The Effects of Social Security on Private Savings: A Reappraisal of the Time Series Evidence

Section I reviews some of the important contributions using time series evidence to estimate Social Security’s impact on private savings. Essential to these studies is the use of a “Social Security Wealth” (SSW) variable, created by Martin Feldstein (1974), which defines the present value of future discounted Social Security benefits for the entire population under the assumption that each working person retires at the normal retirement age. Section II updates this Social Security Wealth series using both an approximation of Martin Feldstein’s original algorithm and two additional methods suggested by Dean Leimer and Selig Lesnoy (1982). The methodology and assumptions are updated to better account for the richer types of data available today. Section III provides details on the regression specifications and data to be used. Section IV follows with the empirical results of these regression specifications and the point estimates of Social Security’s effect on personal savings. Finally, in Section V provides an analysis of the effectiveness of these estimation techniques. This study finds some support that Social Security Wealth creates an adverse effect for private savings, as private savings is reduced by about half when using the most realistic assumptions for estimating Social Security Wealth, but the results are not robust across different types of specifications.