The Invisible Employer: Algorithmic Management and the Reinvention of Control in the Gig Economy

Every employment relationship in labour law has, until now, presupposed a visible employer — a person or entity capable of being identified, questioned, and held to account for the direction it exercises over a worker. The gig economy disturbs this premise. Platforms exercise the same functions traditionally associated with an employer — assigning work, setting pay, monitoring performance, and terminating access — not through a visible supervisor but through an algorithm whose rules the worker cannot see and cannot contest.

This paper asks whether the traditional control test, built on the assumption of an identifiable, human directing mind, can meaningfully regulate a relationship in which the employer’s authority has become functionally invisible, and if not, how the law should respond. The paper adopts a doctrinal and comparative method. It examines how platforms use ratings, personalised pay, gamification and automated deactivation to manage workers in ride-hailing, delivery and freelance work, and places this alongside the legislative and ballot-driven contest over worker status in California under Assembly Bill 5 and Proposition 22, and the European Union’s 2024 Platform Work Directive, which introduces a rebuttable presumption of employment triggered by defined indicators of algorithmic direction and control.

The paper argues that algorithmic management does not eliminate employer control but relocates and conceals it, replacing a visible supervisor with an opaque system that performs the same disciplinary function while evading the doctrinal tests designed to detect it. The paper concludes that the law and workers alike need better tools to hold the invisible employer to account, starting with meaningful access for workers to the criteria used to evaluate and pay them and human review of automated decisions with serious consequences, such as deactivation.