Algorithmic Management and Employee Autonomy: Investigating the Trade-Off Between Efficiency and Well-Being

Abstract

Algorithmic management, the use of software to assign, direct, monitor and evaluate work, raises operational efficiency but has been associated with reduced employee autonomy and well-being, particularly in platform and logistics work. This paper models the trade-off and asks whether it is inherent or a matter of design. Efficiency rises with the intensity of algorithmic management at a diminishing rate; autonomy falls with intensity but less so where workers retain discretion over how to meet algorithmic targets; and well-being depends on autonomy and, secondarily, on efficiency. A simulation of 600 work settings shows that well-being falls with intensity, with a standardised indirect effect through autonomy of −0.65 and a small positive direct effect (+0.13), and that the well-being slope is halved by worker discretion (−0.20 at low discretion against −0.10 at high). Efficiency rises comparably in both cases (slopes 0.34 and 0.36). A combined efficiency–well-being index peaks at 82% intensity under low discretion and at full intensity under high discretion, so that discretion design shifts the frontier rather than merely trading along it. The paper interprets the findings through self-determination and job-design theory and proposes design principles, namely discretion over means, transparency of criteria, contestability of evaluations and limits on monitoring, that preserve well-being without sacrificing efficiency.