Implications of Worker Classification in On-Demand Economy

How should workers in the on-demand economy be classified? We study this policy question focusing primarily on the welfare of long-term (LT) workers, who depend on gig jobs as primary income sources. We develop a queueing model with a service platform and two types of workers: LT workers who base their joining decisions on the long-run earning rate, while ad hoc (AH) workers who participate according to real-time payoffs. We identify two issues with uniform classifications: when all workers previously treated as contractors are reclassified as employees, the profit-maximizing company may undercut workers, and LT workers’ average welfare can decrease; when all are reclassified as “contractors⁺”, an intermediate status that provides incomplete employee benefits but allows workers to self-join, workers can overjoin such that LT workers’ utilization rate will remain low and their welfare may not be enhanced. We then consider a discriminatory scheme that classifies LT workers as employees but leaves AH workers as contractors. This hybrid mode suffers from undercutting but curbs overjoining. More importantly, it can do less harm to consumers and the platform operator. We also study a discriminatory dispatch policy that prioritizes LT workers over AH workers. This operational approach can simultaneously counteract undercutting and overjoining. Finally, we empirically calibrate the model and apply our insights to the ride-hailing market in California.