OBJECTIVE: Assignment of representative payees, third parties responsible for managing clients' funds, has been proposed to counter potential use of public support payments for abused substances by people with severe mental illness and substance use disorders. This study examines substance use outcomes in a sample of homeless persons with serious mental illness and substance use disorders, some of whom were assigned representative payees. METHODS: The subjects were participating in the Access to Community Care and Effective Services and Supports (ACCESS) program, a federally funded demonstration program on integrating service systems. Clients were assessed at baseline and three months after case management services were initiated. Factorial repeated-measures analysis of covariance was used to examine substance use among four client subgroups, two of which had payees and two of which did not. RESULTS: Clients in this sample (N = 1,348) showed significant improvement on all measures of substance use over the first three months in the program. Those with payees showed no greater improvement in substance abuse than those without payees, although they did have fewer days of homelessness. CONCLUSIONS: This study failed to find evidence that merely adding external money management services to existing services improves substance abuse outcomes among clients who had dual diagnoses and were homeless. Besides assigning a payee, structured behavioral interventions may be needed to produce additional clinical benefits.