Abstract:
Freelancers face cold-start problems in online labor markets: getting hired is very difficult without ratings, while obtaining a rating is impossible unless already having been hired. According to economic theory and empirical evidence, advertising can serve as a signal of product quality for experience goods. As such, advertising might help skilled new freelancers without reputation on a platform to obtain a first job, by providing a quality signal to employers. This study empirically explores the role of advertising in online labor markets using transactional data from a major platform. While indeed newer freelancers tend to advertise, I find that buyers dislike ads once I control for the increased visibility of ads. This negative effect is amplified for new and unrated freelancers compared to already rated freelancers. Furthermore, I find that new freelancers who advertise do not perform significantly better in the long-run compared to similar freelancers who do not advertise. Taken together, my results contrast the hypotheses derived from signaling models of advertising.