This study, using panel data of original surveys conducted in June 2020 and July 2021, analyzes the changes in adoption and productivity of working from home (WFH) during the COVID-19 pandemic. The results indicate, first, the mean WFH productivity has improved by more than 10% in the past year, although it is still about 20% lower than when working in the office. 1) "Selection effect" arising from the exit of workers with relatively low WFH productivity from the WFH practice and 2) the improvement in WFH productivity through the "learning effect" contributed almost equally to the productivity growth of WFH. Second, additional working hours extracted from reduced commuting are about 3.0% and 0.7% of the total labor input of WFH workers and all workers, respectively. Even if adjusting for the additional working hours from reduced commuting, the conclusion of relatively low productivity at home is essentially unchanged. Third, the percentage of employees who want to continue frequent WFH after the end of the COVID-19 pandemic has increased substantially, suggesting that WFH may become a popular workstyle.
The English version of this paper is 21-E-078.