There’s an explanation further down in the tweet thread:
So, @JoelMcGuire12 actually used the available data, which is found via this: https://pnas.org/content/118/4/e2016976118#sec-13. In that, there actually is a drop at the end. I think the paper’s author must have merged some of the income brackets at the top end into one, but not 100%.
Oh. Actually, underlying data and graph in paper don’t seem to match up. [...]
There’s an explanation further down in the tweet thread:
Both charts seem to have the same number of data points so I don’t think this is the answer. I think Joel has plotted a different dataset.
Here is it with the points and colors from the original graph (using http://www.graphreader.com/v2 Datapoints and log graph: https://docs.google.com/spreadsheets/d/14huISqo8Q3sfJALqG6D51o4DbWbPeeqgDm5ysrGxSY8/edit?usp=sharing )
Thanks!