Tracking employment shocks using mobile phone data

J. Toole, Y. Lin, E. Muehlegger, D. Shoag, M. Gonzalez, D. Lazer
Journal of the Royal Society Interface
2015;12 (107)
May 26, 2015

Abstract

Can data from  mobile phones be used to observe economic shocks and their consequences at  multiple scales? Here we present novel methods to detect mass layoffs,  identify individuals affected by them, and predict changes in aggregate  unemployment rates using call detail record (CDR) data from mobile phones.  Using the closure of a large manufacturing plant as a case study, we first  describe structural break and Bayesian classification models to detect a mass  layoff and the individuals affected by it by observing changes in calling  behavior. For these affected individuals, we find measure significant  declines in social behavior and mobility following job loss. We then apply  these findings to the macro level and show that the same changes in these  calling behaviors, aggregated at the regional level, can improve forecasts of  unemployment rates.

Related publications