Have 401(K)S Raised Household Saving? Evidence from the Health and Retirement Study

Abstract

401(k)-type pension arrangements are the most popular tax subsidy to household saving in the U.S. This study uses self- and firm-reported pension information, Social Security, and household wealth data from 1992 Health and Retirement Study (HRS) to examine the extent to which 401(k) pension plans have raised household saving. Comparison of self- and firm-reported pension information indicates significant measurement error in self-reported 401(k) eligibility. This error has biased the estimated 401(k) saving effects in all previous studies upward significantly and differentially by income category. There is evidence of significant measurement error in pension assets as well. Overall, the estimates that account for both types of measurement error suggest that 401(k)s have not raised household saving. All of the estimates are significantly lower than those implied by previous studies that have found large effects. The most plausible explanation for the large estimated offset to household saving is firm-level substitution of 401(k)s for other pensions. Even though very little of the average dollar of 401(k) wealth appears to be new household saving, 401(k)s may have stimulated saving significantly for lower-to-middle income households and, hence, increased retirement income security for an important segment of the population.