Ain’t blaming you: Delegation of financial decisions to humans and algorithms
Year: 2025
Type: article
Abstract: This article explores outcome bias---the inclination to disproportionately consider outcomes when evaluating the efficacy of a decision-making process---within the context of choices delegated to either a human or an algorithm. In our experiment, a Principal entrusts a risky financial decision to an Agent, who can handle it independently or delegate it to an algorithm. The Principal subsequently rewards or penalizes the Agent. We allow the Agent to delegate the choice to an algorithm and manipulate the Principal's knowledge about the investment outcome when rewarding the Agent. Results confirm the presence of outcome bias and show that the bias remains consistent whether the choice is made personally or delegated to an algorithm. Notably, reliance on the algorithm and the level of risk in investments are unaffected by whether rewards or punishments are determined before or after the investment outcome.
Authors: Zilia Ismagilova, Matteo Ploner
Institution: University of Trento
Language: English
Subfield: General Decision Sciences
Field: Decision Sciences
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $1,540
OpenAlex ID: W4409039978