The existing social security system in the U.S. has a special provision for married households: a married person can choose between own benefits and half of the spouse's benefits. Another feature of the system is the progressive calculation of benefits: benefits are determined by a concave function of past mean earnings. I develop an equilibrium life-cycle model to quantify the aggregate, cross-sectional, and welfare implications of three alternative scenarios: elimination of the spousal benefits, elimination of the progressivity of benefits, and the two changes combined. Agents start out as permanently married or single and with education levels and wage profiles, where the latter depend both on education and gender. The household is the decision maker and decides on the labor supply of its member(s) and saving. The aggregate production function has as inputs capital and labor aggregated by efficiency. Eliminating the spousal benefit provision has substantial effects. The labor force participation of married women increases by 4.5% and households composed of men with relatively high education and women with relatively low education experience significant welfare losses. When only the progressivity is eliminated, there is a decline in labor force participation of married females and households composed of men with relatively high education and women with relatively low education experience significant welfare gains. When both are eliminated, the labor force participation of married women increases and households composed of two members with high education gain most.
