In this study, we quantify the causal effects of commuting time and working from home (WFH) arrangements on the mental health of Australian men and women. Leveraging rich panel-data models, we first show that adverse effects of commuting time manifest only among men. These are concentrated among individuals with pre-existing mental health issues, and they are modest in magnitude. Second, we show that WFH arrangements have large positive effects on women's mental health, provided that the WFH component is large enough. The effects are once again concentrated among individuals with pre-existing mental health issues. This effect specificity is novel and extends beyond Australia: we show that it also underlies the adverse effects of commuting time on the mental health of British women. Our findings highlight the importance of targeted interventions and support for individuals who are dealing with mental health problems.