Hearing about a Job: A Model of Differential Information Flow and Job Matching

Which people end up in which jobs is not merely a matter of the individual and human capital characteristics of workers and the requirements and rewards of jobs, but is also a function of the process by which persons and jobs are matched with one another (Granovetter 1981; Sørensen and Kalleberg 1981). A poorly understood component of the matching process is how workers and employers find information about each other. We propose a framework for analyzing the dynamics of labor market behavior that emphasizes both the two-sided nature of the matching problem and the potential effects of different information structures. The basic framework for describing labor markets is flexible, and can be used to represent common types of labor markets including spot markets for labor and systems characterized by competition over vacancies. Through simulation, we use this framework to explicitly vary the structure of information flow between workers and employers, thereby allowing us to explore how different recruitment strategies influence labor market performance in different institutional environments. Preliminary results suggest that full information is the most efficient and egalitarian information regime. However, in key respects, information that flows through socially structured networks produces macro-level and individual-level outcomes quite similar to those observed in the full information regime, although with lower mobility rates. In contrast, unstructured limited information produces more egalitarian outcomes than either full information or social networks, by giving workers equally bad opportunities.