Search
Program Calendar
Browse By Day
Browse By Time
Browse By Subject Area
Browse By Session Type
Search Tips
Conference
Virtual Exhibit Hall
Location
About NTA
Personal Schedule
Sign In
We examine changes in retiree income over time using cross-sectional tax data over 2003-2017. Brady and Bass (2024) found that median spendable income fell by 1.0 percent per year of age, on average, from age 70 through age 98. That study also found that the incidence of non-Social-Security retirement income, which held steady at 70 percent or higher from age 71 through age 91, declined after age 91. These results raise the question of whether these age patterns reflect the typical experience of aging individuals or the different life experiences of each birth-year cohort. This study analyzes synthetic cohorts to disentangle these age and cohort effects. We find the decline in income at older ages is a cohort effect. Controlling for birth year, median inflation-adjusted income does not decline with age. For example, individuals born in 1936 had the same median income when aged 85 (in 2017) that they had back when they were aged 71 (in 2003), adjusted for inflation. Nonetheless, their median income was nearly 20 percent less than individuals aged 71 in 2017. Similarly, once we control for birth year, we see much more modest declines in the incidence of non-Social-Security retirement income after age 90. Rather than being a cause for concern, the decline in income with age observed in cross-sectional data is consistent with an improvement in retiree income. Our analysis shows that, on average, all birth-year cohorts maintain their inflation-adjusted income in retirement but that younger cohorts have entered retirement with higher income.