The optimal distribution of income and wealth has always been heatedly discussed in public debate. An important aspect of this discussion is whether and to what extent the state should lower income equality through the redistribution of resources from the rich to the poor. The discourse about how welfare state institutions should be designed is also highly relevant from an economic point of view. Empirical studies have shown that income inequality can hamper economic efficiency and trigger crises. At the same time, the design of welfare state institutions itself has direct economic effects by influencing labor supply through incentive mechanisms. Given their prominent role in the organization of economies and societies, how should we optimally design welfare institutions? From a scientific perspective, the optimal design of the welfare state can only be assessed from efficiency considerations. Equally important, though, is the normative component and the question of what degree of inequality a society considers justifiable and fair. The question of fairness cannot be answered using conventional tools of economics, as such assessments always depend on the worldviews and ethical norms of the members of a society. Essentially, the normative evaluation of income inequality depends on individual preferences. How people evaluate the distribution of resources and opportunities depends on many factors, including education, income, age, and political beliefs. From numerous analyses of international microeconomic data, we know that individuals’ redistribution preferences vary widely both between and within countries. Approximately 70% of citizens living in Germany believe that the state should reduce income differences. Compared to Poland (49.0%) and the Czech Republic (45.5%), this is a significant proportion. In Portugal (92%), on the other hand, support for redistributive policies is much higher (see ESS, 2020). This observation leads to a natural set of questions: What determines these substantial differences in redistribution preferences between nations? And how do these preferences influence the economy? This dissertation focuses on the exploration of the causes and effects of redistribution preferences. The work contributes to a better understanding of the factors that influence preference formation and, consequently, the distribution of income and wealth following government intervention. The implications arising from the research findings that I document in this dissertation can provide valuable insights into the design of social welfare institutions. In the first study of this dissertation (“Global Evidence on Misperceptions and Preferences for Redistribution”), my co-author and I show that many people have erroneous perceptions about their socioeconomic status compared to other members of society. Developing a new machine learning algorithm based on existing techniques of pattern recognition, we calculate individual-level misperceptions on a global scale. Our methodology is based on cross-country survey data for over a quarter of a million people representing around 90% of the world’s population. The results show that the perception biases of socioeconomic status are widespread globally. However, our numbers also reveal substantial variation in misperceptions between countries and socioeconomic characteristics. In particular, individuals at the lower end of the distribution tend to underestimate their status, while those with higher status often underestimate their relative social position. This pattern can be explained by individuals’ social reference groups: those with higher status frequently interact with others from similar backgrounds and are thus unaware of their positions in the overall distribution. In the empirical part of our study, we demonstrate that an upward-biased perception is associated with a lower demand for redistribution. The consequences of these empirical patterns are significant. Aggregating perception biases on a country-year level, we also show that misperceptions have direct effects on government redistribution via the progressivity of the tax and transfer system. The global nature of our study indicates that our findings point to a general pattern in human behavior, which significantly influences policy decisions and redistribution measures. The second study (“Employment and Redistribution Preferences: Evidence from the Women’s Liberation Movement”) examines the influence of labor market participation on the formation of preferences and beliefs. Estimating a causal effect of employment status on preference formation is challenging from a statistical perspective, as microeconomic data and surveys provide little information about whether labor market participation influences preference formation - or if the causal effect runs in the opposite direction. To address this issue, the study examines the relationship between employment and redistribution preferences in the historical context of the Women’s Liberation Movement, which began in the early 1960s. This context offers two significant statistical advantages. First, the Women’s Liberation Movement led to a substantial increase in female labor force participation, with substantial differences between countries. This increase and the country-specific variations provide ample variation that can be used for statistical causal analysis. Second, the Women’s Liberation Movement and the associated increase in female labor force participation were triggered by an exogenous shock: the invention of the contraceptive pill, commonly known as the “pill”. The invention and legalization of the pill were revolutionary for many women, giving them for the first time in history free choices over fertility decisions. The possibility of deciding whether and when to have children brought various improvements in educational opportunities and the ability to participate in the labor market. My empirical analysis is based on a newly compiled dataset on the availability of the pill including roughly 90 countries, linked to socioeconomic characteristics and preferences of over 150,000 women worldwide. Using this new dataset, the study first shows that the invention and legalization of the pill significantly increased female labor force participation, while it led to a stark decrease in fertility rates. Furthermore, I show that the timing of legalization is not driven by other factors. In the second step of the analysis, I use the pill as an instrumental variable to estimate the causal effect of female labor force participation on redistribution preferences. The results indicate that participation in the labor market leads to a significant reduction in preferences for redistribution. The global perspective provides results with high external validity, although it has the drawback that unobservable individual factors may influence pill usage after legalization. To address this challenge, a supplementary case study utilizes temporal differences in the legalization of the pill for legal minors in the United States. The advantage of this analysis is that individuals should differ much less across US states than across the countries included in the global dataset. Moreover, the availability of longitudinal data for the United States allows me to include fixed effects for individuals, which eliminates all time-invariant unobserved characteristics of women. The estimates for the United States provide results that are highly comparable to the results of the global dataset. In the final study of the dissertation (“Intergenerational Transmission of Welfare Benefit Receipt: Evidence from Germany”), my co-author and I examine the effects of state welfare institutions on the intergenerational transmission of welfare benefit receipt in Germany. In democracies, collective preferences in the long run determine the design of institutions. Redistribution preferences of the majority thus influence the generosity of welfare systems. In this study, we ask whether the design of such institutions can also exert adverse effects on individuals. This would be the case, for example, if welfare states led to a high transmission of welfare benefit receipt across generations, meaning that institutions cause parents to pass on their welfare benefit status to their children. The estimates in the third chapter are based on the long-term Socio-Economic Panel (SOEP) survey for Germany, which allows us to explore intergenerational transmission over three and a half decades. We start by describing the correlation between welfare benefit receipt in the parental household and later welfare benefit receipt of young adults. Then, we investigate whether these observed correlations reflect causal effects of past welfare experiences. To address this issue, we use estimates with fixed family effects and the approach of Gottschalk (1996), which considers unobserved family effects by accounting for future parental welfare benefit receipt. Our estimates uncover strong positive correlations between parental and individual welfare benefit receipt, but these patterns cannot be attributed to a causal effect of the social welfare system. Therefore, our results suggest that the observable strong intergenerational correlation of welfare benefit receipt is determined more by family background than by the experience of parental welfare benefit receipt. My dissertation contributes to previous research on the causes and consequences of redistribution preferences. My results consistently show that the individuals’ socioeconomic background, be it objective reality or subjective perception, has a strong influence on preference formation. Moreover, aggregated redistribution preferences correlate strongly with the generosity of welfare states. However, welfare state institutions themselves have a surprisingly limited effect on individuals’ receipt of welfare benefits, at least in the context of the transmission of welfare benefit receipt across generations. My research findings have various economic policy implications, which I discuss in the concluding section of this dissertation.