Using longitudinal survey data from western Kenya, this paper estimates the economic impacts of antiretroviral treatment. The responses in two important outcomes are studied: (1) labor supply of adult AIDS patients receiving treatment; and (2) labor supply of patients' household members. We find that within six months after treatment initiation, there is a 20 percent increase in patients' likelihood of participating in the labor force and a 35 percent increase in weekly hours worked. Since patient health would continue to decline without treatment, these labor supply responses are underestimates of the impact of treatment on the treated. The upper bound of the treatment impact, based on plausible assumptions about the counterfactual, is considerably larger. The responses in household members' labor supply are heterogeneous, with young boys and women work significiantly less after initiation of treatment. The effects on child labor are important since they suggest potential schooling impacts from treatment.