Andreas Gulyas (Universität Mannheim) – Understanding the Sources of Earnings Losses After Job Displacement: A Machine-Learning Approach

We document the sources behind earnings losses of job displacement using machine-learning algorithms. Using administrative data from Austria over three decades, we show that displaced workers face large and persistent earnings losses. We identify substantial heterogeneity in losses across workers using the generalized random forest method by Athey et. al. (2019). The most vulnerable are high-income, high-tenure workers employed at well-paying firms in highly concentrated labor markets. The single most important factor is firm wage premia, implying losses in employer specific wage components are key for the understanding of earnings losses. We further quantify how much losses of employer wage premia and other characteristics contribute to wage loss over time. Using a classical random forest we show that differences in firm-wage premia and job tenure explain the wage gap between displaced workers and the control group in approximately equal proportions. In addition, we find no evidence of a stigma effect, as displaced workers earn more than what would be expected based on their characteristics.

Freitag, 07. Juni 2019 um 11.00 Uhr

Geldzentrum der Oesterreichische Nationalbank
Veranstaltungssaal, 3. OG
Garnisongasse 15
1090 Wien

Um Anmeldung ersuchen wir bis Dienstag, 04. Juni 2019.