I appreciate the discussion about the geographic unit of analysis. How should researchers measure distance migration? Some facts from the paper:
The number of people who change residences within the United States each year is large: roughly 1.5 percent of the population moves between two of the four Census regions (Northeast, Midwest, South, and West) annually, and about the same number of individuals (roughly 1.3 percent of the population) move to a different state within the same region, as shown in Figure 2. In addition, roughly 3 percent move across counties within the same state. All together, in each year between 5 and 6 percent of the population moves across a county boundary, which is often a long-enough distance to make a meaningful difference in their local housing and labor market environment.
Inter-regional migration is as significant as interstate migration within a region. Of course, tracking interstate migration captures both types of movers. I'd hazard to guess that quite a few inter-regional migrants moved to the next state over if it happened to be in a different region. Interstate migration is a pretty good proxy for distance migration. I doubt the Census regions serve much purpose for understanding migration patterns.
I'm interested in this discussion because I want to identify migrants who are likely risk-takers. Think like an immigrant. Crossing a state line is a big deal. Crossing two state lines (or more) might be a good indicator of an entrepreneurial disposition. The further talent has moved away, the more important they might be to the revitalization to the community.