Estimating Trends in Male Earnings Volatility with the Panel Study of Income Dynamics

Abstract

The Panel Study of Income Dynamics (PSID) has been the workhorse dataset used to estimate trends in U.S. earnings volatility at the individual level. We provide updated estimates for male earnings volatility using additional years of data. The analysis confirms prior work showing upward trends in the 1970s and 1980s, with a near doubling of the level of volatility over that period. The results also confirm prior work showing a resumption of an upward trend starting in the 2000s, but the new years of data available show volatility to be falling in recent years. By 2018, volatility had grown by a modest amount relative to the 1990s, with a growth rate only one-fifth the magnitude of that in the 1970s and 1980s. We show that neither attrition or item nonresponse bias, nor other issues with the PSID, affect these conclusions.

Keywords:

• Earnings
• Panel data
• Volatility

Supplementary Materials

The supplementary appendix to this article provides discussion on response error, attrition, and nonresponse and imputation, and additional results discussed in the article.

Acknowledgments

The authors would like to thank Joseph Altonji for comments on a version presented at the 2019 AEA meetings as well as the participants of a July 2019 conference sponsored by Equitable Growth in Cambridge, Massachusetts for their comments. Comments from James Ziliak and Michael Carr are appreciated as well as input from the other team members on this joint project: John Abowd, Christopher Bollinger, Charles Hokayem, Kevin Mckinney, and Emily Wiemers. Assistance from David Johnson was valuable and the comments from an Associate Editor and two referees were also helpful. The authors would like to dedicate this article to the memory of our long-time collaborator, mentor, and friend, Peter Gottschalk, who passed away on March 25, 2021.

Disclosure Statement

The authors have no competing interests to declare.

Additional information

Funding

Zhang would like to thank the financial support from the National Natural Science Foundation of China (NSF72174074) and the National Social Science Fund of China (21FJYB022).