Investor decision making

Abstract

This thesis is comprised of three separate studies which examine different facets of the investment decision making process. A full literature review is included as an additional chapter following each study. The first study takes a broad perspective and focuses on the decision by retail investors to participate in risky asset markets. In particular, it examines whether the positive relation between financial literacy and participation in risky asset markets persists when allowing for variation on a more innate level: an individual’s propensity for financial planning. Propensity for financial planning is shown to be strongly positively associated with stock market participation as well as membership in KiwiSaver, a voluntary workplace retirement savings scheme in New Zealand. The positive association with financial literacy also remains, however, even when controlling for a comprehensive range of demographic characteristics and other control variables. Additional tests suggest that these two variables operate through separate channels. The second study takes a narrower perspective in investigating how social interaction is related to risk taking by retail investors. In contrast to recent theories from the finance literature, the results suggest that investors who tend to purchase stocks with high volatility or skewness moderate their risk taking once they begin to interact with other investors. These results are consistent with evidence from the experimental economics literature which indicates that individuals lower their propensity to accept investments with lottery-like payoffs when they are making decisions in a group setting relative to an individual setting. Finally, the third study concentrates on the investment decisions of short sellers, a sophisticated subgroup of investors, during the days leading up to important announcements in the drug development process. The study finds that on average, short sellers trade successfully during this period. Short sales rise significantly more in the days leading up to very negative drug development announcements relative to very positive announcements. Additional analysis indicates that superior information processing abilities rather than leakage of insider information are the more likely explanation for these results.