Democrat Bruce Braley's campaign for an Iowa's Senate seat has seen its share of stumbles, perhaps none more damaging than his dismissal of the state's longstanding Sen. Chuck Grassley as a "farmer from Iowa." Iowans, needless to say, were not enthusiastic about using their longstanding livelihood as a pejorative. While Braley still trails in the race, there is one bit of data that might work in his favor: Iowa is becoming more urban. And if recent history is a guide, that means that they will likely be more Democratic.
A lengthy piece in Tuesday's New York Times by Michael Barbaro explores the rural/urban dynamic in the state in the frame of the contest between Braley and Republican Joni Ernst. But a look at national trends shows how population density and voting patterns are linked.
Using data from the Census Bureau, we divvied up each county in the United States into one of six categories:
- Counties that are 90 percent urbanized areas or more, which we gave the label SU
- Counties that are 50 to 90 percent urbanized areas (U)
- Counties that are more than half urban clusters (small cities) (S)
- Counties where there is no majority of urban, rural, or urban clusters (M)
- Counties that are 50 to 90 percent rural areas (R)
- Counties that are 90 percent rural areas or more (SR)
When you map that segmentation, it looks like this, with darker colors representing more urban areas.
There's a clear pattern in how those different areas vote. Comparing each county's vote to the national average in each election, we were able to see how the different types of county have changed since 1988. Between the 1988 and 2012 elections, heavily urban counties, the SUs, were 32 percentage points more Democratic in their voting than the average. The heavily rural counties were about 11 percentage points more Republican -- shifting right an average of two percentage points each election.
If you plot every county's urban-versus-rural divide by the per-election average change in the vote, the pattern is clear: more urban areas vote have been voting more Democratic. Also notice that there are far more counties that are more heavily rural. The national average of all counties in presidential elections has become more Republican than the overall vote because there are more rural counties -- but fewer rural voters.
What's particularly interesting is to see how each county has changed over time. Notice that urban areas, counties containing cities, have grown more Democratic; the counties surrounding them (usually suburbs) less so.
This reflects what journalist Bill Bishop calls The Big Sort, the political shifts that have emerged as people of like-minded politics have moved into similar and dense areas. By itself, it's almost certainly not something that will cost Ernst this year's election. But over time, the tension between rural and urban areas in every state could become significant factors in politics, if this polarization continues.