Essays on Behavioral Finance in the Digital Age

Abstract

This dissertation includes four essays on behavioral finance in the digital age. Specifically, the research papers cover the following content:In the first paper, entitled "Self-attribution bias and overconfidence among nonprofessional traders", my co-author and I investigate consequences of the self-attribution bias for nonprofessional traders. By applying a textual analysis of more than 44,000 public comments on a large social trading platform, we contribute to empirical literature on investment and trading behavior in three ways: First, we show that one component of the self-attribution bias, the self-enhancement bias, leads to subsequent underperformance. Second, results support the theory that traders become overconfident due to biased self-enhancement. Third, we find that traders´ social trading portfolios attract higher investment flows from investors when showing self-enhancement biased behavior.In the second paper, entitled "Signaling in initial coin offerings - the key role of entrepreneurs´ self-efficacy and media presence , by analyzing data of more than 1,000 Initial Coin Offerings (ICOs) obtained from seven different ICO information platforms, my co-author and I investigate the effectiveness of projects´ quality signals (human capital, entrepreneurs´ self-efficacy, ambiguity reduction and level of media presence) with regard to ICO funding success. Results imply that media presence and entrepreneurs´ self-efficacy are effective signals of project quality in the ICO market and thus, can foster funding success. Project initiators that communicate (more actively) via social media collect more funds than those who do not.Analogously, entrepreneurs appearing self-efficacious with regard to the quality of their venture receive more funds.In the third paper, entitled "Among peers: the impact of homophily in online investment", my co-authors and I investigate if homophily peoples affinity for similar others is an issue in online investment, too. Drawing on nearly 14,000 loan applications and 65,000 investments obtained from one of the leading online peer-to-peer lending platforms in Europe, we document strong evidence in support of a homophily effect on investors´ financial decision making. Controlling for a host of alternative determinants, being in the same age group as a given investor increases a loan applicant´s odds of being funded by as much as 14%, while same-sex dyads are associated with 6% higher odds of investment. Moreover, any additional demographic similarity increases the average investment amount by nearly 10%. At this, the impact of homophily on investors´ funding propensity proves substantially larger for female investors. Finally, we document a 19bp. difference in risk-adjusted interest rates of loans associated with investor-borrower dyads exhibiting the highest versus lowest number of homophilous ties. This evidence is hard to square with the notion that investors´ affinity for similar borrowers in online peer-to-peer lending follows economic rationale.In the fourth paper, entitled "Occupational self-selection among bankers and financial regulators: evidence from content analysis", using a unique dataset of 532 face-to-face interviews from a German business newspaper, my co-authors and I examine the psychological characteristics of bankers and financial regulators. Applying computerized content analysis techniques, we find that linguistic styles differ significantly among bankers and regulators even when controlling for topics derived from structural topic modeling suggesting differences in well-known psychological characteristics.In particular, we find that bankers´ linguistic style marks them as more selfish and overconfident. Regulators, in contrast, show more linguistic markers of cognitive complexity. Results support existing theoretical considerations that suggest self-selection into different occupations by bankers and regulators due to different psychological characteristics.