Estimating the effect of working from home on parents’ division of childcare and housework: A new panel IV approach

Abstract

This study investigates whether (and how) working from home (WFH) affects the gender division of parental unpaid labor. I use the recent COVID-19 pandemic that brought an unanticipated yet lasting shift to WFH, combined with a measure of occupational WFH feasibility (Alipour et al. 2023) as a quasi-experiment to employ an instrumental variable (IV) approach and estimate causal effects. I use unique longitudinal data from the “Growing up in Germany” (AID:A) panel study, which administered a pre-pandemic wave in 2019, and a post-pandemic wave in 2023. AID:A contains rich information on mothers’ and fathers’ time use for work, commuting, childcare, and housework. I find that the most robust effects emerge for frequent (at least once a week) paternal WFH on parental division of housework: families in which fathers start weekly WFH in the period 2019 to 2023—due to their occupational WFH capacity in combination with the pandemic WFH-boost—experience a significant decrease in the maternal share of housework. Interestingly, this shift appears to be mainly driven by a reduction of maternal time use for housework (combined with an increase in her work hours) and less by an increase in paternal time use for housework, suggesting cross-parent effects of WFH. Further analysis confirms changes in paternal gender role attitudes as a plausible mechanism.