Home office : Causal Evidence on Selection and Location Effects from Telecommuting

The practice of telecommuting has increased during recent years; however, the COVID-19 crisis has accelerated its development even further. Despite various opinions on telecommuting, many companies struggle to establish long-term telecommuting policies; additionally, causal evidence on its implications for companies is largely missing. In a web-based experiment designed to distinguish between selection and pure location effects, we find that individuals who prefer to telecommute have different personality characteristics than those who prefer the workplace (e.g., risk preferences, agreeableness). Moreover, while we do not find a difference in the effort that individuals expend while at the workplace versus while telecommuting, the results indicate that they are less honest when working at home. This reduced honesty can be fully explained by a selection effect, namely, that individuals who are less honest prefer to telecommute. Overall, we provide evidence that telecommuting has important implications for the type of employees that companies attract and retain, but the location itself (home versus workplace) does not affect individuals' behavior