Essays in behavioural and experimental economics: risk and time preference of adolescents

Abstract

There has been increased interest in understanding the economic preferences of adolescents in the last decade. However, the literature is still in its infancy in that we have only an incomplete picture of the economic preferences of young people. This thesis aims to contribute to this literature by investigating the risk and time preferences of adolescents.

In the first chapter, we investigate if people generally show present bias (β) and what factors lead to the heterogeneity across studies using meta-analysis. We construct the largest and most comprehensive dataset of β estimates (62 papers, 81 estimates). The literature shows that people are on average present biased towards money (β=0.82), but there is substantial heterogeneity across studies. People show stronger present bias towards real effort and health outcomes compared to money. Finally, we find evidence of selective reporting and publication bias in the direction of overestimating present-bias, but that present bias still exists after correcting for the biases.

In the second chapter, with a longitudinal experiment, we investigate adolescents’ time preferences towards money and food. We find that adolescents show strong present bias for money and food. The individual measures of present bias are moderately correlated across reward types. Our data also allows us to relate present bias measures elicited in a laboratory task to field behaviours including drinking, smoking, BMI and grades. We find that measures of time preferences over money and food predict field behaviours equally well.

In the third chapter, with a laboratory experiment, we investigate changes in adolescents’ risk preference under peer observation. The experiment allows us to separately estimate participants’ loss aversion and risk tolerance in gains and losses in private and when observed by a peer sitting next to them. We conduct our experiment with 12-17 and 18-24 years old adolescents and find that while observed by peers, 18-24 years old adolescents choose risky prospects more often. This is driven by more risk tolerance both in gains and losses, rather than reduced loss aversion. For 12-17 years old adolescents, neither their risk tolerance nor their loss aversion is affected by peer observation.