Publications

This paper develops a theory of focusing and framing in an intertemporal context with risky choices. We provide a selection criterion between existing theories of focusing by allowing a decision maker to choose her frame such that her attention is either drawn to salient events associated with an option or to the expected utilities an option yields in different time periods. Our key assumption is that a decision maker can choose her frame in a self-serving manner. We predict that the selected frame induces overoptimistic actions in the sense that subjects underrate downside risk but overrate upside risk and accordingly reveal overoptimistic choices. Hence, our theory can explain phenomena such as excessive harmful consumption (smoking, unhealthy diet) and risky investments (entrepreneurship, lotteries, gambling) in one coherent framework. Notably, overoptimistic actions are not universal, but have plausible limits. We characterize under which situations overoptimistic actions are most likely to occur and under which circumstances choices should be rational or even pessimistic.

We experimentally study the role of punishment for cooperation in dynamic public-good problems where past payoffs determine present contribution capabilities. The beneficial role of punishment possibilities for cooperation is fragile: successful cooperation hinges on the presence of a common understanding of how punishment should be used. If high-contributors punish too readily, the group likely gets on a wasteful path of punishment and retaliation. If punishment is administered more patiently, even initially uncooperative groups thrive. Hence, when today’s punishment also determines tomorrow’s cooperation abilities, it seems crucial that groups agree on the right ‘dose’ of sanctions for punishment to support cooperation.

dc.title:

When people sense that another person tries to control their decisions, some people will act against the control, whereas others will not. This individual tendency to control-averse behavior can have far-reaching consequences, such as engagement in illegal activities or noncompliance with medical treatments. Although individual differences in control-averse behavior have been well documented in behavioral studies, their neurological basis is less well understood. Here, we use a neural trait approach to examine whether individual differences in control-averse behavior might be linked to stable brain-based characteristics. To do so, we analyze the association between intrinsic connectivity networks as measured by resting state functional magnetic resonance imaging and control-averse behavior in an economic exchange game. In this game, subjects make choices that are either free or controlled by another person, with real consequences to both interaction partners. We find that the individual level of control-averse behavior can be positively predicted by intrinsic connectivity within the salience network, but not the central executive network or the default mode network. Specifically, subjects with a more prominent connectivity hub in the dorsal anterior cingulate cortex show greater levels of control-averse behavior. This finding provides the first evidence that the heterogeneity in control-averse behavior might originate in systematic differences of the stable functional brain organization.

Shleifer and Vishny (1997) argue that corporate governance should be weak in the absence of powerful residual claimants. We compare foundation owned firms (FoFs) and family firms, with and without codetermination. As foundations have no owners, residual claimants of FoFs are weak. This might strengthen FoF-managers and employees. Codetermination law also strengthens employees. We derive hypotheses about business policy of FoFs and test them. Our findings show that German FoFs are more labor intensive relative to family firms. But their wages and their hiring and firing policy are about the same. Their financing policy is more conservative, their financial performance is slightly weaker. Apart from financing policy, codetermination has similar effects. These findings indicate a stronger impact on corporate governance of employees in firms with weak residual claimants and in codetermined firms, combined with long-term orientation. But, in contrast to Shleifer and Vishny (1997), we do not find evidence of weak corporate governance.

A principal’s production decision imposes a negative externality on an agent. The principal may be a pollution-generating firm, the agent may be a nearby town. The principal offers a contract to the agent, who has the right to be free of pollution. Then the agent privately learns the disutility of pollution. Finally, a production level and a transfer payment are implemented. Suppose there is an upper bound (possibly zero) on payments that the agent can make to the principal. In the second-best solution, there is underproduction for low cost types, while there is overproduction for high cost types. In contrast to standard adverse selection models of pollution claim settlements, there may thus be too much pollution compared to the first-best solution.

To study whether current spending levels and public knowledge of them contribute to transatlantic differences in policy preferences, we implement parallel survey experiments in Germany and the United States. In both countries, support for increased education spending and teacher salaries falls when respondents receive information about existing levels. Treatment effects vary by prior knowledge in a manner consistent with information effects rather than priming. Support for salary increases is inversely related to salary levels across American states, suggesting that higher salaries could explain much of Germans' lower support for increases. Information about the tradeoffs between specific spending categories shifts preferences from class-size reduction towards alternative purposes. Additional German experiments indicate that information effects extend to specific reform proposals and to other areas of public spending.

We study the implications of limited commitment on education and tax policies set by benevolent governments. Consistent with real‐world practices, a government can decide to subsidize different levels of education at different rates. A lack of commitment, however, affects the optimal structure of education subsidies. The direction of the effect depends on how labor taxes are designed. With linear labor tax rates and a transfer for redistribution, subsidies become more progressive. By contrast, if the government is only constrained by informational asymmetries when designing taxes, subsidies become more regressive.

In science education, representations are necessary inter alia for the understanding of relationships between structures and systems. However, several studies have identified difficulties of students when working with representations. In the present study, we investigated students' responses (regarding their preference, test performance, mental load (ML), and stress) toward two kinds of representations: symbolic representations, which only use abstract symbols, versus combined symbolic–textual representations, which additionally comprise textual elements. Therefore, students were randomly assigned to one of two treatment groups: one group worked on test tasks accompanied by symbolic representations, and the others worked on the same tasks, but with symbolic–textual representations. Thereafter, the students' test performance and ML were assessed. The level of perceived stress and the salivary cortisol concentration were measured before and after the test and again a few minutes later. Additionally, heart rate variability parameters were assessed continuously. We found a strong preference of the test version with symbolic representations. Additionally, the students showed better test performance and lower ML when they worked with symbolic representations. However, the level of perceived stress was comparable between both groups and there was no strong physiological stress response: The cortisol concentration decreased in both groups and the heart rate was relatively similar. However, during the second half of the test, we observed a significantly higher ratio between low and high heartbeat frequencies in the group with symbolic–textual representations and we found an indirect influence of the kind of representation on test performance through its effect on ML. The poorer test performance and higher ML in connection with symbolic–textual representations confirm previous studies, which found that symbolic–textual representations pose major problems for students. Thus, teachers should enable students to understand symbolic–textual representations and consider carefully whether they can use symbolic representations instead, especially when they teach complex content.