The current U.S. Social Security program redistributes resources from high-wage workers to low-wage workers through its progressive benefit schedule and from two-earner couples and singles to one-earner couples through its spousal and survivors benefits. This paper extends a standard general-equilibrium overlapping-generations model with uninsurable wage shocks to analyze the effect of the spousal and survivors benefits on the labor supply of married households and the overall economy. The heterogeneousagent model calibrated to the current U.S. economy predicts that, in the long run, removing spousal and survivors benefits would increase the female labor participation rate by 1.4%, the total working hours of women by 1.6–1.7%, and the total output of the economy by 0.5–0.6%. Under the balanced-budget assumption, a phased-in cohort-by-cohort removal of these benefits would make all age cohorts, on average, better off, although the policy change would make a majority of young married households worse off in the short run.