Topics in behavioural economics

© 2017 Dr. Edwin Chi Hin IpThis thesis contains three pieces of original research organised into three separate chapters. The three chapters make contributions to three different topics in Behavioural Economics and can be read independently of each other. The first chapter studies gender quotas. Gender quotas have become prevalent in many labour markets, but the environments in which they are implemented and their impact on hierarchical relationships remain largely unexplored. Using a novel set of laboratory experiments, I study how gender quotas for managers influence wage setting and their workers' subsequent provision of effort. In environments where women face no disadvantage in the selection process, I show that implementing quotas leads to lower wage and effort levels. However, in an environment where women are disadvantaged in a biased selection process, implementing quotas increases wage and effort levels. These results suggest that instead of blindly implementing affirmative action policies across the board, one should first carefully consider the degree of disadvantage that exists in the specific labour market. The second chapter proposes a model of decision making. I introduce a multi-attribute model of decision making whereby a decision maker derives utility from comparing each alternative, attribute by attribute, with all other alternatives that she would have to forgo by choosing one of them. Such comparisons create `pros' or `cons' along each attribute dimension which result in anticipated rejoice or regret for the decision maker. The model can accommodate and characterise well known behavioural irregularities such as choice avoidance, choice overload and decoy effects based on one parameter: the relative weight a decision maker assigns to rejoice and regret. Not only does the model make novel behavioural predictions, its tractibility makes it applicable to a wide range of economic problems to generate new insights into market outcomes. The third chapter investigates the differences between self-reported financial risk attitude and actual financial risk taking behaviour using a large and nationally representative longitudinal survey. I find that while the self-reported financial risk attitude is a very good predictor of financial risk taking behaviour in terms of risky asset ownership, its determinants and stability characteristics are very different from those of risk taking behaviour. Financial risk taking behaviour is consistent with the life cycle model of investing and varies strongly over time. It is also heavily associated with stock market movements. Self-reported risk attitude, by contrast, is stable, even throughout the Global Financial Crisis, and are correlated with personality traits. Our findings suggest that self-reported risk attitude, a measurement widely used in surveys and experiments, provides information about intrinsic preferences rather than risk taking behaviour