Terrorism imposes substantial economic costs on affected populations. In this paper, we argue that conventional studies on the costs of terrorism overlook a crucial factor: costly behavioral responses. We demonstrate that terrorism alters the risk attitudes of individuals in affected regions. Using a representative German sample, we utilize a staggered difference-in-differences (DiD) design comparing individuals in counties impacted by terrorism with those unaffected. Results indicate an immediate and significant decrease in risk propensity post-attack. The magnitude of this effect is influenced by news sentiment; more negative reporting amplifies the impact on risk attitudes. Additionally, we observe happiness as a potential mediator in the relationship between exposure to terror attacks and risk attitudes.

We aim to understand whether language causally affects people's attitudes towards private and public debt. We provide descriptive evidence along the German/French language border running through Switzerland and causal evidence from online survey experiments in Austria, Germany, Switzerland, the Netherlands, and Sweden. The experimental component of our research design consists of the randomization of wording in our outcome questions. Exploiting the fact that the commonly used word for ``debt" also means ``guilt" in German, Dutch and Swedish, we analyze how respondents' attitudes towards private and public debt are affected if those negatively connotated words are substituted with neutral ones. Our results confirm our main hypothesis that the use of guilt-connotated wording triggers lower approval for public debt and lower borrowing in an incentivized investment decision. Finally, we provide evidence that language is strategically used in the political arena using natural language processing and machine learning techniques on parliamentary speeches in the German Bundestag.

While the impact of adverse health shocks and the factors influencing charitable giving have been widely studied, research that combines these topics, especially focusing on a specific disease, is limited. Using a staggered treatment setup and Swedish register data, this study examines "ingroup altruism born of suffering": whether cancer patients are more inclined to donate to cancer charities and if they donate more. The results show a significant increase in the probability of donating to cancer charities after a diagnosis, a decrease in donations to non-cancer health charities, and no impact on non-health charities. Overall charitable giving increases significantly, mainly driven by donations to cancer charities. This pattern is consistent for both the probability and amount of giving. Additionally, donations to cancer charities continue to increase over time post-diagnosis, although the rate of increase slows down.