This paper attempts to empirically assess how advice may reduce suboptimality in a portfolio choice experiment with risk-neutral participants induced via binary lottery incentives. Previous studies (---2017 and 2018) with a larger set of choice tasks report overwhelming evidence of suboptimality and how it is slightly reduced by learning and experience. Participants confront 15 randomly ordered portfolio choices, which they experience again in two successive phases. Intermediate advice between phases alerts participants that less-risky investments can improve the outcome for at least one chance event without harming their success chances in the other random event. Compared to the pure choice treatment, another cognitively more demanding treatment additionally asks participants to form event-specific success aspirations that allow us to test satisficing and its optimality. The results show that intermediate advice increases the share of satisficing but not of optimal behavior beyond learning through experience. However, it significantly lowers the average distance from optimality.
Experimental Evidence of Behavioral Improvement by Learning and Intermediate Advice
Noemi Pace;
2020-01-01
Abstract
This paper attempts to empirically assess how advice may reduce suboptimality in a portfolio choice experiment with risk-neutral participants induced via binary lottery incentives. Previous studies (---2017 and 2018) with a larger set of choice tasks report overwhelming evidence of suboptimality and how it is slightly reduced by learning and experience. Participants confront 15 randomly ordered portfolio choices, which they experience again in two successive phases. Intermediate advice between phases alerts participants that less-risky investments can improve the outcome for at least one chance event without harming their success chances in the other random event. Compared to the pure choice treatment, another cognitively more demanding treatment additionally asks participants to form event-specific success aspirations that allow us to test satisficing and its optimality. The results show that intermediate advice increases the share of satisficing but not of optimal behavior beyond learning through experience. However, it significantly lowers the average distance from optimality.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
