Online Labor Markets: Reputation Transferability, Career Development Paths and Hiring Decisions

In an online labor marketplace (OLM) employers post jobs, receive freelancer applications, and make hiring decisions. Once hired, freelancers complete the tasks online and receive their payment along with feedback about their performance. Because of the natural heterogeneity that appears in task categories, skills, and the latent abilities of freelancers and employers, these markets stuffer from a series of inefficiencies. In this work I focus on understanding these inefficiencies and propose solutions. In particular, I study three problems: (1) Reputation Transferability, (2) Skills Recommendations, and (3) Hiring Decisions. I start by proposing and evaluating different approaches that explain how freelancers’ reputation transfers across different task categories. I then propose to examine the utility of skills in an OLM, given the level of expertise of the freelancer and the demand of each skill in the marketplace. Based on this analysis, I further propose to build a career development framework. Next, I build a series of explanatory and predictive models that describe employers’ hiring decisions. In addition I study how each one of the available freelancers-employers characteristics affect hiring decisions. Finally, I conclude by highlighting the impact of my work on OLMs, and pose a series of questions that need to be addressed next.