Diversify or Specialize? Demand-Reputation Trade-Offs and Career Paths in Online Labor Markets
43 Pages Posted: 5 Sep 2019 Last revised: 12 Sep 2019
Date Written: August 28, 2019
In online labor markets (e.g., Upwork, Freelancer) the ability of contractors to charge for their services largely depends on their skills and expertise. The value of each skill is bound to dynamic market conditions, i.e., the skill's demand and supply. To keep up with shifting demands, contractors need to continuously keep reskilling themselves. This dynamic nature of market trends often presents contractors with the dilemma to either diversify their skillsets and significantly alter their career paths, or specialize by learning (and improving on) similar skills to their current skillsets.
This work tackles this dilemma by investigating how different career choices of online workers affect their market values. Through a mathematical modeling, we argue that, compared to diversification, specialization yields reputation gains and demand losses. These two competing forces define the net effect of each career choice on a contractor's market value, which we capture through a contractor's hourly wages and hiring rates. To empirically study this demand-reputation tension we propose a three-component methodological framework that combines word embeddings, Hidden Markov Models, and fixed effects specifications to isolate the effects of different career choices. Analysis of 693,260 job applications from a major online labor market shows that, for hourly wages, reputation gains do not compensate for demand losses; hence, compared to diversification, specialization leads to lower wages. However, for hiring rates, reputation gains overcompensate for demand losses. As a result, compared to diversification, specialization increases contractor hireability.
Keywords: Online labor markets, Career choices, Career paths, Skillset specialization, Skillset diversification, Empirical analysis
Suggested Citation: Suggested Citation