Banking architecture and algorithmic intelligence in asset management: The precision–discretion paradox in UITF and mutual fund institutions

Abstract

This study examines how banking architecture and algorithmic intelligence influence asset management outcomes in banking-managed Unit Investment Trust Funds (UITFs) and mutual fund institutions, with particular focus on the Precision–Discretion Paradox. Specifically, it investigates the effects of algorithmic intelligence on professional decision discretion and asset management outcomes, as well as the moderating role of governance structures. A quantitative explanatory research design was employed using survey data collected from 214 professionals involved in asset management, including fund managers, analysts, and investment officers. Data were analyzed using correlation, regression, mediation, and moderation techniques to test the proposed relationships among variables. The results indicate that algorithmic intelligence has a significant positive effect on asset management outcomes (β = 0.62, p < 0.001), reflecting improvements in decision consistency and effectiveness. However, it also exhibits a significant negative effect on professional decision discretion (β = –0.41, p < 0.01), suggesting reduced managerial autonomy. Mediation analysis reveals that professional discretion partially mediates the relationship between algorithmic intelligence and asset management outcomes, while moderation results show that strong banking architecture weakens the negative impact of algorithmic intelligence on discretion (β = 0.28, p < 0.05). While algorithmic intelligence enhances performance, its effectiveness depends on governance structures that preserve professional judgment. These findings highlight the importance of balancing technological precision with institutional oversight in asset management.