Prospective Social Security Wealth Measures

This data set contains calculated and imputed Social Security wealth measures for HRS respondents at three points in time: 1992, 1998, and 2004. The data files were created using information from HRS respondents who authorized the University of Michigan's Institute for Social Research to obtain administrative records from the Social Security Administration.

These data are organized in three files, each containing wealth information at a separate time point: 1992, 1998 and 2004. The unit of observation is an individual, uniquely identified by the concatenation of the household identifier (HHID) and person number (PN). For each wave, benefits are calculated assuming that the respondent has not attained claim age as of the wave date. If the respondent is older than a particular claim age, missing values are assigned.

Wealth variables are first rounded to the nearest $100 and then top-coded at the top 2 percent. PIAs are rounded to the nearest $5 and then top-coded at the top 2 percent. Values exceeding the top 2% have been replaced with the mean of the top 2% for each wave.

Product Details

1992: 12,651. 1998: 22,879. 2004: 21,820.
Product Type
Social Security Administration (SSA) Data
Entry Cohort Info
Latest Release October 2009 (v4.0)
Version Notes

This version uses an improved imputation strategy for household wealth values.


It should be noted that for each public release data set all geographic variables have been aggregated to the Census Division level. Files derived from SSA administrative data may not be merged with any data containing geographic identifiers at a level of aggregation finer than Census Division. In addition they may not be merged with any other Restricted Data Set(s) without the explicit written permission of the Health and Retirement Study.