We apply principal-agent theory to the man-to-machine context and investigate the schism between consumers’ desire to maximize convenience and minimize risk in automated shopping. Agents in the form of self-ordering washing machines can provide additional convenience to consumers by restocking on detergent on behalf of their principals. The added convenience, however, comes at the cost of outsourcing decision-making power to machine agents. We show that technological agency increases perceived convenience, but also increases perceived risk. We further illustrate the process in which perceived risk inhibits perceived convenience in loss situations, where agents conduct purchases to their principals’ disadvantage. Controlling for human agents, we find that dissimilarity between principal and agent increases perceived risk. Our research calls for the inclusion of contractual elements in an automated shopping environment in order to avoid alienating consumers.
