The Nature of Countercyclical Income Risk

This 2014 Journal of Political Economy paper (circulated as Federal Reserve Bank of Minneapolis Staff Report 476, December 2012; the Minneapolis Fed’s own hosting of this PDF is currently unreachable, so the text was retrieved via a Wayback Machine snapshot of that same document) studies how individual labor-earnings risk varies over the business cycle, using a confidential U.S. Social Security Administration panel of uncapped earnings histories for a 10% random sample of American men, 1978-2010 (165 million individual-year observations). Decomposing earnings growth into a systematic (“between-group”) component predictable from a worker’s pre-recession characteristics and a residual (“within-group”) component representing idiosyncratic shocks among otherwise-similar workers, the paper’s central finding directly contradicts an influential earlier result (Storesletten, Telmer, and Yaren 2004): the variance of idiosyncratic earnings shocks is not strongly countercyclical. Instead, what varies sharply over the cycle is the skewness of shocks – during recessions the upper end of the earnings-growth distribution compresses toward the median (large upward moves become rarer) while the lower end expands (large downward moves become more common), so overall dispersion changes little even as shocks become systematically more left-skewed and risky. On the systematic side, the paper finds that a worker’s own average earnings just before a recession is a powerful predictor of how that recession will treat him: earnings losses rise roughly linearly with pre-recession earnings rank up to about the 90th percentile, but this pattern reverses sharply at the very top, with workers in the top 1% suffering disproportionately large and persistent losses in the two most recent recessions (though not in the 1980-83 downturn) – a result the authors describe as “dramatically different” from the pattern for everyone else, even those just below them in the income distribution.

Jae Song | Macro Paper Warehouse

The Nature of Countercyclical Income Risk