Flows and Boundaries: A Network Framework for Understanding Occupational Mobility in the Labor Market

• In Event: 2204 - New Approaches to Understanding and Addressing Inequality

Abstract

Building upon the long-established thesis on the usefulness of mapping out the underlying boundaries that govern worker flows for understanding the labor market structure, this article formulates a network framework for analyzing intragenerational occupational mobility. We argue that current the literature faces three major challenges: (1) the determination of boundaries, (2) the incorporation of over-time changes in boundaries, and (3) the incorporation of multi-step flows. These challenges motivated us to propose a network framework that helps address them, one that utilizes the detailed occupation-to-occupation flows to uncover the latent boundaries and analyze mobility patterns within and across them. In this network, the ``nodes'' are the detailed occupations and the ``weighted, directed edges'' are defined by the volume and direction of workers who flow between the nodes. We then use a community detection algorithm to detect the mobility boundaries based on the observed mobility network. We apply this network approach to the analysis of trends in intragenerational occupational mobility in the United States from 1989 to 2015 and compare the recovered boundaries with those defined by the macro-, meso-, and micro- class schemes proposed by Weeden and Grusky (2005). Contrary to the time-invariant class schemes assumed in previous work, our results suggest that the boundaries that constrain mobility opportunities change over time. Further, we show that failure to account for these changes may lead to an overstatement of the level of fluidity in intragenerational occupational mobility and an understatement of the growth in the rigidity of mobility boundaries over time.

Authors

• Barum Park

• Siwei Cheng, New York University