We have very little research on how substance use impacts employment among welfare mothers. But welfare reform's emphasis on moving aid recipients into the workforce has brought this issue to the fore. Using Cox proportional hazard and logistic regression in a longitudinal study of California welfare mothers in 2001-2003, we examine how substance use impacts the ability to move from welfare to work and to remain economically independent after welfare. While education, work history, and family size consistently predict transitions from welfare to work and back again, substance use-related problems consistently do not. However, the jobs obtained by welfare mothers are short-term and poorly paid regardless of whether they misuse alcohol or use drugs. We argue that, if all that is open to welfare mothers are short-lived work assignments, substance use may have little time to impact job retention. Limitations of the study are noted.