Publications

» Publications » Consumption and Differential Mortality

Consumption and Differential Mortality

Authors

Abstract

It is well-established that differential mortality according to wealth or income introduces bias into age profiles of these variables when estimated on cross-sectional or synthetic cohort data.  However, little is known about whether this association is also found with consumption, and if so, how strong this association is.  In this paper we use panel data on total household spending from the Health and Retirement Study (HRS) and its supplemental study, the Consumption and Activities Mail Survey (CAMS), to estimate differences in consumption by survival status to the next survey wave. We quantify the bias in age profiles of consumption that results from differential mortality when estimating the age profiles on cross-sectional data or on synthetic cohort data.  We find that the bias is smaller than that found for wealth or income.

Key Findings

  • We find that those who are wealthier tend to live longer.
    • Analysis of HRS data for 2000-2004 showed that among single persons, average wealth two years prior to death was 81 percent of that of survivors. The median wealth of single people who died was only 45 percent that of survivors.
    • The mean wealth of couples where one member became deceased was 66 percent that of couples where both survived. The disparity in median wealth was 65 percent.
  • While those who consume more tend to be wealthier and live longer, the dying spend a greater proportion of their wealth than do those who survive. This proportion of wealth spent is greater for singles than for married persons.
  • We find that cross-sectional data on consumption does not accurately reflect life-cycle spending trajectories: the cross-sectional profiles decline much more slowly with age than the panel profiles. The results of cross-sectional profiles overestimate the needs of individuals and households for economic resources in retirement.
  • Synthetic panels provide a fairly close approximation to true panels, which means that studies based on the CEX are likely to approximate studies based on panel data.

Full Text

Download PDF

Project

Paper ID

WP 2011-254

Publication Type

Working Paper

Publication Year

2011