The Scarring Effects of Firm Shutdowns on Workers’ Wages: A Distributional Perspective

Abstract

To shed light on the differential impact of firm shutdowns across the distribution of workers, we adopt the wage determination framework of Bonhomme, Lamadon, and Manresa (2019) to uncover workers’ unobserved types. Worker types relate to workers’ position in the wage distribution: all else equal, a higher type implies higher wages. We use the universe of social security records of Italy’s Veneto region, one of the leading Italian regional economies, from 1975 to 2001. Based on this rich matched employer-employee data, we measure wage losses after firm shutdowns for different worker types using an event-study framework. Aggregate wage losses directly after a shutdown are 4.5% of the daily wage, which almost halves after six years. This aggregate trajectory masks stark heterogeneity: top-type workers face initial losses of 12.4%, which remain persistent even after six years. Conversely, initial losses for bottom-type workers are 2.6%, which become statistically insignificant after six years. We identify losses in firm tenure as the main source of wage reductions following the shutdown of a worker’s firm. Finally, we show that the AKM model (Abowd, Kramarz, and Margolis, 1999), the current workhorse model of wage determination, does not capture this heterogeneity and can lead to misleading conclusions regarding the sources of wage losses.