Voter Migration and the Geographic Sorting of the American Electorate

This study shows that party ID voters relocate based on destination characteristics and constraints but also prefer to locate to destinations with people who share their political party’s ideology. While aiming to address the corresponding migration of individuals with locations that fit their political preferences the authors do take to consideration that geographic patterns may occur due to social processes and environments. They also note that relocation decisions are usually and primarily made due to constrained choices which are usually economic. However, if there is a desire for individuals to move to places that suit them socially and economically then the same can go for politically.
The authors find that 30% of those surveyed in their data take partisanship to be an important factor to their desired place of living. Liberal residential settings are typically urban, whereas conservatives prefer suburban and rural areas. For example, democrats are found to prefer densely populated areas more than Republicans.
According to the article republicans tend to sort rather than mix where as democrats don’t mind mixing, sometimes democrats will relocate to areas that are more conservative than their origin but this is probably because the most democratic locations are in big cities that have several social problems. The same negative features can go to the most republican places as well which are rural and have limited employment opportunities. Republicans prefer moving to more republican locations which keeps increasing the further the distance is. On the other hand Democrats can have similar preferences as republicans except their preferences are not as strong as the preferences republicans have. This can be seen in figure 1 which shows the migration in and from Portland, Oregon.

Cho, W. K., Gimpel, J. G., & Hui, I. S. (2013). Voter Migration and the Geographic Sorting of the American Electorate. Annals of the Association of American Geographers, 103(4), 856-870. doi:10.1080/00045608.2012.720229