Henry Farrell at Crooked Timber has a graph by Lane Kenworthy. I think this is better, aside from the legend-based labels:
The data is the same CBO average household income data as Kenworthy uses. I indexed it.
Note that the lower quintiles are below their 1979 levels much longer than the top 1%. The lowest quintile doesn’t exceed 1979 levels until 1994, which is the same time that the top 1% really takes off. The volatility of the top 1% is amazing, however, which leads me to believe that it is related to stock market movements, given the 2000 and 2001 drop-offs.
