An important note, though: While this data is useful for relative measurements, the absolute gain or loss in income % is less meaningful. For one thing, they're comparing ~30-yo's today to their parents in the "late '90s". Let's say I was 32 in 2012, and we compare my income with my parents' in 1998. In 1998, I would have been 18, which would mean my parents were likely at least 38, and had a few more prime years of earnings growth racked up.
Relatedly, my understanding is that there is a long-term trend towards later earnings years, tied to more professional and graduate education. Today's 30-yo just graduating from medical or business school might be primed to make great money later even if their parents were making more at age 30.
Just fascinating data.
The middle tier is mainly made up of areas where moving from a suburb to a major city does not involve getting on a plane to visit relatives. (Chicago being the exception, not sure why, maybe its proximity to Michigan?).
The lowest areas are primarily the deed south without access to oil/nat gas reserves. The deep south has a triple challenge of low income to reinvest in to school systems, limited opportunities or areas near Detroit which have seen a complete collapse of jobs.
This map is basically a heat map of areas which saw rapid changes to their underlying economies because of forces they could not control.
IMO access to efficient public transportation has an outsized impact on people's ability to move up the economic ladder.
I'd like to see a map of how many of these income reporters moved away from their locales and to where and also what professions they picked up.
Of course, that "gotta get out of here" feeling isn't limited to lower income kids, so it could be a factor. Certainly it's typically good for your earnings if you're willing to move where you can make more money. But then why isn't upstate New York dark blue? (Sorry, upstate New York!)
which I think explains the results you see almost entirely