Journal of Pension Economics and Finance


Redistribution under the Social Security benefit formula at the individual and household levels, 1992 and 2004  *


a1 Department of Economics, Dartmouth College and NBER, Hanover, NH 03755-3514, USA (e-mail:

a2 Department of Economics, Texas Tech University, Lubbock, TX 79409, USA

a3 Department of Economics, Dartmouth College, Hanover, NH 03755-3514, USA


Studies using data from the early 1990s suggested that while the progressive Social Security benefit formula succeeded in redistributing benefits from individuals with high earnings to individuals with low earnings, it was much less successful in redistributing benefits from households with high earnings to households with low earnings. Wives often earned much less than their husbands. As a result, much of the redistribution at the individual level was effectively from high earning husbands to their own lower earning wives. In addition, spouse and survivor benefits accrue disproportionately to women from high income households. Both factors mitigate redistribution at the household level. It has been argued that with the increase in the labor force participation and earnings of women, Social Security now should do a better job of redistributing benefits at the household level. To be sure, when we compare outcomes for a cohort with a household member age 51 to 56 in 1992 with those from a cohort born twelve years later, redistribution at the household level has increased over time. Nevertheless, as of 2004 there still is substantially less redistribution of benefits from high to low earning households than from high to low earning individuals.

(Online publication July 09 2012)


  • D31;
  • H55;
  • J11;
  • J14;
  • J16;
  • J18;
  • J26;
  • J38


  • Social Security;
  • redistribution;
  • benefits;
  • spouse benefits;
  • survivor benefits;
  • benefit formula;
  • Health and Retirement Study;
  • women's earnings;
  • family


*  This research was supported by a grant from the U.S. Social Security Administration (SSA) through the Michigan Retirement Research Center (MRRC) under grant number UM11-06. The findings and conclusions expressed are solely those of the authors and do not represent the views of SSA or the MRRC. David Olson of the Social Security Administration was extremely helpful to us in dealing with the ANYPIA program. We also thank Mike Hurd, Olivia Mitchell and participants at the MRRC workshop for their helpful comments.