LFIN Seminar

Béatrice Boulu-Reshef

(University of Orleans)

will give a presentation on :

Algorithmic vs. Human Portfolio Choice

Robo-advisors that provide investment advice online on the basis of risk profiling questionnaires have recently made a breakthrough in the investment management industry. The validity and reliability of these questionnaires is crucial as profiling inaccuracies can lead to a mismatch between investment proposals and retail investors’ preferences. This paper uses data from a robo-advisor that makes portfolio recommendations to its potential users and lets them choose their risk exposure after having received this recommendation. Comprehensive information about savers’ characteristics allow us to investigate how the robo-advisor’s algorithm maps questionnaire’s answers into recommendations and to what extent users follow or deviate from the recommendation. The results provide evidence that risk profiles recommended by the robo-advisor are qualitatively aligned with financial portfolio theory. Although a variety ofinformation is used by the algorithm, the recommendation is heavily based on answers about financial risk taking. A large majority of users follow the recommendation and as a result are also strongly influenced by their declared propensity to take financial risk. Factors influencing the recommendation like age, financial wealth and short-term liquidity needs are downplayed by savers. Factors not exploited by the algorithm like saving’s goals or professional occupation impact investors’ portfolio choice. Gender, although not taken into account by the algorithm, still influences savers’ choice after controlling for a wealth of potential confoundings.

• Béatrice Boulu-Reshef : http://www.beatrice-boulu-reshef.info