Consumption, retirement and social security: Evaluating the efficiency of reform that encourages longer careers

This paper proposes and analyzes a Social Security reform in which individuals no longer face the OASI payroll tax after, say, age 54 or a career of 34 years, and their subsequent earnings have no bearing on their benefits. We first estimate parameters of a life-cycle model. Our specification includes non-separable preferences and possible disability. It predicts a consumption–expenditure change at retirement. We use the magnitude of the expenditure change, together with households' retirement-age decisions, to identify key structural parameters. The estimated magnitude of the change in consumption–expenditure depends importantly on the treatment of consumption by adult children of the household. Simulations indicate that the reform could increase retirement ages one year or more, equivalent variations could average more than $4000 per household, and income tax revenues per household could increase by more than $14,000.

► We offer a quantitative evaluation of a plausible Social Security reform.
► Simulation of the reform shows that careers lengthen and welfare gains are large.
► Estimation draws new inference from the change in expenditure at retirement.