FSRDC Presentation Series — April 24th, 2024
“Interpreting Cohort Profiles of Lifecycle Earnings Volatility”
Presented by Christopher Bollinger and Charles Hokayem
April 24th @ 1:00 pm ET
To join online click the “Join Us” link below.
Join Us!
Abstract:
We present new estimates of earnings volatility over time and the lifecycle for men and women by race and human capital. Using a long panel of restricted-access administrative Social Security earnings linked to the Current Population Survey, we estimate volatility with both transparent summary measures, as well as decompositions into permanent and transitory components. From the late 1970s to the mid 1990s there is a strong negative trend in earnings volatility for both men and women. We show this is driven by a reduction in transitory variance. Starting in the mid 1990s there is relative stability in trends of male earnings volatility because of an increase in the variance of permanent shocks, especially among workers without a college education, and a more attenuated trend decline among women. Cohort analyses indicate a strong U-shape pattern of volatility over the working life, which comes from large permanent shocks early and later in the lifecycle. However, this U-shape shifted downward and leftward in more recent cohorts, the latter from the fanning out of lifecycle transitory volatility in younger cohorts. These patterns are more pronounced among White men and women compared to Black workers.
View the Presentation flyer at:
FSRDC Presentation Series 4-24-24 (002)
*The FSRDC Presentation Series offers a virtual platform for researchers who have conducted research in an FSRDC using Census Bureau data to share their work. Though the Census Bureau received a programmatic benefit from this work, extramural researcher(s) were neither employed nor funded by the U.S. Census Bureau. The Census Bureau and its partner agencies have reviewed the statistical output contained in these publications and presentations to ensure against any unauthorized disclosure of confidential information. However, all findings and interpretations of these research results are made exclusively by the researcher and do not represent the views or positions of the U.S. Census Bureau or our partner agencies or institutions.