The Impact of the Covid-19 Outbreak on theGeographical Labour Mobility in Sweden’sMunicipalities

Abstract: This study examines the impact of the COVID-19 pandemic on the geographical labour mobility of employees in Sweden. The analysis employs a difference-indifference (DiD) approach to compare the mobility patterns of employees in treated municipalities with higher COVID-19 infectious rates to those in controlled municipalities with lower infectious rates. The study utilizes two primary datasets: population-wide micro-level data on COVID-19 infectious rates provided by the Swedish Public Health Agency, and aggregate data on employed commuters from Statistics Sweden. The latter dataset includes information on the residential and employed municipalities of individuals aged 16 and above between 2014 and 2021. The analysis focuses on the geographical mobility ratio which represents the proportion of employed commuters with different residential and employed municipalities. Regression models are used to estimate the treatment effect of COVID-19 on geographical labour mobility, controlling for municipality-specific factors and potential confounding variables. The results indicate that the treated municipalities with higher COVID-19 infectious rates exhibit significantly higher geographical labour mobility than the controlled municipalities. However, the main effect of time, representing the impact of the COVID-19 outbreak, is found to be statistically non-significant, suggesting that the pandemic did not have a direct effect on workers' geographical mobility. Gender is found to have an equal influence on commuters' mobility patterns, irrespective of the impact of COVID-19. The study contributes to the understanding of labour mobility dynamics during a pandemic and highlights the importance of considering contextual factors such as infection rates and demographic characteristics. It also underscores the need for further research to explore the nuanced factors influencing remote work preferences among different demographic groups.