The Structure and Dynamics of U.S. Labor Market Networks

• Activity Number: 150
• Type: Invited
• Date/Time: Monday, August 4, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Kevin McKinney*+ and John M. Abowd
• Companies: U.S. Census Bureau and Cornell University, ILR School
• Keywords: Labor ; Economics ; Network ; Graph

Abstract

The advent of large-scale administrative longitudinal linked worker and firm datasets has the potential to greatly increase our understanding of the labor market. These new data sources allow researchers to follow workers over many years, capturing earnings at their initial as well as all subsequent covered employers. The observed worker mobility combined with a virtual universe sample of both workers and firms enables the estimation of labor market networks or graphs for an entire region or country. In this paper we focus our analysis on the worker-worker and the firm-firm graphs. Using data from the U.S. Census Bureau's LEHD program, we examine the evolution of both graphs, reporting the number of new edges and nodes generated each quarter. New edges that replicate an existing edge do not alter the structure of the graph, while completely new edges and nodes have the potential to transform it. Once we have a better understanding of the graphs' dynamics, we examine the connectedness, the degree distribution, and various distance measures; both for the entire graph and by various worker and firm characteristics.