Understanding Changes in the Duration of Widowhood in the United States

Sven E. Wilson, Brigham Young University
Joseph Price, Brigham Young University
Adriana Lleras-Muney, University of California, Los Angeles
Robert A. Pollak, Washington University in St. Louis
Josh Nicholls, Brigham Young University
Sami Adamson, Brigham Young University
Charles Hutchinson, UCLA

The past century and a half has seen vast improvements in life expectancy that have implications for the quality of life in old age, and that quality is influenced by the duration of marriage and the length of widowhood. Though much is known about the determinants of individual life expectancy, much less is known about the joint life expectancy of couples because life tables typically do not account for 1) whether individuals are married and 2) to whom they are married. This analysis uses the new Census Tree data source to estimate cohort-based couple life expectancy for the cohorts born between 1960 and 1920 in the United States and compare those tables to period and cohort tables from the SSA. Our preliminary results reveal that estimates of the time in widowhood are quite different between the cohort couple tables and both the period and cohort tables from individual life tables. These comparisons allow us to decompose changes over this critical period between changes in individual life expectancy for men and women; changes due to marital selection on the extensive margin (selection into marriage); and changes due to selection along the intensive marriage (assortative mating).

No extended abstract or paper available

 Presented in Session 26. Mortality Over the Life Course: Effects of Social, Family, and Historical Contexts