Involuntary job loss and changes in personality traits

- We investigate changes in personality traits after involuntary job loss.
- The German Socio-Economic Panel Study provides repeated personality trait measures.
- We find a causal increase in openness for the average displaced worker.
- The effect is driven by highly educated workers and direct job-to-job transitions.
- Involuntary job loss does not affect the other dimensions of the Big Five inventory.

Economists consider personality traits to be stable, particularly throughout adulthood. However, evidence from psychological studies suggests that the stability assumption may not always be valid, as personality traits can respond to certain life events. Our paper analyzes whether and to what extent personality traits are malleable over a time span of eight years for a sample of working individuals. Furthermore, we specifically look at changes in personality traits after a major adverse life event: involuntary job loss. We use data from the German Socio-Economic Panel Study (SOEP) from 2004 to 2014 - a period over which individuals' Big Five personality inventory was measured three times. Our dataset allows us to exploit detailed employment information, particularly reasons for job termination and unemployment spells. We focus solely on plant closures as a reason for job termination. Job loss due to plant closure is widely used as a relatively exogenous event to identify causal effects. Our results suggest that personality traits are indeed malleable during adulthood. Although the Big Five measures are relatively stable within the overall population of workers, we find an increase in openness, that is, the willingness to seek new experiences, for the average displaced worker. This increase, however, is fully driven by individuals with high educational attainment and by those who find a new job immediately after dismissal. The other dimensions of the Big Five personality inventory remain nearly unchanged after an involuntary job loss. Our findings hold for a number of robustness checks and are supported by the results of a falsification test using a placebo treatment.