Would Greater Awareness of Social Security Survivor Benefits Affect Claiming Decisions?
Most Americans enter retirement as married couples, and one spouse, typically the wife, outlives the other. Many widows lack the income needed to maintain the standard of living they had when their husbands were alive. Widows would generally have more adequate incomes if their husbands, who are typically the higher earner in the couple, delayed claiming Social Security. This project uses the Health and Retirement Study (HRS) to test the extent to which husbands consider their wives’ well-being as widows when making claiming decisions. It then uses an online experiment to determine whether raising a husband’s awareness of the risks that his widow faces, and how delayed claiming can reduce those risks, affect his claiming behavior.
The paper found that:
- Husbands do not seem to consider the prospective drop in income experienced by their widows when choosing a Social Security claiming age.
- Husbands respond instead to immediate concerns such as pension incentives and health conditions.
- A simple information intervention that highlights the likelihood and consequences of widowhood, and demonstrates how delayed claiming enhances survivor benefits, may be insufficient to change the stated claiming intentions of older husbands.
- The framing of these information interventions (e.g., presenting the full retirement age as the default) seems to affect claiming at least as much as the content presented.
The policy implications of the findings are:
- Informing husbands that they can improve their widows’ financial well-being by claiming later may be ineffective in alleviating widows’ poverty.
- An alternate approach would develop policies to protect widows that at least partially decouple survivor benefits from the husband’s claiming age.
- For example, one common proposal would set survivor benefits at 75 percent of the couple’s combined Social Security benefit, funded by a reduction in spousal benefits while the husband is still living.