective types, and the observers of the signals adopt non-decreasing beliefs vis-à-vis the signalers' relative quality. In view of their sharp divergence, the second part of the thesis investigates experimentally the precision of these predictions. The results provide support for many elements of the theory. Among others, the behaviour of individuals in the treatments with beneficent signals is much more closely in line with the theoretical predictions than expected given the complexity of their behavioural implications, especially when it comes to the similarity of the participants' behaviour within them. Behaviour in the treatment with neutral signals, in turn, is consistent with multiple equilibria. The final part of the thesis explores what kind of signaling activity individuals trying to communicate their intentions to potential interaction partners will use in various social settings if given a choice. To this end, the framework developed in the first part is extended to allow the signalers to choose endogenously a signal from a “menu” of signaling activities rather than exogenously prescribing a messaging tool. Besides revealing that the uniqueness result of the framework without choice no longer obtains, the results indicate that the players may, under some conditions, opt for inefficient signals.

This dissertation is comprised of three essays that investigate consumer behavior and health outcomes. The first essay uses experimental economic techniques to explore consumers' preferences and willingness to pay (WTP) for nutritionally differentiated grass-fed beef. Our findings suggest that consumers' nutrition knowledge about the functions of Vitamin A, Vitamin E, CLA, and Omega 3 could positively affect their WTP for grass-fed beef while the knowledge about the main food sources of these nutrients negatively affects their WTP for grass-fed beef. Furthermore, a higher sensory evaluation score of grass-fed beef compared to conventional beef will lead to a higher probability for a consumer to choose grass-fed beef and a higher monetary value she/he is willing to pay for grass-fed beef. Using the same experimental data collected in the first study, the second essay investigates the impacts of consumers' nutrition knowledge on their WTP by accommodating the potential endogeneity problem using an instrumental variable approach and a non-instrumental variable approach. Our results suggest the existence of the endogeneity of nutrition knowledge and indicate that ignoring the endogeneity problem in econometric modeling will downwardly bias the estimates of the true effects of nutrition knowledge. The estimates obtained from different estimation strategies in the study indicate the robustness of our findings about the effects of nutrition knowledge on consumers' food purchasing behavior. The third essay investigates the impacts of primary care giver (PCG)'s time allocation patterns and household food expenditure choices on childhood obesity using the national panel study of income dynamics data. Our results do not suggest significant impacts of PCG's labor force participation, involvement in children's outdoor activity, and household food expenditures on children's Body Mass Index (BMI). However, the estimates from iterated seemingly unrelated regression (SUR) and semi-parametric polynomial estimation indicate that parents' BMI significantly influence children's BMI. Interestingly, physical activity appears to have weak correlation with children's BMI. / Ph. D.

The diploma thesis deals with analysis of risks in relation to various focus (or types) of economic experiments. On the basis of background research suggests entirely new classification of economic experiments and subsequently examines general structure of economic experiment. Performed analysis then identifies major risks of experimental economics, analyses them and proposes possible countermeasures. Outputs of the work substantially enrich and extend current theory. Their utilization is expected in theory and practice, for purpose of scientific research or specific objectives of companies and institutions.

This work statistically analyses hypotheses about the presence of antisocial punishment in study of economic experiments. Analysed data source is author's own ABM simulation of antisocial punishment environment. Tested hypotheses and ABM simulations are evolutionary oriented. This represents natural selection of evolving agents (and their groups) interacting with one another. Work is divided by two independent parts of presumptions. Crucial hypothesis of the first part is evolutional advantage from retaliatory punishment, the second part tests hypotheses of evolutional benefits resulting from group selection. Findings of the analysis allow us to accept these crucial hypothesis, and it may be concluded, that antisocial punishment can be regard as an evolutionary advantage, which directly advantages individuals, as well as whole groups.

It is well-known that taxes affect risky investment decisions. Analytical studies indicate that tax rate increases (decreases) can foster (hinder) investment if there is flexibility, in particular when an exit option is available. We design an experiment based on an analytical model with binomial random walk and entry and exit flexibility. Contrasting the underlying model, we find accelerated investment, which is often considered as an increased willingness to invest, on tax rate increases to be independent of the existence of an exit option. However, we observe this investor reaction only for a tax increase, not for a tax decrease. This behavior is driven possibly by tax salience and the mechanisms known from the theory of irreversible choice under uncertainty. Our empirical evidence suggests that the at-first-sight unexpected tax reform effects are more common than is predicted by the theoretical literature. Policy makers should therefore carefully consider the behavioral aspects when anticipating taxpayer reactions. (authors' abstract) / Series: WU International Taxation Research Paper Series