Working Paper: NBER ID: w2888
Authors: Robert Barsky; Gary Solon
Abstract: This paper is an examination of cyclical real wage behavior in the United States since World War II. Like most previous aggregate studies. ours finds little cyclicalitv in aggregate industry real wage data. On the other hand, our analysis of longitudinal microdata from the Panel Study of Income Dynamics reveals substantial procyclicality. We find that this procyclicality is obscured in industry average wage statistics, and to a lesser extent in economywide averages, because those statistics are constructed in a way that gives greater weight to low-wage workers during expansions. The almost complete absence of evidence for countercyclical real wages suggests that movements along labor demand curves have not played a dominant role in cyclical employment fluctuations over the last 40 years. Instead, the procyclicality of real wages indicates that cyclical employment fluctuations have been generated mainly by shifts in labor demand. The sources of these shifts and of the positive slope of the effective labor supply curve, however, remain open to alternative interpretations.
Keywords: Real wages; Business cycle; Labor demand; Procyclicality
JEL Codes: E24; E32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| unemployment rate (J64) | real wages (J31) |
| labor demand shifts (J23) | employment fluctuations (J63) |
| low-wage workers (J46) | aggregate wage statistics (E10) |
| cyclical employment fluctuations (E32) | labor demand shifts (J23) |
| real wages (J31) | cyclical employment fluctuations (E32) |
| real wages (J31) | procyclical behavior (E32) |