In my first book, Voting and Migration Patterns in the U.S., I wrote quite a bit of text on the migration patterns of various age cohorts during the 1990s. One of my key findings was that young people (under 30) were flocking to those big, overwhelmingly Democratic counties, and older people tended to move out of these places. At the time, I had not yet looked at net migration rates by county for 2000 and 2010. It is now pretty easy to do, now that we have access to accurate net migration rates by age, race, and county (see here).
I was very curious to know if the same trends we saw in the 1990s continued, or if those people who moved into those big blue metropolises decided to stay there.
As it turns out, we see the exact same pattern between 2000 and 2010 that we saw between 1990 and 2000. That is, young people moved into these overwhelmingly Democratic counties, and older people moved out, often settling in overwhelmingly Republican counties.
For this analysis, I focused specifically on those "landslide counties" -- those that gave more than 60 percent of the vote to one of the two major parties. This is the standard that Bill Bishop used in his book, The Big Sort. I used the 2000 presidential election when classifying counties.
There are a couple of things to note about these two county types. There are far more landslide Republican counties than landslide Democratic counties, but the landslide Democratic counties typically have far more people. That is, landslide Democratic counties tend to be places like King County, WA (home of Seattle), whereas landslide Republican counties tend to be places like Garfield County, WA (whose largest city is Pomeroy, population 1,425).
I should therefore note that I don't think young people are flocking to these big Democratic counties because they are Democratic (though maybe some of them are). Rather, these are the big fun cities with lots of things to do and more job opportunities.
In any event, as it was in the 1990s, these big cities lose their appeal once someone reaches a certain age. I remember my apartments on Capitol Hill in Washington, DC. When I was 23, I was willing to pay an outlandish sum for a room the size of a cubicle. Now that I'm married with kids, that would be ridiculous -- especially since I don't have time to make use of big city cultural amenities anyway.
This potentially matters politically, since where one lives does seem to have an impact on a person's political views. A move during formative years to a big liberal metropolis will probably nudge people in a more Democratic direction. (See Thad Brown's dated, but still excellent, book on the subject.) This may be playing a role in the degree to which Millennials lean Democratic.
There is another consideration to keep in mind. It is possible that these different migration patterns by age cohort actually reflect different migration patterns of different racial groups; non-Hispanic whites are, on average, much older than racial and ethnic minorities. For that reason I present these migration patterns for all groups, as well as non-Hispanic whites alone. The results are below.
If anything, the results are even more dramatic when we look just at whites. Like everyone else, young whites move in the millions to these overwhelmingly Democratic counties. At around the age of 30, however, the trend reverses dramatically, and far more whites move to overwhelmingly Republican counties. I should note that I could have also included trend lines for those politically competitive counties, but I chose not to because it cluttered the figures too much. In all cases, the trend line for the competitive counties fell in between the other two.