This graph is floating around a bit after Lee Drutman submitted some interesting analysis to The Monkey Cage. I think Andrew Gelman makes a good point that we're looking at some serious selection bias here, and it would be nice to see more zero bailout points to get a better idea of overall lobbying expenditures. Also, as one of the commenters points out, without AIG, there is arguably not much of a relationship there. Pretty graph, but I wouldn't worry about it too much. The AIG lobbying-bailout outlier simply as a relationship between size of bailout and size of assets (Y-axis only) is particularly interesting though.
