Four Essays on Economic Preferences

Abstract

This thesis consists of four independent chapters. Each chapter contributes experimental evidence to our knowledge on economic preferences. Chapter one shows that a preference for truth-telling per se is even more prevalent than previous research suggests. Chapter 2 investigates the relationship between economic preferences and psychological personality measures and arrives at the conclusion that the degree of association between the two concepts is rather small and that they are complementary in explaining heterogeneity in life outcomes. Chapter 3 validates non-incentivized survey measures for key economic preferences, i.e. risk taking, time discounting and social preferences, by examining their predictive power for behavior in incentivized economic choice experiments. Chapter 4 shows that the variation in preferences across countries as documented in Falk, Becker, Dohmen, Enke, Huffman, and Sunde (2015) has deep cultural origins. <br /> Chapter 1 attends to what is often called a non-standard preference: a preference for truth-telling per se.2 We implement a truth-telling experiment, in which misreporting cannot be detected and participants have a strong monetary incentive to misreport, with a representative population sample which we call at home. We nd that aggregate reporting behavior closely resembles the distribution that would result if everyone reported truthfully. This contrasts previous evidence from laboratory experiments which also documented substantial levels of truthful reporting as well, but consistently found considerable degrees of cheating. Since our partici pants made their reports via the phone while participants in laboratory experiments typically entered their reports into the computer, we conduct an additional laboratory experiment to rule out the possibility that the difference between behavior in our study and previous research is mainly driven by the difference in communication modes. Similarly, we can rule out that it is the difference in the subject pools - a representative sample versus the typical student participants in laboratory experiments - that explains the much higher level of truth-telling in our study: the behavior of the students in our representative sample does not differ from the behavior of the rest of the sample. <br /> Chapter 2 examines the relationship between economic preferences and psychological personality measures. Using data from incentivized laboratory experiments and representative samples of the German population it shows that the association between the two concepts is rather low and that the two concepts are complementary in explaining heterogeneity in life outcomes. <br /> Chapter 3 validates survey measures for the six key economic preferences - risk taking, time discounting, trust, altruism, positive and negative reciprocity - by assessing their (joint) explanatory power in explaining behavior in incentivized choice experiments. This results in a preference module consisting of two items per preference - one typically being a hypothetical version of the incentivized experiment and the other one being a subjective self-assessment. Next, we adjust the module by reducing complexity and excluding culturally loaded wording to allow implementability across heterogeneous participants, e.g. in terms of cultural or educational background, and across survey modes. We test this "streamlined" module in the field in 22 countries of diverse cultural backgrounds. The resulting feedback calls for only minor adjustments and overall confirms a good implementability of our preference module. <br /> Chapter 45 explores whether differences in culture can explain part of the varia tion in economic preferences we see across countries around the globe as documented in Falk, Becker, Dohmen, Enke, Huffman, and Sunde (2015) by using a specific feature of languages as a proxy for culture. Speakers of languages which require the speaker to grammatically mark the future when talking about future events are less patient and less prosocial than speakers of languages which lack such a grammatical requirement. Heterogeneity in preferences across countries and cultures seems to be partly driven by deep cultural differences.