This study examines the role of car ownership in facilitating employment among recipients under the current welfare-to-work law. Because of a potential problem with simultaneity, the analysis uses predicted car ownership constructed from two instrumental variables, insurance premiums and population density for car ownership. The data come from a 1999–2000 survey of TANF recipients in the Los Angeles metropolitan area. The empirical results show a significant independent contribution of car ownership on employment. The presence of an predicted ownership is associated with a 9 percentage point increase in the odds of being employed. Moreover, the results indicate that lowering insurance premiums by $100 can increase the odds of employment by 4 percentage points. © 2002 by the Association for Public Policy Analysis and Management.